From joshmccormack at travelersdiary.com Mon Apr 2 14:01:59 2007 From: joshmccormack at travelersdiary.com (Josh McCormack) Date: Mon, 2 Apr 2007 13:01:59 -0500 Subject: [nycbug-talk] quickbooks alternative Message-ID: Does anyone use an open source solution, ideally on a *BSD, as an alternative to QuickBooks? I'd like something that can handle my service based business, with various pricing, to track invoices, and perhaps do more, like accounting, even CRM & newsletters and such. Thanks, Josh McCormack -------------- next part -------------- An HTML attachment was scrubbed... URL: From george at ceetonetechnology.com Mon Apr 2 14:10:14 2007 From: george at ceetonetechnology.com (George Rosamond) Date: Mon, 02 Apr 2007 14:10:14 -0400 Subject: [nycbug-talk] quickbooks alternative In-Reply-To: References: Message-ID: <46114706.9010002@ceetonetechnology.com> Josh McCormack wrote: > Does anyone use an open source solution, ideally on a *BSD, as an > alternative to QuickBooks? > I'd like something that can handle my service based business, with > various pricing, to track invoices, and perhaps do more, like > accounting, even CRM & newsletters and such. I know some people are using gnucash. . . although a while back, at least, it wasn't primetime.. . George From mark at dueck.bz Mon Apr 2 14:11:22 2007 From: mark at dueck.bz (Mark Dueck) Date: Mon, 02 Apr 2007 12:11:22 -0600 Subject: [nycbug-talk] quickbooks alternative In-Reply-To: References: Message-ID: <6.2.5.6.0.20070402120734.03022878@dueck.bz> I use phpCoin to do a lot of my accounting. I'm a consultant for multiple businesses. I am currently becoming a reseller for Quasar, a very good quickbooks replacement. Check it out on www.linuxcanada.com. Runs on linux, can have Firebird or Postgres as the DB, and has tons of features. Regards, Mark At 12:01 PM 4/2/2007, Josh McCormack wrote: >Does anyone use an open source solution, ideally on a *BSD, as an >alternative to QuickBooks? >I'd like something that can handle my service based business, with >various pricing, to track invoices, and perhaps do more, like >accounting, even CRM & newsletters and such. > >Thanks, > >Josh McCormack >_______________________________________________ >% NYC*BUG talk mailing list >http://lists.nycbug.org/mailman/listinfo/talk >%Be sure to check out our Jobs and NYCBUG-announce lists >%We meet the first Wednesday of the month > >Internal Virus Database is out-of-date. >Checked by AVG Free Edition. >Version: 7.5.446 / Virus Database: 268.18.17/731 - Release Date: >3/23/2007 3:27 PM From jca at sdf.lonestar.org Mon Apr 2 14:20:18 2007 From: jca at sdf.lonestar.org (Jonathan C. Allen) Date: Mon, 2 Apr 2007 18:20:18 +0000 Subject: [nycbug-talk] quickbooks alternative In-Reply-To: References: Message-ID: <20070402182018.GB27408@SDF.LONESTAR.ORG> On Mon, Apr 02, 2007 at 01:01:59PM -0500, Josh McCormack wrote: > Does anyone use an open source solution, ideally on a *BSD, as an > alternative to QuickBooks? > I'd like something that can handle my service based business, with various > pricing, to track invoices, and perhaps do more, like accounting, even CRM & > newsletters and such. I've used GNUCash before, it is probably a good package for you to start with: GNUCash http://www.gnucash.org/ Nice CLI double-entry accounting ledger: Ledger http://newartisans.com/ledger.html jca From nycbug at cyth.net Mon Apr 2 20:44:35 2007 From: nycbug at cyth.net (Ray Lai) Date: Mon, 2 Apr 2007 20:44:35 -0400 Subject: [nycbug-talk] OpenCVS questions? Message-ID: <20070403004458.GC12761@cybertron.cyth.net> Hi Nycbuggers, In preparation for my upcoming OpenCVS talk, please send me or the list any questions you may want answered during the talk. Thanks! -Ray- From swygue at gmail.com Wed Apr 4 15:31:57 2007 From: swygue at gmail.com (Rodrique Heron) Date: Wed, 04 Apr 2007 15:31:57 -0400 Subject: [nycbug-talk] Apache rewrite rules Message-ID: <4613FD2D.3090100@gmail.com> I am responsible for a web server that several departments publish content to. I want to delegate the creation of rewrite rules to a non privileged user. Can I accomplish this safely without giving them shell access or using .htaccess ? Are there any tools available ? From pete at nomadlogic.org Wed Apr 4 15:55:57 2007 From: pete at nomadlogic.org (Peter Wright) Date: Wed, 4 Apr 2007 12:55:57 -0700 (PDT) Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <4613FD2D.3090100@gmail.com> References: <4613FD2D.3090100@gmail.com> Message-ID: <55843.160.33.20.11.1175716557.squirrel@webmail.nomadlogic.org> > I am responsible for a web server that several departments publish > content to. I want to delegate the creation of rewrite rules to a non > privileged user. Can I accomplish this safely without giving them shell > access or using .htaccess ? Are there any tools available ? > ahh it's been a while since i've had to deal with mod_rewrite rules - but i think you can break out the rules from the main httpd.conf file. at the end of your main config file you can put a directive like this in: Include conf/conf.d/*.conf or what ever directory structure you would like to use. in the conf.d directory you could then setup what ever access control mechanisms you want for your various users. we use svn and rcs along side triggers and Makefile's to publish changes to these directories. I think this should work for you - unless I misunderstood your task. -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From jlam at pkgsrc.org Wed Apr 4 16:12:35 2007 From: jlam at pkgsrc.org (Johnny C. Lam) Date: Wed, 04 Apr 2007 16:12:35 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <4613FD2D.3090100@gmail.com> References: <4613FD2D.3090100@gmail.com> Message-ID: <461406B3.3000709@pkgsrc.org> Rodrique Heron wrote: > I am responsible for a web server that several departments publish > content to. I want to delegate the creation of rewrite rules to a non > privileged user. Can I accomplish this safely without giving them shell > access or using .htaccess ? Are there any tools available ? Not that I endorse its use, but Webmin has an Apache module that probably does what you're asking. Cheers, -- Johnny Lam From swygue at gmail.com Wed Apr 4 16:15:35 2007 From: swygue at gmail.com (Rodrique Heron) Date: Wed, 04 Apr 2007 16:15:35 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <55843.160.33.20.11.1175716557.squirrel@webmail.nomadlogic.org> References: <4613FD2D.3090100@gmail.com> <55843.160.33.20.11.1175716557.squirrel@webmail.nomadlogic.org> Message-ID: <46140767.6000105@gmail.com> Peter Wright wrote: >> I am responsible for a web server that several departments publish >> content to. I want to delegate the creation of rewrite rules to a non >> privileged user. Can I accomplish this safely without giving them shell >> access or using .htaccess ? Are there any tools available ? >> >> > > ahh it's been a while since i've had to deal with mod_rewrite rules - but > i think you can break out the rules from the main httpd.conf file. at the > end of your main config file you can put a directive like this in: > > Include conf/conf.d/*.conf > > > or what ever directory structure you would like to use. in the conf.d > directory you could then setup what ever access control mechanisms you > want for your various users. we use svn and rcs along side triggers and > Makefile's to publish changes to these directories. > > I think this should work for you - unless I misunderstood your task. > > -p > > > Thanks Peter, That's exactly what I thought of, unfortunately I have no one whose competent enough to even ssh into the server and edit the file. What I am looking for is a web front-end, which I certainly lack the coding skills to do. For example, inside conf.d, a file called rewrite.conf and allow modifications to this file via the web front-end. Then have a script run from cron that checks if the file has been updated and restarts apache. From rodrique_heron at baruch.cuny.edu Wed Apr 4 16:19:51 2007 From: rodrique_heron at baruch.cuny.edu (Rodrique Heron) Date: Wed, 04 Apr 2007 16:19:51 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <461406B3.3000709@pkgsrc.org> References: <4613FD2D.3090100@gmail.com> <461406B3.3000709@pkgsrc.org> Message-ID: <46140867.60407@baruch.cuny.edu> Johnny C. Lam wrote: > Rodrique Heron wrote: > >> I am responsible for a web server that several departments publish >> content to. I want to delegate the creation of rewrite rules to a non >> privileged user. Can I accomplish this safely without giving them shell >> access or using .htaccess ? Are there any tools available ? >> > > Not that I endorse its use, but Webmin has an Apache module that > probably does what you're asking. > > Cheers, > > -- Johnny Lam > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > > Same here, I'm considering webmin but I am aware of it. -- Rodrique Heron System Administrator Baruch College New York, NY (646) 312-1055 From okan at demirmen.com Wed Apr 4 16:24:18 2007 From: okan at demirmen.com (Okan Demirmen) Date: Wed, 4 Apr 2007 16:24:18 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <46140767.6000105@gmail.com> References: <4613FD2D.3090100@gmail.com> <55843.160.33.20.11.1175716557.squirrel@webmail.nomadlogic.org> <46140767.6000105@gmail.com> Message-ID: <20070404202418.GD29570@clam.khaoz.org> On Wed 2007.04.04 at 16:15 -0400, Rodrique Heron wrote: > Peter Wright wrote: > >> I am responsible for a web server that several departments publish > >> content to. I want to delegate the creation of rewrite rules to a non > >> privileged user. Can I accomplish this safely without giving them shell > >> access or using .htaccess ? Are there any tools available ? > >> > >> > > > > ahh it's been a while since i've had to deal with mod_rewrite rules - but > > i think you can break out the rules from the main httpd.conf file. at the > > end of your main config file you can put a directive like this in: > > > > Include conf/conf.d/*.conf > > > > > > or what ever directory structure you would like to use. in the conf.d > > directory you could then setup what ever access control mechanisms you > > want for your various users. we use svn and rcs along side triggers and > > Makefile's to publish changes to these directories. > > > > I think this should work for you - unless I misunderstood your task. > > > > -p > > > > > > > Thanks Peter, > > That's exactly what I thought of, unfortunately I have no one whose > competent enough to even ssh into the server and edit the file. What I > am looking for is a web front-end, which I certainly lack the coding > skills to do. For example, inside conf.d, a file called rewrite.conf and > allow modifications to this file via the web front-end. Then have a > script run from cron that checks if the file has been updated and > restarts apache. asking for trouble, imho... if "they" can't even ssh in and edit one file, do you really trust them to write rewrite rules? you use "safely" in your requirements...may have to re-think your process ;) From nycbug-list at 2xlp.com Wed Apr 4 16:28:49 2007 From: nycbug-list at 2xlp.com (Jonathan Vanasco) Date: Wed, 4 Apr 2007 16:28:49 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <4613FD2D.3090100@gmail.com> References: <4613FD2D.3090100@gmail.com> Message-ID: <0AD63071-E683-4167-965A-D0ABCBAF6D21@2xlp.com> On Apr 4, 2007, at 3:31 PM, Rodrique Heron wrote: > I am responsible for a web server that several departments publish > content to. I want to delegate the creation of rewrite rules to a non > privileged user. Can I accomplish this safely without giving them > shell > access or using .htaccess ? Are there any tools available ? if you're not using a .htaccess file, then you'll need to restart apache -- which means you do need a priv'd user, and you'll have to let them restart the app suggestions: 1) use mod_perl / mod_python / etc to handle the rewrite rules off a static file , have a web page read/validate/write to that file . configure it to be read periodically or on update. pro- keep apache con- modperl/modpython bloat apache ; security implications unless you take time to isolate rules/ranges for each section. 2) run nginx on port80, apache on 8000. proxy each dept. to its own ip port range like so: all 80 - nginx dept1 8010 - nginx dept2 8011 - nginx have a webscript manage the 8010/8011 conf files. proxy all valid requests to apache on port 8000. they're both halfassed ideas, but you're going to run into issues making sure that one department can't screw up another. From jlam at pkgsrc.org Wed Apr 4 16:41:12 2007 From: jlam at pkgsrc.org (Johnny C. Lam) Date: Wed, 04 Apr 2007 16:41:12 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <46140867.60407@baruch.cuny.edu> References: <4613FD2D.3090100@gmail.com> <461406B3.3000709@pkgsrc.org> <46140867.60407@baruch.cuny.edu> Message-ID: <46140D68.40108@pkgsrc.org> Rodrique Heron wrote: > Johnny C. Lam wrote: >> Rodrique Heron wrote: >> >>> I am responsible for a web server that several departments publish >>> content to. I want to delegate the creation of rewrite rules to a non >>> privileged user. Can I accomplish this safely without giving them shell >>> access or using .htaccess ? Are there any tools available ? >>> >> Not that I endorse its use, but Webmin has an Apache module that >> probably does what you're asking. > > Same here, I'm considering webmin but I am aware of it. Just out of curiosity, why would some department even need special URL rewrite rules? Does the webapp demand their use? Cheers, -- Johnny Lam From thareja.lists at gmail.com Wed Apr 4 16:58:25 2007 From: thareja.lists at gmail.com (Sachin Thareja) Date: Wed, 4 Apr 2007 16:58:25 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <46140D68.40108@pkgsrc.org> References: <4613FD2D.3090100@gmail.com> <461406B3.3000709@pkgsrc.org> <46140867.60407@baruch.cuny.edu> <46140D68.40108@pkgsrc.org> Message-ID: <9bdd47ef0704041358i4be4346dwecee5e880b264514@mail.gmail.com> > Just out of curiosity, why would some department even need special URL > rewrite rules? Does the webapp demand their use? If they want to do this: http://company.com/aweseome-product -> Actually rewrites to: http://company.com/some/deep/link/possibly/even/to/not/such/an/awesome/product.htm and the department creates/does away with their products at a rate of >once a week, enough for the local sysadmin to look for alternatives to this weekly cycle of pestering... But that's just my point of view... ;-) -Sachin From nycbug-list at 2xlp.com Wed Apr 4 17:18:59 2007 From: nycbug-list at 2xlp.com (Jonathan Vanasco) Date: Wed, 4 Apr 2007 17:18:59 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <46140767.6000105@gmail.com> References: <4613FD2D.3090100@gmail.com> <55843.160.33.20.11.1175716557.squirrel@webmail.nomadlogic.org> <46140767.6000105@gmail.com> Message-ID: <54357A0C-8044-4459-8886-B997C40959D5@2xlp.com> > That's exactly what I thought of, unfortunately I have no one whose > competent enough to even ssh into the server and edit the file. What I > am looking for is a web front-end, which I certainly lack the coding > skills to do. For example, inside conf.d, a file called > rewrite.conf and > allow modifications to this file via the web front-end. Then have a > script run from cron that checks if the file has been updated and > restarts apache. the problem with that is that a single bad line will kill all of apache if you're going to do that, you need to segment out the risk to per- department. ie: they can only break their own sections. you also need to add in some sort of testing system for the config // Jonathan Vanasco | - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - | SyndiClick.com | - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - | FindMeOn.com - The cure for Multiple Web Personality Disorder | Web Identity Management and 3D Social Networking | - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - | RoadSound.com - Tools For Bands, Stuff For Fans | Collaborative Online Management And Syndication Tools | - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - From lists at genoverly.net Wed Apr 4 17:35:14 2007 From: lists at genoverly.net (michael) Date: Wed, 4 Apr 2007 17:35:14 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <9bdd47ef0704041358i4be4346dwecee5e880b264514@mail.gmail.com> References: <4613FD2D.3090100@gmail.com> <461406B3.3000709@pkgsrc.org> <46140867.60407@baruch.cuny.edu> <46140D68.40108@pkgsrc.org> <9bdd47ef0704041358i4be4346dwecee5e880b264514@mail.gmail.com> Message-ID: <20070404173514.0d10821f@dt.genoverly.com> On Wed, 4 Apr 2007 16:58:25 -0400 "Sachin Thareja" wrote: > > Just out of curiosity, why would some department even need special > > URL rewrite rules? Does the webapp demand their use? > > If they want to do this: > > http://company.com/aweseome-product > > -> Actually rewrites to: > > http://company.com/some/deep/link/possibly/even/to/not/such/an/awesome/product.htm > > and the department creates/does away with their products at a rate of > >once a week, enough for the local sysadmin to look for alternatives > to this weekly cycle of pestering... But that's just my point of > view... ;-) > > -Sachin Wow, it seems like continuously editting the web server config (and restarting/reloading the config) each time somebody builds a new web page.. and wants a purdy URL; is a senseless waste of time for the sysadmin. On top of others saying that a poorly written (or poorly tested) config can reak havoc; it is probably a maintenance nightmare. The web application should handle that. I'll echo a prior post.. "you may have to re-think your process". -- michael (this address does not accept public email) From swygue at gmail.com Wed Apr 4 20:42:01 2007 From: swygue at gmail.com (Rodrique Heron) Date: Wed, 04 Apr 2007 20:42:01 -0400 Subject: [nycbug-talk] Apache rewrite rules In-Reply-To: <20070404173514.0d10821f@dt.genoverly.com> References: <4613FD2D.3090100@gmail.com> <461406B3.3000709@pkgsrc.org> <46140867.60407@baruch.cuny.edu> <46140D68.40108@pkgsrc.org> <9bdd47ef0704041358i4be4346dwecee5e880b264514@mail.gmail.com> <20070404173514.0d10821f@dt.genoverly.com> Message-ID: <461445D9.9080706@gmail.com> michael wrote: > On Wed, 4 Apr 2007 16:58:25 -0400 > "Sachin Thareja" wrote: > > >>> Just out of curiosity, why would some department even need special >>> URL rewrite rules? Does the webapp demand their use? >>> >> If they want to do this: >> >> http://company.com/aweseome-product >> >> -> Actually rewrites to: >> >> http://company.com/some/deep/link/possibly/even/to/not/such/an/awesome/product.htm >> >> and the department creates/does away with their products at a rate of >> >>> once a week, enough for the local sysadmin to look for alternatives >>> >> to this weekly cycle of pestering... But that's just my point of >> view... ;-) >> >> -Sachin >> > > Wow, it seems like continuously editting the web server config (and > restarting/reloading the config) each time somebody builds a new web > page.. and wants a purdy URL; is a senseless waste of time for the > sysadmin. On top of others saying that a poorly written (or > poorly tested) config can reak havoc; it is probably a maintenance > nightmare. The web application should handle that. > > I'll echo a prior post.. "you may have to re-think your process". > > I am changing the way this is done. I inherited this web server and I'm in the process of rebuilding the entire web infrastructure. So I welcome your suggestions. re: Johnny C. Lam Why would some department even need special URL rewrite rules? There's a requirement to have all URL conform to http://www.company.com, so http://department.company.com/events/welcome.htm must be accessible by http://www.company.com/events and as Michael pointed out, to make really long URL shorter. Does the webapp demand their use ? No. This is all static contents, regular user renames a folder but still wants it accessible from the old url. This happens often enough to byte into my time, so I want to delegate it to the department whose responsible for web content. And since I'm in the process of a overhaul I would appreciate some best practice feedback. The redirects are the most pestering of the problems, another is, all users publish directly to the main web server. Most of them are using Adobe Contribute and the rest access it via ftp and ssh. I've gotten the support to change this, and put up a staging server and I'm killing ftp. I would like to know how nycbug members layout their web infrastructure, where do you place your proxy servers and what do you proxy ? From KReiter at insidefsi.net Thu Apr 5 09:36:51 2007 From: KReiter at insidefsi.net (Kevin Reiter) Date: Thu, 5 Apr 2007 09:36:51 -0400 Subject: [nycbug-talk] FreeBSD syslog.conf quickie In-Reply-To: <54357A0C-8044-4459-8886-B997C40959D5@2xlp.com> Message-ID: <184B0715C3D74243B86F872B55C340E703A9126E@fsi32.fsidp.insidefsi.com> All, A quick question on the syntax for defining a remote system logging to a remote syslog server. I have a number of routers I want to send their syslogs to a FreeBSD system. I've checked all the firewall rules and know that not to be a problem. My question is, in the syslog.conf file itself, how to I define a host to log to a specific logfile? I've RTFM, checked Google, and checked everything else I thought might have been mucking up the waters, but can't seem to get this straight. I'm looking to get all syslogs for hostname "ssr" to go to /var/log/router.log In my /etc/syslog.conf, I have the following entry: +ssr.* /var/log/router.log >From man 5 syslog.conf: A hostname specification of the form `#+hostname' or `+hostname' means the following blocks will be applied to messages received from the speci- fied hostname. Alternatively, the hostname specification `#-hostname' or `-hostname' causes the following blocks to be applied to messages from any host but the one specified. If the hostname is given as `@', the local hostname will be used. As for program specifications, multiple comma-separated values may be specified for hostname specifications. I've tried using +ssr, +ssr.*, and a few other variations with no luck. Am I just reading the examples wrong? (In my defense, I've had a headache for the past 3 days, which hasn't helped..) Thanks, Kev This message may contain confidential or proprietary information and is intended solely for the individual(s) to whom it is addressed. If you are not a named addressee you should not disseminate, distribute or copy this e-mail or act upon the information contained herein. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. From swygue at gmail.com Thu Apr 5 10:13:44 2007 From: swygue at gmail.com (Rodrique Heron) Date: Thu, 05 Apr 2007 10:13:44 -0400 Subject: [nycbug-talk] FreeBSD syslog.conf quickie In-Reply-To: <184B0715C3D74243B86F872B55C340E703A9126E@fsi32.fsidp.insidefsi.com> References: <184B0715C3D74243B86F872B55C340E703A9126E@fsi32.fsidp.insidefsi.com> Message-ID: <46150418.3060506@gmail.com> Kevin Reiter wrote: > All, > > A quick question on the syntax for defining a remote system logging to a remote syslog server. > > I have a number of routers I want to send their syslogs to a FreeBSD system. I've checked all the firewall rules and know that not to be a problem. My question is, in the syslog.conf file itself, how to I define a host to log to a specific logfile? I've RTFM, checked Google, and checked everything else I thought might have been mucking up the waters, but can't seem to get this straight. > > I'm looking to get all syslogs for hostname "ssr" to go to /var/log/router.log > > In my /etc/syslog.conf, I have the following entry: > > +ssr.* /var/log/router.log > > >From man 5 syslog.conf: > > A hostname specification of the form `#+hostname' or `+hostname' means > the following blocks will be applied to messages received from the speci- > fied hostname. Alternatively, the hostname specification `#-hostname' or > `-hostname' causes the following blocks to be applied to messages from > any host but the one specified. If the hostname is given as `@', the > local hostname will be used. As for program specifications, multiple > comma-separated values may be specified for hostname specifications. > > I've tried using +ssr, +ssr.*, and a few other variations with no luck. > > Am I just reading the examples wrong? (In my defense, I've had a headache for the past 3 days, which hasn't helped..) > > Thanks, > Kev > > This message may contain confidential or proprietary information and is intended solely for the individual(s) to whom it is addressed. If you are not a named addressee you should not disseminate, distribute or copy this e-mail or act upon the information contained herein. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. > > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > > Use minirsyslogd instead, I am don't know why BSD syslog won't work. I created a /va/log/cisco/{device1,device2}, replace device with the address with ip address of your switch or router. in == /etc/rc.conf == minirsyslog_enable="YES" minirsyslog_flags="--rootdir /var/log/cisco --umask 022" syslogd_enable="YES" syslogd_flags="-ss -b 192.168.2.1" That works for me, minirsyslog only writes when the buffer is full, so you may not see events right away. From KReiter at insidefsi.net Thu Apr 5 10:51:44 2007 From: KReiter at insidefsi.net (Kevin Reiter) Date: Thu, 5 Apr 2007 10:51:44 -0400 Subject: [nycbug-talk] FreeBSD syslog.conf quickie In-Reply-To: <46150418.3060506@gmail.com> Message-ID: <184B0715C3D74243B86F872B55C340E703A91273@fsi32.fsidp.insidefsi.com> Rodrique Heron wrote: : Kevin Reiter wrote: :: All, :: :: A quick question on the syntax for defining a remote system logging :: to a remote syslog server. :: :: I have a number of routers I want to send their syslogs to a FreeBSD :: system. I've checked all the firewall rules and know that not to be :: a problem. My question is, in the syslog.conf file itself, how to I :: define a host to log to a specific logfile? I've RTFM, checked :: Google, and checked everything else I thought might have been :: mucking up the waters, but can't seem to get this straight. :: :: I'm looking to get all syslogs for hostname "ssr" to go to :: /var/log/router.log :: :: In my /etc/syslog.conf, I have the following entry: :: :: +ssr.* /var/log/router.log :: ::: From man 5 syslog.conf: :: :: A hostname specification of the form `#+hostname' or `+hostname' :: means :: the following blocks will be applied to messages received from the :: speci- :: fied hostname. Alternatively, the hostname specification :: `#-hostname' or `-hostname' causes the following blocks to be :: applied to messages from :: any host but the one specified. If the hostname is given as `@', the :: local hostname will be used. As for program specifications, multiple :: comma-separated values may be specified for hostname specifications. :: :: I've tried using +ssr, +ssr.*, and a few other variations with no :: luck. :: :: Am I just reading the examples wrong? (In my defense, I've had a :: headache for the past 3 days, which hasn't helped..) :: :: Thanks, :: Kev :: :: This message may contain confidential or proprietary information and :: is intended solely for the individual(s) to whom it is addressed. :: If you are not a named addressee you should not disseminate, :: distribute or copy this e-mail or act upon the information contained :: herein. Please notify the sender immediately by e-mail if you have :: received this e-mail by mistake and delete this e-mail from your :: system. :: :: _______________________________________________ :: % NYC*BUG talk mailing list :: http://lists.nycbug.org/mailman/listinfo/talk :: %Be sure to check out our Jobs and NYCBUG-announce lists :: %We meet the first Wednesday of the month :: :: : Use minirsyslogd instead, I am don't know why BSD syslog won't work. I : created a /va/log/cisco/{device1,device2}, replace device with the : address with ip address of your switch or router. : : in == /etc/rc.conf == : minirsyslog_enable="YES" : minirsyslog_flags="--rootdir /var/log/cisco --umask 022" : : syslogd_enable="YES" : syslogd_flags="-ss -b 192.168.2.1" : : That works for me, minirsyslog only writes when the buffer is full, so : you may not see events right away. Thanks, but I'd rather use the built-in syslogd for this, without having to install anything else. I know it *should* work, once I get the right syntax for the configs figured out. I've changed rc.conf to include: syslogd_flags="-b loghost.intranet.com -a ip_of_router:* -a ssr:* -a ssr" (ssr being the hostname of the router that resolves via DNS) Restarted syslogd, but still nothing. root at darkhorse [~]# ls -l /var/log/router.log -rw-r--r-- 1 root wheel 64 Apr 5 00:00 /var/log/router.log root at darkhorse [~]# sockstat | grep 514 root syslogd 21191 6 udp4 my_ip:514 *:* ## Router logging starts +ssr *.* /var/log/router.log ## Router logging ends Am I missing something? From nikolai at fetissov.org Thu Apr 5 11:15:51 2007 From: nikolai at fetissov.org (nikolai) Date: Thu, 5 Apr 2007 11:15:51 -0400 (EDT) Subject: [nycbug-talk] April 2007 meeting audio Message-ID: <37596.63.66.6.15.1175786151.squirrel@www.geekisp.com> Folks, Audio of Ray's talk is up at http://www.fetissov.org/public/nycbug/ Cheers. -- Nikolai From KReiter at insidefsi.net Thu Apr 5 13:10:00 2007 From: KReiter at insidefsi.net (Kevin Reiter) Date: Thu, 5 Apr 2007 13:10:00 -0400 Subject: [nycbug-talk] FreeBSD syslog.conf quickie In-Reply-To: <184B0715C3D74243B86F872B55C340E703A91273@fsi32.fsidp.insidefsi.com> Message-ID: <184B0715C3D74243B86F872B55C340E703A91284@fsi32.fsidp.insidefsi.com> talk-bounces at lists.nycbug.org wrote: : Rodrique Heron wrote: :: Kevin Reiter wrote: ::: All, ::: ::: A quick question on the syntax for defining a remote system logging ::: to a remote syslog server. ::: ::: I have a number of routers I want to send their syslogs to a FreeBSD ::: system. I've checked all the firewall rules and know that not to be ::: a problem. My question is, in the syslog.conf file itself, how to I ::: define a host to log to a specific logfile? I've RTFM, checked ::: Google, and checked everything else I thought might have been ::: mucking up the waters, but can't seem to get this straight. ::: ::: I'm looking to get all syslogs for hostname "ssr" to go to ::: /var/log/router.log ::: ::: In my /etc/syslog.conf, I have the following entry: ::: ::: +ssr.* /var/log/router.log ::: :::: From man 5 syslog.conf: ::: ::: A hostname specification of the form `#+hostname' or `+hostname' ::: means the following blocks will be applied to messages received ::: from the speci- fied hostname. Alternatively, the hostname ::: specification `#-hostname' or `-hostname' causes the following ::: blocks to be applied to messages from any host but the one ::: specified. If the hostname is given as `@', the local hostname ::: will be used. As for program specifications, multiple ::: comma-separated values may be specified for hostname ::: specifications. ::: ::: I've tried using +ssr, +ssr.*, and a few other variations with no ::: luck. ::: ::: Am I just reading the examples wrong? (In my defense, I've had a ::: headache for the past 3 days, which hasn't helped..) ::: ::: Thanks, ::: Kev ::: ::: This message may contain confidential or proprietary information and ::: is intended solely for the individual(s) to whom it is addressed. ::: If you are not a named addressee you should not disseminate, ::: distribute or copy this e-mail or act upon the information contained ::: herein. Please notify the sender immediately by e-mail if you have ::: received this e-mail by mistake and delete this e-mail from your ::: system. ::: ::: _______________________________________________ ::: % NYC*BUG talk mailing list ::: http://lists.nycbug.org/mailman/listinfo/talk ::: %Be sure to check out our Jobs and NYCBUG-announce lists ::: %We meet the first Wednesday of the month ::: ::: :: Use minirsyslogd instead, I am don't know why BSD syslog won't work. :: I created a /va/log/cisco/{device1,device2}, replace device with the :: address with ip address of your switch or router. :: :: in == /etc/rc.conf == :: minirsyslog_enable="YES" :: minirsyslog_flags="--rootdir /var/log/cisco --umask 022" :: :: syslogd_enable="YES" :: syslogd_flags="-ss -b 192.168.2.1" :: :: That works for me, minirsyslog only writes when the buffer is full, :: so you may not see events right away. : : Thanks, but I'd rather use the built-in syslogd for this, without : having to install anything else. I know it *should* work, once I get : the right syntax for the configs figured out. : : I've changed rc.conf to include: : syslogd_flags="-b loghost.intranet.com -a ip_of_router:* -a ssr:* -a : ssr" : : (ssr being the hostname of the router that resolves via DNS) : : Restarted syslogd, but still nothing. : : root at darkhorse [~]# ls -l /var/log/router.log : -rw-r--r-- 1 root wheel 64 Apr 5 00:00 /var/log/router.log : : root at darkhorse [~]# sockstat | grep 514 : root syslogd 21191 6 udp4 my_ip:514 *:* : : ## Router logging starts : +ssr : *.* /var/log/router.log : ## Router logging ends : : : Am I missing something? OK, I now see messages from my test system, but they're showing up in /var/log/messages as opposed to the logfile I specified in /etc/syslog.conf: root at tracker [~]# cat /var/log/messages | grep darkhorse Apr 5 12:01:09 darkhorse kevin: testing123 Apr 5 12:01:44 darkhorse kevin: testing1234 Apr 5 12:04:25 darkhorse kevin: test 1234 Apr 5 12:06:05 darkhorse kevin: test 12 Apr 5 12:07:37 darkhorse kevin: why are they showing up in messages? Apr 5 12:29:07 darkhorse kevin: why are they showing up in messages instead of darkhorse? Apr 5 12:34:08 darkhorse kevin: hmmm... Apr 5 12:58:46 darkhorse kevin: another test Here is what I have configured on my test syslog server: Darkhorse = host sending syslog entries to syslog server Tracker = test syslog server root at tracker [~]# ls -l /var/log/darkhorse.log -rw-r--r-- 1 root wheel 5 Apr 5 11:37 /var/log/darkhorse.log (It's readable by others so a Perl script I have parsing the file and displaying it in a browser can actually read it.) In /etc/syslog.conf on tracker: ################################### # Remote syslog testing starts here ################################### `+darkhorse' *.* /var/log/darkhorse.log In /etc/rc.conf on tracker: # Syslog testing syslogd_flags="" (when I use: syslogd_flags="-a darkhorse" I get nothing in any log from that host) The only thing currently not working as expected is the logs not going into the specified logfile. Any ideas? Thanks, Kev From skreuzer at f2o.org Fri Apr 6 12:09:06 2007 From: skreuzer at f2o.org (Steven Kreuzer) Date: Fri, 6 Apr 2007 12:09:06 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. Message-ID: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> Thought I would post this to the list since there has been alot of talk about ZFS. ZFS support was just commited to HEAD as an experimental feature and will be available in 7.0-RELEASE From the commit log: Please welcome ZFS - The last word in file systems. ZFS file system was ported from OpenSolaris operating system. The code in under CDDL license. I'd like to thank all SUN developers that created this great piece of software. Supported by: Wheel LTD (http://www.wheel.pl/) Supported by: The FreeBSD Foundation (http:// www.freebsdfoundation.org/) Supported by: Sentex (http://www.sentex.net/) Limitations. Currently ZFS is only compiled as kernel module and is only available for i386 architecture. Amd64 should be available very soon, the other archs will come later, as we implement needed atomic operations. Missing functionality. - We don't have iSCSI target daemon in the tree, so sharing ZVOLs via iSCSI is also not supported at this point. This should be fixed in the future, we may also add support for sharing ZVOLs over ggate. - There is no support for ACLs and extended attributes. - There is no support for booting off of ZFS file system. Other than that, ZFS should be fully-functional. Lets give a very big round of applause for Pawel Jakub Dawidek. SK From spork at bway.net Fri Apr 6 15:02:57 2007 From: spork at bway.net (Charles Sprickman) Date: Fri, 6 Apr 2007 15:02:57 -0400 (EDT) Subject: [nycbug-talk] ZFS committed to FreeBSD base. In-Reply-To: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> References: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> Message-ID: On Fri, 6 Apr 2007, Steven Kreuzer wrote: > Thought I would post this to the list since there has been alot of > talk about ZFS. > > ZFS support was just commited to HEAD as an experimental feature and > will be available in 7.0-RELEASE [...] > - There is no support for booting off of ZFS file system. For anyone who's following this closely, is bootable ZFS supposed to eventually make it's way into 7.x, or will that be for 8.x? Thanks, Charles > Lets give a very big round of applause for Pawel Jakub Dawidek. Hell yeah! > SK > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > From carton at Ivy.NET Fri Apr 6 15:14:20 2007 From: carton at Ivy.NET (Miles Nordin) Date: Fri, 06 Apr 2007 15:14:20 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. In-Reply-To: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> (Steven Kreuzer's message of "Fri, 6 Apr 2007 12:09:06 -0400") References: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> Message-ID: >>>>> "sk" == Steven Kreuzer writes: sk> - We don't have iSCSI target daemon in the tree, so sharing sk> ZVOLs via iSCSI is also not supported at this point. Is there a working iSCSI initiator for FreeBSD? sk> There is no support for booting off of ZFS file system. nor in Solaris. There is some clumsy mod to have a ZFS root on Solaris/x86 only, but it's based on something like a Linux initrd (ugly), and also not working on SPARC. This is great, though! I've been using ZFS for about a year, and it does solve real problems. The AFS-ish separation between pools and filesystems is important for me, as are reliable, persistent, performant shapshots and clones. And of course being able to make and to fsck a multi-TB filesystem in a BSD plugs a big gap and one of the things for which I had to use Linux. Also maybe it will light a fire under Sun to prove that using their FMA daemon to manage faults is really a benefit, not a liability, a spaghetti disaster, and a bug mill compared to the old Unix way of handling hardware failures and other exceptions in-context like FreeBSD. Maybe Sun will clean up their disklabel mess, too (that's another thing which is much less irritating in FreeBSD). -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: From jeroen at unfix.org Fri Apr 6 18:34:29 2007 From: jeroen at unfix.org (Jeroen Massar) Date: Fri, 06 Apr 2007 23:34:29 +0100 Subject: [nycbug-talk] IPv6 NY-US Roll Call In-Reply-To: References: <5A00D735-3806-4FD9-B5BC-19AFC46D8C15@lesmuug.org> Message-ID: <4616CAF5.1070405@spaghetti.zurich.ibm.com> Hi, First of Isaac, nice writeup, I also tend to hear that in Japan they have quite some advanced usages already of the Internet in general. We are just getting left out in the cold in the rest of the world with it because in the western countries the companies want to earn money from you and as long as they can sell the old crap, why would they try and sell the new crap? They can in a couple of years make loads of money off all the problems that will occur when IPv4 does become harder to get and when tools suddenly have to talk IPv6 because people are having to use it they can charge very well for it, as it also has to happen in a hurry. In other words folks: brush up on your IPv6 skills and in a couple of years you will be consulting the big companies who are giving you money to help them out of their misery. Anyway, somebody pointed me to this thread as the person noted that there where a couple of FUD's trying to be spread. And as this is publically archived, I would like to avoid any confusion and clarify what is being said wrongly. Of course, like always, if you have any questions never ever hesitate to contact SixXS using: info at sixxs.net As clearly detailed on http://www.sixxs.net/contact/ I tend to type long elaborate messages and yes they are all PGP signed so that context can't be removed or words twisted. As it is a long reply, here is a (long ;) summary: * SixXS != OCCAID * SixXS does allow IRC, but people who just want IPv6 to hide on the internet are out of luck: whois.sixxs.net shows your IPv4. * There is no IRC channel for SixXS (except noc.sixxs.net which is for monitoring purposes and is +m with only a bot saying stuff). * IRC servers and shell servers are indeed not allowed. Simple rule. * Abusing the service isn't either, just like any other ISP asks you to be a good netizen and to behave. * It works and is stable. * If you want to define your own rules/requirements: pay for it. * Read the FAQ. * Of course info at sixxs.net for questions/comments/flames. Here goes, oh for replyers, use [..] cut's there is a lot of useless text in there. Replying to the summary is also an option of course, but do mind it doesn't capture the whole wording. Miles Nordin wrote: [..] > The alternative now is SixXS, which gets transit from OCCAID but has > lots of automation for serving Interweb users. Unfortunately they > also have fewer pops. SixXS doesn't only get transit, OCCAID, or actually a couple of ISP's that do that for OCCAID, host the PoPs. They take care of the machine, the power, upgrading it, connectivity, everything. They *pay* for it. SixXS merely provisions (read: configure) the tunnels on those machines. How the PoP routes the traffic, what policies apply, how it is used etc all that is based on how the owner of the PoP, in this case OCCAID, defines it. If they say "let everybody DDoS the world", then the SixXS tools simply ignore that. We most would not accept such a PoP application though as we have an interest in the better good of the Internet and to provide a solid base for people to actually use IPv6. [..] > OCCAID, and their new partners SixXS, seem to be generally technically > competent, but they have crippling layer 10 issues. It is stuffed > full of those people who claim they want to be ``apolitical'' and then > run their organization like some kind of pathetic nerd mafia. > problems like: > > o SixXS problems > > o AUP forbids irc clients Not true and I really wonder why you think that. First of all there is no "SixXS AUP" (only the Belnet PoP has one). We do expect people to be good Internet citizens, but one is expected to do that for all Internet services that I know of. See also: http://www.sixxs.net/faq/sixxs/?faq=aup Also a quick google(sixxs irc) will reveal to you: http://www.sixxs.net/faq/misc/?faq=irc Which clearly shows that we even have special agreements in place with the two largest (EFnet & IRCnet) and arranged I-lines with them specifically so that people can use IPv6 IRC. None of the servers are hosted in address space provided by SixXS though. The operators of those services have special permission to host them and generally have a 24/7 team standby to resolve DDoS issues and other strange attacks. BUT: We don't see a major reason for people to use it especially for that. You clearly have IPv4, otherwise you could not have an IPv6-in-IPv4 tunnel, so we do hint that people should be using that instead. > o AUP forbids irc servers > > o AUP forbids shell servers even when not used for irc This really is solely because of the simple nature that those services exist solely for one reason: Hiding on IRC and thus causing trouble. Which is something we don't like, as when people are able to hide, they will DDoS our services, and thus hurt a lot of people, instead of the annoying person hiding behind it. Do note that we do publish, in whois.sixxs.net, the IPv4 endpoint of every user and with even a slight breeze of an abuse ticket we will terminate the account, of course after verifying the facts naturally. If you want to have an IRC server, go to one of the many "IRC Shell Hosting" companies who have "bulletproof" services etc blabla. They tend to have you pay for them btw. If you really have a 'shell service' which is not meant for IRC, can you explain what that service entails? I checked, but you clearly never signed up to SixXS and never send an email either. As such, are you sure that what you think you read is really what you think you read? from http://www.sixxs.net/contact/ in a big green box: "In case you are having issues with our services, don't hesitate to contact us." Just so that you know.... > o a series of kiddie-heuristic harassment tests when you sign up > (``the ten easy steps to ipv6!'') Those "10 easy steps" (http://www.sixxs.net/faq/account/?faq=10steps) are very easy and seem to be followed by a lot of people. Nothing in there is actually against the 'kiddies'. The portion that is against the kiddies though is that they have to provide their real information. Just like when you sign up to an ADSL service or a colo etc. You are also held fully accountable for any traffic that flows from the delegations that get assigned to you, just like with your normal ISP. > o horror stories from many people who try to use them. If they cut > you off, they always cut you off first and ``inform'' you later > or not at all. One guy said they kept deleting his account > because they didn't believe the name he gave sounded real enough > (it was his real name). Which part of "One signup per person" is unclear? Can you also show which person didn't provide his real name and got rejected? "Kept deleting" sounds weird too, you really don't get a second account and deletion only works once. Also, if we have clearly indicated to somebody to *NOT* use our service for IRC servers and then they signup as somebody else and start using it anyway for IRC servers, then what should we do? Ask *again* to not do it? Indeed, they then nicely get terminated. Services provided are on a best basis, nobody pays for it (and there is no option to pay for it btw), if you are a good net citizen all will work fine, just as a couple of thousand people are already experiencing for the last 5 years nearly, and that is only the portion that it was called SixXS. IPng.nl was already in the air 2 years before that and we merged it into SixXS to provide a better and larger service for the users who actually want to use, not abuse, IPv6. > o OCCAID/SixXS problems I've had in the past Only covering SixXS, I don't know how OCCAID does/runs this thus can't say anything about that. I suggest that you ask them about it. > o putting mailing lists under ``emergency moderation'' when they > feel embarassed by the discussion There is no SixXS mailinglist, unless somebody has something up that I don't know about, but then what would it have to do with SixXS? There is info at sixxs.net and there is a Forum which is free for everybody to post on when you have an account. And of course, not to forget, the ticket system. > o offering better support to ``insiders'' on an unpublished irc > channel and worse support to people on the mailing list, then > banning and klining people from that channel if they're > insufficiently sycophantic There is no IRC channel for SixXS and there is no 'secret' channel either. Unless you mean a specific channel which covers all kinds of operational networking concepts which is about routing in the internet (IPv4 and IPv6) and resolving it, but that has totally no relation whatsoever to SixXS. There is #noc.sixxs.net on IRCNet of course, but that is a moderated channel where only a bot speaks to say how many requests are in the queue and which pops are broken, in the cases that they are broken. > o AUP forbids so-called ``DNS spam'' which is any DNS reverse > lookup that spells an English phrase or sounds excessively cute. > I guess this is another anti-irc thing, but I'm not happy about > indignities like this and don't think it's in the spirit of the > Internet, and I really bristle at the idea of passing on a > restriction like that to _my_ users. The FAQ mentions it, there is no AUP. If you don't like that rule, then don't use the service. Over the years having this little rule in place has saved us from a lot of problems. Ever since we introduced this rule somewhere in 2000 at IPng.nl and at first started warning people, later disabling them when they did it anyway, DDoSses started to stop and now they nearly don't occur anymore. We could have allowed all of this and then went the www.ipng.org.uk way: http://216.239.59.104/search?q=cache:1lPla2CNyLYJ:ipng.org.uk/+%22IPng.org.uk%22&hl=en&ct=clnk&cd=1 aka "DDoSSed to death". This regulation avoided this and that way we can still provide a service to people. Note that the IPng.org.uk folks gave up only after a couple of months of 'operation', I only found out that they even existed, with more or less a similar name to our IPng.nl when there suddenly where messages of help and cries flying around that they where turning it off unless they found something to do about it. Clearly you know what DDoS does to your network, and you have had to, according to wikipedia, change ISP's because you where a DDoS target. So why should we, as a free (gratuit, free as in beer) provider allow you to attracked it and then let *OUR* systems get hurt by it? > so, if you can live with that, SixXS is the way to go. I can't stand > them You clearly seem to have a more personal issue then than anything else. If you want to talk about it, try info at sixxs.net. No pun intended. What is soooo difficult of coming up with a decent hostname that you actually use!? I really never have somebody do a ssh to a host like: my.mother.did.nasty.things.with.your.father.and.that.is.why.you.exist.org Unless you alias that locally of course. It is useless and was never the intention of the DNS in the first place. Cute URL's to read up on: http://www.dnsspam.nl http://www.sixxs.net/faq/sixxs/?faq=dnsspam There used to be http://spamcalc.net with even more information, also specifically on the subject how IRC Networks take care of them. Personally I enjoy it much more in coming up with good punny short names, see http://unfix.org/projects/network/ For people in theology, or who simply read the bible once; purgatory is my 'gateway', 'heaven' is a Linux box, while hell is the Windows box. Heaven/paradise/eden are used for storage as they have lots of beautiful content (cam pix :). Limbo is a laptop, which is in a complete state of flux. Magrathea (per HHGTTG) is a BSD box which does "make world" a lot (to put in on topic for this list a bit ;) I know people using demonolatry to name their hosts, using the name of the demon who watches the gates to hell as the name of their gateway etc. Those things are way more creative then a stupid sentence, which doesn't reflect what the hostname is nor it's function, but is only used to show penis length on the Internet. IMHO the higher the spamcalc the more you have to compensate for something. > but I will probably sign up soon to get back some kind of > censored politicized v6 (albeit without BGP now). Why do you have a need for BGP? Do you have your own ASN? Do you have multiple circuits to the Internet? Do you have a prefix to announce? If not, then why BGP, as it for sure isn't helping connectivity for a bit for you in that case. So please any arguments? I have to note that we did have an idea of adding BGP to the various PoPs, then allowing people to make a tunnel to multiple PoPs and then allowing them to announce their /48 to the PoP. Aggregation would still block the prefix to be globally visible though but it would allow for some backup/failover in the odd case that the PoP breaks. But that is something for the todo-list, which is quite long and involves other things which are more or less more important and useful to fix up, especially as the PoPs usually simply work(tm). > We'll see how long > it takes them to find an excuse to boot some guy who posts on public > mailing lists that he has major problems with their attitude. The problem I have with your writings is the simple fact that what you write is wrong, and even worse it doesn't contain any arguments, not even false ones. It is good though that you wrote it in public and fortunately that somebody attended it to me so that I can actually answer those false claims. I think they call it "slander" what you are trying to do, or was it "libel"? Oh well, dunno, I am no lawyer and that is also not what this is about. Clarification is a good thing though so that other people do not read this wrongly. > A year ago I reported the ``DNS spam'' to Declan's politech list. I googled and found the rather odd message, again containing no real arguments. You reported there that OCCAID had taken a decision that on THEIR network (thus not yours) to apply a certain policy. Simple solution: if you don't like it -> don't use it. > Hurricane Electric tunnels are, for me, not even worth looking at, > because they block irc. I'm not spending $600/mo on Internet so I can > put up with this T-Online/Verizon port blocking bullshit. HOLD ON. So you are spending money for IPv4, but require your IPv6 to be perfect and for free? Come on, that is laughable. > I *will* > pay for IPv6, but this second-class interweb crap completely defeats > the purpose of an experimental protocol. Then PAY for it. Go to NTT, C&W, and a lot of other ISP's who can provide it to you. For the ARIN region, take your pick: http://www.sixxs.net/tools/grh/dfp/arin/ Call them up and ask what they can do for you. As for HE.net itself, I have outed enough nastyness about that already: http://lists.cluenet.de/pipermail/ipv6-ops/2007-March/001024.html note that that message actually contains arguments, next to them being valid too, and I cc'd them so they have a easy chance to respond, unfortunately they are never at home. [..] > some people make a big deal about ``native'' v6, meaning v6 over > the Ethernet cable. Not having tunnels definitely makes routing > problems easier to track down, but I really don't think it's faster > or intangibly ``better'' somehow. The problem is when the two ends > of the tunnel are far apart. The tunnel should only be a couple ms > long, not spanning countries or oceans, so routing is still close > to optimal. It's the rtt, not the tunnel itself, that sucks. Correct, as long as you simply TRY to make it native wherever possible. When it is not possible, then at least try to keep the tunnel inside ASN bounds or only hop over one ASN. In case of end-sites though, one can drop this rule as it can be quite hard to get things arranged and setup with the IPv4 provider as they simply haven't get around to it. > 2. bad neighbor ISP's (cough *Abliene* cough) that fuck up the v6 > routing table. with OCCAID I had packets crossing the Atlantic > twice to get to Hurricane Electric. stupid. OCCAID blamed HE and > said they were doing something wrong and ignoring OCCAID's > complaints. Who knows what the real story is. The true story: Abilene doesn't peer with commericial entities (they are changing this though, finally) as such the only way to get their prefixes is using either a rogue NREN who does pass it on or through one of the commercial transits. As for HE.net<->OCCAID, see the link above. > 3. not maintaining your v6 well. If your site depends on some > ``tunnel broker'' with a dynamic address on your end, then > inevitably the broker machine gets rebooted a couple times a month > and loses your site's state. Over generalization, but I understood that some tunnel brokers indeed do that. SixXS PoPs tend to keep on running for months on and are very stable. If they where not stable we would not let users on them. Also note that GRH (http://www.sixxs.net/tools/grh/) and a number of other tools are in place to monitor that the routing to the rest of the Internet is actually working and yes we actively pursue problems, track them down and get them fixed. > If your tunnel broker client is buggy > and crashes, or isn't running at all, then your v6 goes down for > weeks until someone notices. Problems can be reported and they will be fixed, at least for SixXS that works. Although it is run as best-effort we tend to try and keep it up 24/7 and have adequate monitoring and fixups in place. No guarantees though; if you want an SLA, go to an ISP and pay for it. [..] > a> There's no demand, cause, well, why'd you *want* ipv6. > > I'll pay you $50 extra per month for v6 right now. I want it so I can > reach and be reached by v6-centric friends in Germany (and apparently > also Japan). Ever wondered where most CCC people get their IPv6 from? :) And of course all of that said, in case people do want to really start using IPv6 and testing it out, join the club of a lot of other people going before you, who have no issue with being good netizens and simply join up, it is free as in beer and free as in speech too. Enjoy. Lastly, in case of questions/comments/flames etc, as long as you use arguments, don't hesitate to actually send them on to info at sixxs.net. Thanks & Greets, Jeroen -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 311 bytes Desc: OpenPGP digital signature URL: From alex at pilosoft.com Fri Apr 6 19:20:09 2007 From: alex at pilosoft.com (alex at pilosoft.com) Date: Fri, 6 Apr 2007 19:20:09 -0400 (EDT) Subject: [nycbug-talk] IPv6 NY-US Roll Call In-Reply-To: <4616CAF5.1070405@spaghetti.zurich.ibm.com> Message-ID: On Fri, 6 Apr 2007, Jeroen Massar wrote: > And of course all of that said, in case people do want to really start > using IPv6 and testing it out, join the club of a lot of other people > going before you, who have no issue with being good netizens and simply > join up, it is free as in beer and free as in speech too. Enjoy. This is some serious SCIENCE being dropped here. Many thanks Jeroen. -alex From bonsaime at gmail.com Sat Apr 7 13:33:46 2007 From: bonsaime at gmail.com (Jesse Callaway) Date: Sat, 7 Apr 2007 13:33:46 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. - iSCSI Message-ID: On 4/6/07, Miles Nordin wrote: > >>>>> "sk" == Steven Kreuzer writes: > > sk> - We don't have iSCSI target daemon in the tree, so sharing > sk> ZVOLs via iSCSI is also not supported at this point. > > Is there a working iSCSI initiator for FreeBSD? I was messing around and was using the NetBSD code on OpenBSD for iSCSI. Totally worked. WAs very simple... just had to follow a couple of patches and compile. Email me if you have any questions. It was just as fast as Samba, so I scrapped it since I wasn't doing anything interesting with it and there's no authentication. -jesse From carton at Ivy.NET Sat Apr 7 02:55:23 2007 From: carton at Ivy.NET (Miles Nordin) Date: Sat, 07 Apr 2007 02:55:23 -0400 Subject: [nycbug-talk] IPv6 NY-US Roll Call In-Reply-To: <4616CAF5.1070405@spaghetti.zurich.ibm.com> (Jeroen Massar's message of "Fri, 06 Apr 2007 23:34:29 +0100") References: <5A00D735-3806-4FD9-B5BC-19AFC46D8C15@lesmuug.org> <4616CAF5.1070405@spaghetti.zurich.ibm.com> Message-ID: >>>>> "jm" == Jeroen Massar writes: jm> I think they call it "slander" what you are trying to do, or jm> was it "libel"? .....r...right. jm> Ever wondered where most CCC people get their IPv6 from? :) CCCC is SixXS, but I think it is hard to run a server there because they are trying so much to save power. It's mostly just laptop clients. so I guess the server policy is no problem for them. They mentioned some problem with a ``tunnel points'' hoop they had to jump through, but it sounded like they had it worked out. Berlin is from Packbart I think? some ISP-ish guy who runs a few AS's? I couldn't get a clear answer in Berlin. I think they were working on setting up some new things that were not finished. The Congress gets dark fiber and collects transit from a bunch of friendly ISP's from scratch every year, so I would expect one of the peers throws in v6. not sure what's going on in Hamburg, if they have v6 or not. I wish I'd asked. They did have a workin ChaosPhone though! most of the individuals I know use xs26, though, because they accept anyone, and it ``just works.'' jm> * If you want to define your own rules/requirements: pay for jm> it. [...] jm> Then PAY for it. Go to NTT, C&W, and a lot of other ISP's who jm> can provide it to you. jm> For the ARIN region, take your pick: jm> http://www.sixxs.net/tools/grh/dfp/arin/ If you read the thread here, I'm trying to do just that, as are several other people in NYCBUG. I haven't found any good options. I sent an email to Hurricane asking to pay for a nonsilly tunnel a month ago---no response. I discussed getting v6 from Alex on this list---not happening right away. I sent an email to nLayer today, but I doubt they handle such small pipes. Anyone on this list have experience with NTT or C&W in T1 sizes? wasn't someone saying NTT has v6 but not at all POPs? jm> Do note that we do publish, in whois.sixxs.net, the jm> IPv4 endpoint of every user and with even a slight breeze of jm> an abuse ticket we will terminate the account, yes, this ``slight breeze'' thing is one of my main concerns. My irc server and shell server are available on v4 as well. I'm not ``hiding'' from anyone. And when I ran it under OCCAID, I agreed when I signed up and repeatedly thereafter that I'd take the irc server off IPv6 if there were any DDoS problems with it. Such agreements were the foundation of the Internet I grew up with. It's really sad and frustrating for me to see us moving to this new ``slightest breeze'' system. jm> go to one of the many "IRC Shell Hosting" companies jm> who have "bulletproof" services etc blabla. They tend to have jm> you pay for them btw. Yeah, I know about those. We use those, too. I think we have three or four of them. And yes, we pay for them. And they're a lot cheaper than the T1+colo I was using for OCCAID. And they don't have IPv6. But you're under no obligation to provide a complete solution to everyone's problems. ``I want to run an irc server'' doesn't mean you have to either help me do it yourself or else tell me how to do it without you, and I also don't have to take your advice. so I don't think ``irc shell hosting'' has much to do with the problems of getting reasonable v6 transit in the US. I'm just saying, I'm embarassed to offer some degraded port-blocked or condescending-AUP'ed v6 to my friends in the same way that I wouldn't buy Internet that didn't permit wireless sharing. Also, it doesn't fit my needs because I want to run ipv6 irc and shell servers, and others who want to do the same should be aware. jm> If you really have a 'shell service' which is not meant for jm> IRC, can you explain what that service entails? mail, webpage-serving, hosting 'screen' sessions for irc. There's other chat stuff on there like mcabber and silc. emacs. A few people may write C or Perl programs there if they don't have their own Unix. finger and ytalk still work on my box, like in the old days. . I don't know what other people are doing with their shell servers. Some were doing audio streaming at one point. One guy runs a Jabber server. There's always talk about VoIP, but no one finishes it. but I'm really not interested in second-class Internet access for any price whether it =0 or not, so I'm not sure why I need to justify to you what I'm doing with my shell server, and endorse the IMO silly implication that irc is ``abuse''. jm> I checked, but you clearly never signed up to SixXS and never jm> send an email either. I think you and I had a discussion on the OCCAID list, so it's not exactly fair to imply I just appeared out of nowhere. I was going to sign up until a couple of my friends pointed out irc and shell servers were forbidden, which makes SixXS uninteresting to me, so what's the point? jm> Can you also show which person didn't provide his real name jm> and got rejected? yeah, I could try. I'm a little concerned though because I feel like I'm basically blacklisted from OCCAID/SixXS folk at this point, and I don't want to make my friend into another squeaky wheel to go on your shit list. The stories are common enough that I find them a credible irritant to working with and getting reliable drama-free service from your crew. If others find my experiences too vague without the names and dates, I think I'll leave them with their skepticism. jm> If you don't like that [DNS spam] rule, then don't use the jm> service. That's exactly what I'm doing, for the moment. However, one thing that gets lost in these discussions sometimes is that it's completely reasonable for me to use a service but still complain about it. James has asked me over and over again, if I have such a problem with him claiming to be ``apolitical'' or with ``layer 10 issues,'' why did I keep using OCCAID, and the answer is very simple: there isn't a good alternative, and, as I never fail to mention, by some metrics OCCAID was very good. I think it's important to accept criticism. Based on the way James and other OCCAID folk have responded to my criticism in the past, I'm not really comfortable speaking with full frankness until (1) I'm not doing it on his list where he can emergency-moderate it, and (2) I'm already cut off from OCCAID, so I've got nothing more to lose by failing to grovel to the nerd mafia. jm> Over the years having this little rule in place has saved us jm> from a lot of problems. well, it's not just one rule. It's the DNS ``spam'' rule, which you equate with ``network abuse''. It's the ``slightest breeze'' policy. It's the blocking people whose names don't sound real enough. It's the complicated email address validation stuff. and who knows what else. I just don't want to deal with it, because even after I do get it working, it's just going to go down again when some gentle zephyr trips one of your auto-kiddie-hoop tests and flags me as an abuser, after which my account gets permanently blacklisted. As for the rules saving you from problems, yeah, I get it. You have to be practical. You have to build something that doesn't fall apart. And if we look around, we don't see much of that. but _post hoc ergo propter hoc_: how do you really know which rules have done what when you have so many? My intuition says BGP is a better hoop than any of this other stuff. And the value system you've built around what should be practical value-agnostic rules for which, if anything, you regrettably appologize is definitely not necessary. Instead the definition of this word ``abuse'' has spread like a hungry mold. jm> Clearly you know what DDoS does to your network, and you have jm> had to, according to wikipedia, change ISP's because you where jm> a DDoS target. So why should we, as a free (gratuit, free as jm> in beer) provider allow you to attracked it and then let *OUR* jm> systems get hurt by it? I can't think of a good reason why you would offer v6 for free at all, much less put up with non-customers getting DDoSed. However, I do exactly that same thing, also for free, for a few other people who host shell servers on my shelf, so maybe the answer is ``for the same reason I do,'' whatever that is. >> but I will probably sign up soon to get back some kind of >> censored politicized v6 (albeit without BGP now). jm> Why do you have a need for BGP? Do you have your own ASN? Do jm> you have multiple circuits to the Internet? Do you have a jm> prefix to announce? If not, then why BGP, as it for sure jm> isn't helping connectivity for a bit for you in that case. So jm> please any arguments? I need to learn BGP. This is what OCCAID was invented _for_. only as a second step did it get involved in IPv6. no, obviously you don't have to provide BGP to your users to be a valuable service, but there's no reason I need to justify myself with ``arguments'' before I can be sad that I used to have BGP, and now I can't get it any more. Seriously....this is a Unix user's group. We do not have a cost/benefit analysis for everything we learn, use, or play with. I think SixXS is against the Internet ethic for other reasons, not because it doesn't offer BGP. But I'm...well first, I'm not saying you're doing this exactly. But I've seen others do it, and I suspect it's common. I'm frustrated when netadmin types talk about ``kiddies wanting access to the elite default free zone'' or some such, basically making fun of people for being curious and enthusiastic like that implies some sort of weakness or desperation. It's sort of like a father calling his nine-year-old son a skinny whimp and knocking him around and pushing him down and not letting him win at sports. It's not just unreasonable or harmful. As a geek, I find it absolutely disgusting. You guys seem to start by thinking BGP is some kind of privilege that has to be earned or deserved (which I don't agree with), and then extend it _further_ suggesting that even trying to _earn_ it is pathetic---it has to fall in your lap, or be forced upon you, and only in that way does it become deserved. Imagine if the source code for BSD were a privilege given only to people who proved they had both need and ability to write drivers. We'd be the laughingstock of our community. the awkward ugly girl at the catholic school wearing the abstinence T-shirt that says ``I'm worth waiting for.'' Sorry, that's just not my culture. jm> You clearly seem to have a more personal issue then than jm> anything else. If you want to talk about it, try jm> info at sixxs.net. or I could just post on any mailing list anywhere and wait for you guys to google yourselves. Then you'll yuck it up in private irc channels so I hear back from my friends ``haha those guys are LOLing it up over you again Miles'', and then eventually, as I predicted in my original post, you'll track me down from wherever outside NYC you operate and engage me on our little local user's group list. such service! :) here's what I would say to info@: Q. I would like to sign up, but I run an irc and shell server. Do the rules apply to everyone else, but not me? A. No. Q. shit. BTW I think the rules suck. Anyway seriously thanks for the invitation and the reply. And thanks for the work you've done taking the first few small steps for IPv6 in the many, many years since the basic tools have been available. I just hope we can get some viable competition in this space soon, because I think that will make everyone's attitude mellow out a bit, and better yet will give me the chance to take your repeated ``then PAY for it'' jab seriously. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: From pete at nomadlogic.org Sat Apr 7 14:35:02 2007 From: pete at nomadlogic.org (Pete Wright) Date: Sat, 7 Apr 2007 14:35:02 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. In-Reply-To: References: <28608836-6F48-44BA-B216-7179BE56B298@f2o.org> Message-ID: <20070407183457.GA81395@sunset.nomadlogic.org> On Fri, Apr 06, 2007 at 03:02:57PM -0400, Charles Sprickman wrote: > On Fri, 6 Apr 2007, Steven Kreuzer wrote: > > > Thought I would post this to the list since there has been alot of > > talk about ZFS. > > > > ZFS support was just commited to HEAD as an experimental feature and > > will be available in 7.0-RELEASE > [...] > > - There is no support for booting off of ZFS file system. > > For anyone who's following this closely, is bootable ZFS supposed to > eventually make it's way into 7.x, or will that be for 8.x? > there are a couple things preventing this from happening currently - i'll spare talk@ the details as you can get them from the current@ list. in a nut shell: -zfs is available only as a .ko -zfs needs to read files in /etc/ so the FreeBSD bootloader will need to be modified for that to work. yea i'm pretty excited for sure - i'd like to get this working with FreeNAS :) -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From carton at Ivy.NET Sat Apr 7 15:05:53 2007 From: carton at Ivy.NET (Miles Nordin) Date: Sat, 07 Apr 2007 15:05:53 -0400 Subject: [nycbug-talk] [Potentially Spam] Re: IPv6 NY-US Roll Call In-Reply-To: <007501c7789f$1ab9dcf0$1efc5dd8@HCMC.local> (James Jun's message of "Fri, 6 Apr 2007 18:58:34 -0400") References: <5A00D735-3806-4FD9-B5BC-19AFC46D8C15@lesmuug.org> <4616CAF5.1070405@spaghetti.zurich.ibm.com> <007501c7789f$1ab9dcf0$1efc5dd8@HCMC.local> Message-ID: >>>>> "jj" == James Jun writes: jj> the '404' of URLs happened as we upgrade our web sites. well, here's the politech post: http://seclists.org/politech/2005/Jan/0031.html the URL's worked when I submitted it to Declan, and they quit working a few hours after you discovered it. Also, one of the URLs was for an irc log on my web server, and you privately asked me to 404 it because it embarassed you, and I did. You're welcome! Then I accidentally brought it back a year later because of a CVS screwup, and you asked me to remove it again, and I did. Sorry about that! Anyway it was a long time ago so I'm almost embarassed to remember it more clearly than you, so sorry to drag it up again. It's just that I'm still kind of upset and distrustful after the whole experience though. I guess I'm often a little abrasive and absolutist to a fault in all of these geek groups, but I seem to get along much better with NetBSD and NYCBUG people than netadmin people, whether it's at conferences or OCCAID or ISP sales guys or even job interviews. I interviewed for a Cisco spot once, and the guy must have taken five phone calls, right in front of me, at a 1-hour interview. I was like ``ha ha, that's uh, not funny.'' He got all puffed up and defensive when I suggested something about his WAN might not be optimal, while it seems like a Unix guy would either tell me why he did it, or say ``yeah that makes sense. Oh well, always so much to do.'' or both. He cut me off with ``yeah yeah yeah that's good enough'' in the middle of answers. I felt totally unfree to volunteer things or be casual and honest. i dunno, for me a very different and unfriendly culture. Except Alex and the Quagga people---they're easy to get along with. so I guess I will hang out here until I find a better way back into the world of network stuff. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: From carton at Ivy.NET Sat Apr 7 15:51:15 2007 From: carton at Ivy.NET (Miles Nordin) Date: Sat, 07 Apr 2007 15:51:15 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. - iSCSI In-Reply-To: (Jesse Callaway's message of "Sat, 7 Apr 2007 13:33:46 -0400") References: Message-ID: >>>>> "jc" == Jesse Callaway writes: jc> I was messing around and was using the NetBSD code on OpenBSD jc> for iSCSI. Totally worked. WAs very simple... this is the NetBSD iSCSI target, right? not an initiator? meaning, it offers iSCSI disks to other machines? I was wondering if FreeBSD can mount iSCSI disks off other machines. Long-term I want a machine with ZFS and a good iSCSI initiator, so that I can: 1. build ZFS pools with more disks than I can fit on a single motherboard 2. make vdev stripes across several chassis so I can power down a whole tower case without taking the pool offline 3. cluster the work of geli/cgd/dmcrypt across several CPUs I think it's good to use iSCSI rather than geom-gate or nbd because from watching mailing lists, it sounds like the iSCSI guys are more aggressive about getting tagged queueing, SYNCHRONIZE CACHE, and other weird SCSI commands to work in a precisely correct way that won't corrupt fancy RAIDs and filesystems. I'm not sure they're there yet, though. The iSCSI initiator in Solaris Nevada b44 is not good. It panics the kernel when an iSCSI target that's in use disappears, and sometimes when Linux IET reboots I need to 'iscsiadm discovery-address remove ; iscsiadm discovery-address add ' before it'll work again. but I think there is actually source code for it---kind of rare for big chunks of ``open''solaris. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: From bonsaime at gmail.com Sat Apr 7 16:51:37 2007 From: bonsaime at gmail.com (Jesse Callaway) Date: Sat, 7 Apr 2007 16:51:37 -0400 Subject: [nycbug-talk] ZFS committed to FreeBSD base. - iSCSI In-Reply-To: References: Message-ID: On 4/7/07, Miles Nordin wrote: > >>>>> "jc" == Jesse Callaway writes: > > jc> I was messing around and was using the NetBSD code on OpenBSD > jc> for iSCSI. Totally worked. WAs very simple... > yeah, just the target. I remember when gigE came out people started thinking about iSCSI seriously. > this is the NetBSD iSCSI target, right? not an initiator? meaning, > it offers iSCSI disks to other machines? I was wondering if FreeBSD > can mount iSCSI disks off other machines. > > Long-term I want a machine with ZFS and a good iSCSI initiator, so > that I can: > > 1. build ZFS pools with more disks than I can fit on a single > motherboard > > 2. make vdev stripes across several chassis so I can power down a > whole tower case without taking the pool offline > > 3. cluster the work of geli/cgd/dmcrypt across several CPUs yummy > > I think it's good to use iSCSI rather than geom-gate or nbd because > from watching mailing lists, it sounds like the iSCSI guys are more > aggressive about getting tagged queueing, SYNCHRONIZE CACHE, and other > weird SCSI commands to work in a precisely correct way that won't > corrupt fancy RAIDs and filesystems. > > I'm not sure they're there yet, though. The iSCSI initiator in > Solaris Nevada b44 is not good. It panics the kernel when an iSCSI > target that's in use disappears, and sometimes when Linux IET reboots > I need to 'iscsiadm discovery-address remove ; iscsiadm > discovery-address add ' before it'll work again. but I think > there is actually source code for it---kind of rare for big chunks of > ``open''solaris. If that's the only hangup with Linux IET, I'd run with it. > > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > > > From mikel.king at techally.com Fri Apr 13 12:30:18 2007 From: mikel.king at techally.com (Mikel King) Date: Fri, 13 Apr 2007 12:30:18 -0400 Subject: [nycbug-talk] Printer Epson Stylus c64 USB Ink-jet Message-ID: Greetings all, I figured I give anyone on this list first dibs. I will be reposting the following to FreeCycleNYC in approximately an hour. Offer: Epson Stylus C64 USB Ink-jet Printer In very good condition, although the ink cartridges may be dried up. The printer was taken out of service several months ago and placed in storage. Includes power and USB cable, I have no idea where the install media is therefore the recipient will have to download it from Epson's web site. This printer was used exclusively on a MAC running OS X Tiger, but should work on a PC with Windows XP. If you've foolishly upgraded to Vista, then your mileage will vary. Pick-up Only: M-F (near Union Square) Obviously first come first served. Cheers, Mikel From techneck at goldenpath.org Fri Apr 13 13:42:30 2007 From: techneck at goldenpath.org (Tim A.) Date: Fri, 13 Apr 2007 13:42:30 -0400 Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. Message-ID: <461FC106.8090404@goldenpath.org> In about a years time, the company I work for is moving just over the river to the South Bronx, East 138th street. Shopping for communication services, I can't find any attractive offerings that are worth a damn. Ideally, I'd find the nearest carrier neutral colo and a fiber contractor to connect us. Any of you have any experience with that? I'd like a map of NYC colo's and a list of recommended fiber contractors. From bonsaime at gmail.com Fri Apr 13 14:42:40 2007 From: bonsaime at gmail.com (Jesse Callaway) Date: Fri, 13 Apr 2007 14:42:40 -0400 Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: <461FC106.8090404@goldenpath.org> References: <461FC106.8090404@goldenpath.org> Message-ID: What type of fiber connection are you looking for? Unless you're looking for FIOS you have to dig a tunnel and lay it down if the fiber is not already at the colo. Realistically you have to find a colo which already has the connectivity or peering with your favorite carrier. Why must it be fiber optic, what you want is a dedicated connection, right? Since you don't say you need any specific fiber provider, then you probably don't need fiber. Don't take this as an insult, I just think you need to reevaluate which medium is necessary to support your needs. Likely you could connect using 802.11g if you get down to brass tacks. Nobody would do this, but you get what i'm saying. Look at a colo you like, and ask them what connections are available. Finding a carrier you like and asking them to meet you at a colo is a very big operation where you have to strike a special deal with the colo and the carrier. Unless you are looking for something a tech can just drop by and install like a cable-modem connection, dsl, or a T circuit. In other words, if it's not on the telephone pole outside with the line you want to tap into on it things get way too complicated. questions you need to ask 1. how much bandwidth do I need 2. what do I want to connect to -jesse On 4/13/07, Tim A. wrote: > In about a years time, the company I work for is moving just over the > river to the South Bronx, East 138th street. > Shopping for communication services, I can't find any attractive > offerings that are worth a damn. > > Ideally, I'd find the nearest carrier neutral colo and a fiber > contractor to connect us. > > Any of you have any experience with that? > > I'd like a map of NYC colo's and a list of recommended fiber contractors. > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > From techneck at goldenpath.org Fri Apr 13 17:54:06 2007 From: techneck at goldenpath.org (Tim A.) Date: Fri, 13 Apr 2007 17:54:06 -0400 Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: <461FC106.8090404@goldenpath.org> References: <461FC106.8090404@goldenpath.org> Message-ID: <461FFBFE.1010606@goldenpath.org> Not Manhattan, everything there's underground and more expensive. Bronx we can do poles. I guess ConEd are the guys to talk to. I'm talking about contracting a fiber run to a carrier neutral colo with either title or a long term right. Yes, I know it's a PITA. Right of Way, Municipality Access Agreements, and a seriously expensive installation with maintenance to boot. But looking over the books, if it costs less than 100k to do, it'd pay itself off in 2-3 years and leave us with a hell of a lot more bang for the buck. Tim A. wrote: > In about a years time, the company I work for is moving just over the > river to the South Bronx, East 138th street. > Shopping for communication services, I can't find any attractive > offerings that are worth a damn. > > Ideally, I'd find the nearest carrier neutral colo and a fiber > contractor to connect us. > > Any of you have any experience with that? > > I'd like a map of NYC colo's and a list of recommended fiber contractors. > _______________________________________________ > From alex at pilosoft.com Fri Apr 13 23:31:35 2007 From: alex at pilosoft.com (alex at pilosoft.com) Date: Fri, 13 Apr 2007 23:31:35 -0400 (EDT) Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: <461FFBFE.1010606@goldenpath.org> Message-ID: On Fri, 13 Apr 2007, Tim A. wrote: > Not Manhattan, everything there's underground and more expensive. Bronx > we can do poles. I guess ConEd are the guys to talk to. > > I'm talking about contracting a fiber run to a carrier neutral colo with > either title or a long term right. > > Yes, I know it's a PITA. Right of Way, Municipality Access Agreements, > and a seriously expensive installation with maintenance to boot. > > But looking over the books, if it costs less than 100k to do, it'd pay > itself off in 2-3 years and leave us with a hell of a lot more bang for > the buck. Let me put it this way. It is much cheaper to bring servers to your bandwidth than to bring bandwidth to your servers. Can you briefly explain your requirements? How much bandwidth you need? Is it hundreds of gigabits? How much equipment do you have? That will determine whether it is really cost effective to build. And no, 100k will roughly cover trenching of a single block, certainly not enough to cover a new build from bronx to 111 8th. That'll be in tens of millions of dollars. If you are really serious about getting DF to your facility, you need to get fiber maps from major players. If its "sufficiently near" (as in <10 blocks) to an existing route, it might be worth a try to get them to do some trenching to get to you. But you'll still be paying 10K+$/mo for use of their fiber. Offhand, not that many carriers have dark in the bronx area. CV does, but they probably won't sell dark, only lit services. VZB does, but also, they no longer sell dark, only lit service. MFNX *might* have something that goes to albany, it might be passing somewhere through bronx. Frankly, I think its a very very long shot. With regard to carrier neutrality - are you sure you need that? Generally, carrier neutral facilities are more expensive and generally more hassle to deal with than a facility operated by carrier (because carriers can expect to get some profit on miscellaneous services, and carrier-neutral facility will only make money on your colocation + crossconnects). What you really want is being in a building with large number of carriers in them ("Carrier hotel"). Note that these are usually specific buildings, which have multiple facilities in them, some neutral, most not. In NYC, these are the important ones, in roughly decreasing importance order: * 111 8th: Probably the most important carrier hotel, all major players have their own suites in the building. Unfortunately, getting between the suites in there is complicated. You want to be in NYCC, the building-sanctioned meet-me-room (which was recently sold to TelX), where almost everyone is. * 60 Hudson: Mostly voice-oriented. Number of carrier-neutral facilities: TelX, S&D and building-sanctioned MMR (operated by fibernet). You want to be in TelX - most carriers are there. * 25 Broadway: Only one facility, Telehouse, which is carrier-neutral. There are good and bad sides about telehouse...And its beyond scope of this discussion. You probably don't want to be here. * 32 AoA: There's building meetmeroom, with some selection of carriers. But you don't want to be here. * 601 W 26: Nothing carrier-neutral really. You don't want to be here. * 75 Broad: AccessIT is the only carrier-neutral, but even more ghetto than telehouse. You dont want to be there. * Bunch of other third-rate buildings: Brooklyn Army Terminal, 470 Vanderbilt, with a sparse selection of carriers. You don't want to be there. Disclaimer: we operate a non-neutral facility in 55 Broad St, and we do have dark fiber to 60 Hudson, and 25 Broadway (soon to 111 8th). -- Alex Pilosov | DSL, Colocation, Hosting Services President | alex at pilosoft.com 877-PILOSOFT x601 Pilosoft, Inc. | http://www.pilosoft.com From spork at bway.net Sat Apr 14 00:03:04 2007 From: spork at bway.net (Charles Sprickman) Date: Sat, 14 Apr 2007 00:03:04 -0400 (EDT) Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: References: Message-ID: On Fri, 13 Apr 2007, alex at pilosoft.com wrote: > * 111 8th: Probably the most important carrier hotel, all major players > have their own suites in the building. Unfortunately, getting between the > suites in there is complicated. You want to be in NYCC, the > building-sanctioned meet-me-room (which was recently sold to TelX), where > almost everyone is. Not only complicated to get stuff (I gave up trying to get something from Level3 to Broadwing - they are across the hall from each other), but also expensive. Maybe TelX will be cheaper than the building-management-run facility was. I think part of Alex's point is that you can pick a decent carrier and deal with them. The carrier neutral stuff will not be financially very smart if you're just going to hook up to one or two other people. It seems some of these places also get full at some point, and there's no bargains at that point. > * 25 Broadway: Only one facility, Telehouse, which is carrier-neutral. > There are good and bad sides about telehouse...And its beyond scope of > this discussion. You probably don't want to be here. If you understand the value of the facility, and you're not putting all your eggs in one basket, it can't have gotten that much worse than it was. I have to imagine that by now they got that whole electricity thing figured out (or not...). When I used to be there they had something I've not seen since: one-time cross connect fees. You pay $800 or so for a cat5 run to a neighboring cabinet or $2k for fiber to someone *once* and that's it. Over in Level3 we're paying $250/month on each DS3 to a cabinet that's less than 50' away from our cabinet. Quite a racket, but I guess that's how most places do it. Like Alex said, why build a datacenter when you can rent? If you want bandwidth for the office, you can certainly get something affordable out of your rented datacenter space to your office. I've been really happy with ConEd (now RCN) and their metro ethernet service. We get 10Mb from 111 8th to our soho office for $400 (plus $200 or so in cross-connect fees to Level3). We can bump that to 100Mb for another $600. Much more reliable than any telco-provided circuit I've seen. Pricing may not be as good after RCN bought them, but that's just an example of another option... Charles > -- > Alex Pilosov | DSL, Colocation, Hosting Services > President | alex at pilosoft.com 877-PILOSOFT x601 > Pilosoft, Inc. | http://www.pilosoft.com > > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > From alex at pilosoft.com Sat Apr 14 00:10:43 2007 From: alex at pilosoft.com (alex at pilosoft.com) Date: Sat, 14 Apr 2007 00:10:43 -0400 (EDT) Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: Message-ID: On Sat, 14 Apr 2007, Charles Sprickman wrote: > On Fri, 13 Apr 2007, alex at pilosoft.com wrote: > > > * 111 8th: Probably the most important carrier hotel, all major > > players have their own suites in the building. Unfortunately, getting > > between the suites in there is complicated. You want to be in NYCC, > > the building-sanctioned meet-me-room (which was recently sold to > > TelX), where almost everyone is. > > Not only complicated to get stuff (I gave up trying to get something > from Level3 to Broadwing - they are across the hall from each other), > but also expensive. Maybe TelX will be cheaper than the > building-management-run facility was. Right. To get from one to the other, you will have to go through NYCC. No, the NYCC *is* the building-management-run facility, and that's where you really want to be if you are a carrier. > I think part of Alex's point is that you can pick a decent carrier and > deal with them. The carrier neutral stuff will not be financially very > smart if you're just going to hook up to one or two other people. Exactly. > It seems some of these places also get full at some point, and there's > no bargains at that point. Yes, colo prices are going through the *roof*. 100% increase in about 2 years. > > * 25 Broadway: Only one facility, Telehouse, which is carrier-neutral. > > There are good and bad sides about telehouse...And its beyond scope of > > this discussion. You probably don't want to be here. > > If you understand the value of the facility, and you're not putting all > your eggs in one basket, it can't have gotten that much worse than it > was. I have to imagine that by now they got that whole electricity > thing figured out (or not...). ...not really... > When I used to be there they had something I've not seen since: > one-time cross connect fees. You pay $800 or so for a cat5 run to a > neighboring cabinet or $2k for fiber to someone *once* and that's it. Yes, that's great. :) Plus, since its so ghetto, they won't know if you ran crossconnect yourself at night, etc. Anyway. > Over in Level3 we're paying $250/month on each DS3 to a cabinet that's > less than 50' away from our cabinet. Quite a racket, but I guess that's > how most places do it. > > Like Alex said, why build a datacenter when you can rent? > > If you want bandwidth for the office, you can certainly get something > affordable out of your rented datacenter space to your office. I've > been really happy with ConEd (now RCN) and their metro ethernet service. > We get 10Mb from 111 8th to our soho office for $400 (plus $200 or so in > cross-connect fees to Level3). We can bump that to 100Mb for another > $600. Much more reliable than any telco-provided circuit I've seen. > Pricing may not be as good after RCN bought them, but that's just an > example of another option... Yep, if you are in a lit building, you generally get decent options. If you are in da bronx, eh, "not so much". -alex From spork at bway.net Sat Apr 14 00:33:28 2007 From: spork at bway.net (Charles Sprickman) Date: Sat, 14 Apr 2007 00:33:28 -0400 (EDT) Subject: [nycbug-talk] Carrier Neutral Colocation Facilities, and Fiber Contractors. In-Reply-To: References: Message-ID: On Sat, 14 Apr 2007, alex at pilosoft.com wrote: > On Sat, 14 Apr 2007, Charles Sprickman wrote: >> When I used to be there they had something I've not seen since: >> one-time cross connect fees. You pay $800 or so for a cat5 run to a >> neighboring cabinet or $2k for fiber to someone *once* and that's it. > Yes, that's great. :) Plus, since its so ghetto, they won't know if you > ran crossconnect yourself at night, etc. Anyway. See: http://spork.us/uh-oh-spaghetti-os.jpg That was at least four years ago, I have to imagine the floor is even more stuffed now. C > -alex > > From yusuke at cs.nyu.edu Sat Apr 14 16:45:30 2007 From: yusuke at cs.nyu.edu (Yusuke Shinyama) Date: Sat, 14 Apr 2007 16:45:30 -0400 Subject: [nycbug-talk] How to raise openfiles ulimit in FreeBSD? Message-ID: <20070414204530.27745.11760.yusuke@grape.cs.nyu.edu> Hi, I'm trying to raise the per-user ulimit for openfiles. (FreeBSD) /etc/login.conf defines it unlimited and sysctl kern.maxfilesperproc=11095. But when I login as a normal user, the openfiles is 1024, whereas I get 11095 as root. How this value is determined? (I checked my bashrc.) Thanks, Yusuke From bob at redivi.com Sat Apr 14 16:59:10 2007 From: bob at redivi.com (Bob Ippolito) Date: Sat, 14 Apr 2007 13:59:10 -0700 Subject: [nycbug-talk] How to raise openfiles ulimit in FreeBSD? In-Reply-To: <20070414204530.27745.11760.yusuke@grape.cs.nyu.edu> References: <20070414204530.27745.11760.yusuke@grape.cs.nyu.edu> Message-ID: <6a36e7290704141359p3b1ce5d5i7f47b248bd604ed6@mail.gmail.com> On 4/14/07, Yusuke Shinyama wrote: > Hi, > > I'm trying to raise the per-user ulimit for openfiles. (FreeBSD) > /etc/login.conf defines it unlimited and sysctl kern.maxfilesperproc=11095. > > But when I login as a normal user, the openfiles is 1024, > whereas I get 11095 as root. > > How this value is determined? (I checked my bashrc.) ulimit -n 11095 -bob From yusuke at cs.nyu.edu Sat Apr 14 17:52:19 2007 From: yusuke at cs.nyu.edu (Yusuke Shinyama) Date: Sat, 14 Apr 2007 17:52:19 -0400 Subject: [nycbug-talk] How to raise openfiles ulimit in FreeBSD? In-Reply-To: <6a36e7290704141359p3b1ce5d5i7f47b248bd604ed6@mail.gmail.com> References: <6a36e7290704141359p3b1ce5d5i7f47b248bd604ed6@mail.gmail.com> <20070414204530.27745.11760.yusuke@grape.cs.nyu.edu> Message-ID: <20070414215219.1694.97960.yusuke@grape.cs.nyu.edu> On Sat, 14 Apr 2007 13:59:10 -0700, "Bob Ippolito" wrote: > On 4/14/07, Yusuke Shinyama wrote: > > Hi, > > > > I'm trying to raise the per-user ulimit for openfiles. (FreeBSD) > > /etc/login.conf defines it unlimited and sysctl kern.maxfilesperproc=11095. > > > > But when I login as a normal user, the openfiles is 1024, > > whereas I get 11095 as root. > > > > How this value is determined? (I checked my bashrc.) > > ulimit -n 11095 I can do this as root. But I'd like to do this as a normal user. grape$ su - Password: grape# bash [root at grape ~]# ulimit -n 11095 [root at grape ~]# ulimit -n 11095 [root at grape ~]# exit grape# exit logout grape$ ulimit -n 11095 bash: ulimit: open files: cannot modify limit: Operation not permitted grape$ ulimit -n 1024 Any hints? Thanks Yusuke From josh at rivels.org Sun Apr 15 09:57:46 2007 From: josh at rivels.org (Josh Rivel) Date: Sun, 15 Apr 2007 09:57:46 -0400 Subject: [nycbug-talk] Server hosting? Message-ID: <20070415135746.GB27716@rivels.org> Hello all- So the former employer where I've been hosting my OpenBSD box for the past 5 or 6 years is moving and needs me to get the server out by the end of April. I'm wondering if anyone knows of a facility (hopefully in NYC) that will allow me to host a 2u server. I don't use a lot of bandwidth, it's really just a web site for family photographs and the like, and email for my vanity domain (plus an IRC server, also very low bandwidth, maybe 20 users max) I searched the archives and couldn't find anything, so I apologize if this has been covered already. Thanks, Josh From alex at pilosoft.com Sun Apr 15 10:05:30 2007 From: alex at pilosoft.com (alex at pilosoft.com) Date: Sun, 15 Apr 2007 10:05:30 -0400 (EDT) Subject: [nycbug-talk] Server hosting? In-Reply-To: <20070415135746.GB27716@rivels.org> Message-ID: On Sun, 15 Apr 2007, Josh Rivel wrote: > Hello all- > > So the former employer where I've been hosting my OpenBSD box for the > past 5 or 6 years is moving and needs me to get the server out by the > end of April. > > I'm wondering if anyone knows of a facility (hopefully in NYC) that will > allow me to host a 2u server. I don't use a lot of bandwidth, it's > really just a web site for family photographs and the like, and email > for my vanity domain (plus an IRC server, also very low bandwidth, maybe > 20 users max) > > I searched the archives and couldn't find anything, so I apologize if > this has been covered already. We do that, 75$/mo for anything up to 1A of power. (If your server has 1 CPU and up to 2 drives, it will qualify). Alex Pilosov | DSL, Colocation, Hosting Services President | alex at pilosoft.com 877-PILOSOFT x601 Pilosoft, Inc. | http://www.pilosoft.com From josh at rivels.org Sun Apr 15 10:12:17 2007 From: josh at rivels.org (Josh Rivel) Date: Sun, 15 Apr 2007 10:12:17 -0400 Subject: [nycbug-talk] Server hosting? In-Reply-To: References: <20070415135746.GB27716@rivels.org> Message-ID: <20070415141217.GA21950@rivels.org> Alex- > We do that, 75$/mo for anything up to 1A of power. (If your server has 1 > CPU and up to 2 drives, it will qualify). Great, thanks. I am trying to see if my current employer will let me host it, but I'm trying to keep my options open otherwise. And as a bonus you are not very far from where I work now. I'll be in touch! Josh From jkeen at verizon.net Sun Apr 15 14:51:41 2007 From: jkeen at verizon.net (James Keenan) Date: Sun, 15 Apr 2007 14:51:41 -0400 Subject: [nycbug-talk] [ANNOUNCE]: Stevan Little Speaks on Moose Tues April 17 Message-ID: <87768FD3-5F18-4587-863A-D48AEE749BCC@verizon.net> Those of us who follow developments in Perl may have heard of "Moose" but not know much about it. This month, Perl Seminar NY will have Moose's lead developer speaking. Perl Seminar NY Tuesday, April 17, 6:15-8:15 pm (EDT) at office of NYPC User Group 481 8th Ave (Ramada New Yorker Hotel) Suite 550 between West 34 and 35 Streets Manhattan Speaker: Stevan Little Topic: "Moose: An Object-Oriented Framework for Perl 5" Moose is an object oriented framework for Perl 5 inspired by the Perl 6 object system. Moose gives you the benefits of real OO with a solid meta model -- all built on top of Perl?s own object system. Moose?s main goal is to make Perl 5 OO clean and easy to write without sacrificing the power and flexibility that you have come to expect. Stevan Little is, along with Christian Hansen and Yuval Kogman, the author of Moose: http://search.cpan.org/~stevan/Moose/ Perl Seminar NY: http://tech.groups.yahoo.com/group/perlsemny/ Jim Keenan From nikolai at fetissov.org Mon Apr 16 15:13:34 2007 From: nikolai at fetissov.org (nikolai) Date: Mon, 16 Apr 2007 15:13:34 -0400 (EDT) Subject: [nycbug-talk] How to raise openfiles ulimit in FreeBSD? In-Reply-To: <20070414215219.1694.97960.yusuke@grape.cs.nyu.edu> References: <6a36e7290704141359p3b1ce5d5i7f47b248bd604ed6@mail.gmail.com> <20070414204530.27745.11760.yusuke@grape.cs.nyu.edu> <20070414215219.1694.97960.yusuke@grape.cs.nyu.edu> Message-ID: <25406.63.66.6.15.1176750814.squirrel@www.geekisp.com> > On Sat, 14 Apr 2007 13:59:10 -0700, "Bob Ippolito" wrote: >> On 4/14/07, Yusuke Shinyama wrote: >> > Hi, >> > >> > I'm trying to raise the per-user ulimit for openfiles. (FreeBSD) >> > /etc/login.conf defines it unlimited and sysctl >> kern.maxfilesperproc=11095. >> > >> > But when I login as a normal user, the openfiles is 1024, >> > whereas I get 11095 as root. >> > >> > How this value is determined? (I checked my bashrc.) >> >> ulimit -n 11095 > > I can do this as root. But I'd like to do this as a normal user. > > grape$ su - > Password: > grape# bash > [root at grape ~]# ulimit -n 11095 > [root at grape ~]# ulimit -n > 11095 > [root at grape ~]# exit > grape# exit > logout > > grape$ ulimit -n 11095 > bash: ulimit: open files: cannot modify limit: Operation not permitted > grape$ ulimit -n > 1024 > > Any hints? > ulimit -aH to see hard limits ulimit -aS to see soft limits and check out getrlimit(2) -- nikolai From pete at nomadlogic.org Fri Apr 20 13:55:58 2007 From: pete at nomadlogic.org (Peter Wright) Date: Fri, 20 Apr 2007 10:55:58 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers Message-ID: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> hi all, i've finally lost it and can no longer use my companies exchange solution for work related email. i'm thinking of setting up a solution using fetchmail. ideally i'd like to spool the mail in a courier-imapd store, not local pop. i believe this is possible. does anyone have any pointers for this task? it looks like the fetchmail site at berlios.de is down (you don't have permission to read index.html - how rude)...i've installed fetchmail on my mac and am reading the man page now. thanks! -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From lego at therac25.net Fri Apr 20 23:38:26 2007 From: lego at therac25.net (Andy Michaels) Date: Fri, 20 Apr 2007 23:38:26 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> Message-ID: Hi peter! I'm currently using the uw-imapd server combined with fetchmail for my home mail. The setup is quite simple: client (running Apple Mail) ----> server running imapd and fetchmail -----> internet the server in the middle runs fetchmail and the mail is delivered to that local user's inbox (mbox format). by default, the imap server uses the user's inbox as the imap inbox. The biggest drawback is the delay inherent in this setup. Fetchmail runs every 30 seconds, and checks 2 accounts (one with my old ISP and one with my own domain). It's really not a big deal for me, but I know some people want instant notice of mail arriving :) Installing fetchmail and imapd on your mac will probably work just fine. Although you could also enable imap on the exchange server. Anyway, I'd be glad to provide you with more details. -Andy On Apr 20, 2007, at 1:55 PM, Peter Wright wrote: > > hi all, i've finally lost it and can no longer use my companies > exchange > solution for work related email. i'm thinking of setting up a solution > using fetchmail. ideally i'd like to spool the mail in a courier-imapd > store, not local pop. i believe this is possible. > > does anyone have any pointers for this task? it looks like the > fetchmail > site at berlios.de is down (you don't have permission to read > index.html - > how rude)...i've installed fetchmail on my mac and am reading the man > page > now. > > thanks! > > -p > > -- > ~~oO00Oo~~ > Peter Wright > pete at nomadlogic.org > www.nomadlogic.org/~pete > 310.869.9459 > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month From pete at nomadlogic.org Mon Apr 23 11:30:13 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 08:30:13 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers In-Reply-To: References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> Message-ID: <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> > Hi peter! I'm currently using the uw-imapd server combined with > fetchmail for my home mail. The setup is quite simple: > > client (running Apple Mail) ----> server running imapd and fetchmail > -----> internet > > the server in the middle runs fetchmail and the mail is delivered to > that local user's inbox (mbox format). by default, the imap server > uses the user's inbox as the imap inbox. > > The biggest drawback is the delay inherent in this setup. Fetchmail > runs every 30 seconds, and checks 2 accounts (one with my old ISP and > one with my own domain). It's really not a big deal for me, but I know > some people want instant notice of mail arriving :) > > Installing fetchmail and imapd on your mac will probably work just > fine. Although you could also enable imap on the exchange server. > > Anyway, I'd be glad to provide you with more details. > (please don't top post - no biggie but it helps others on this list. see the guideline section at: http://www.nycbug.org/index.php?NAV=MailingLists) Thanks Andy - I'm pretty comfortable with setting up IMAP and SMTP servers - guess I should have been more specific. 1) any known gotcha's with using fetchmail with exchanges broken IMAP implementation (i.e. non-standards compliant implementation) 2) ideally i'd like to be able to sync b/w my exchange mail store and my fetchmail/imapd store. i.e. - i delete from my fetchmail'd account and those changes get pushed to exchange. this one is asking a bit much - but was hoping someone had run into this problem before. imapd is enabled on our exchange server and is what most of us use - but it's basicly broken (no offence to NT admins out there - it's just the microsoft imap implementation does not work). -p > -Andy > > On Apr 20, 2007, at 1:55 PM, Peter Wright wrote: > >> >> hi all, i've finally lost it and can no longer use my companies >> exchange >> solution for work related email. i'm thinking of setting up a solution >> using fetchmail. ideally i'd like to spool the mail in a courier-imapd >> store, not local pop. i believe this is possible. >> >> does anyone have any pointers for this task? it looks like the >> fetchmail >> site at berlios.de is down (you don't have permission to read >> index.html - >> how rude)...i've installed fetchmail on my mac and am reading the man >> page >> now. >> >> thanks! >> >> -p >> >> -- >> ~~oO00Oo~~ >> Peter Wright >> pete at nomadlogic.org >> www.nomadlogic.org/~pete >> 310.869.9459 >> _______________________________________________ >> % NYC*BUG talk mailing list >> http://lists.nycbug.org/mailman/listinfo/talk >> %Be sure to check out our Jobs and NYCBUG-announce lists >> %We meet the first Wednesday of the month > > _______________________________________________ > % NYC*BUG talk mailing list > http://lists.nycbug.org/mailman/listinfo/talk > %Be sure to check out our Jobs and NYCBUG-announce lists > %We meet the first Wednesday of the month > -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From jlam at pkgsrc.org Mon Apr 23 11:59:39 2007 From: jlam at pkgsrc.org (Johnny C. Lam) Date: Mon, 23 Apr 2007 11:59:39 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> Message-ID: <462CD7EB.3000709@pkgsrc.org> Peter Wright wrote: > 1) any known gotcha's with using fetchmail with exchanges broken IMAP > implementation (i.e. non-standards compliant implementation) You can always fall back to using POP if Exchange supports it. > 2) ideally i'd like to be able to sync b/w my exchange mail store and my > fetchmail/imapd store. i.e. - i delete from my fetchmail'd account and > those changes get pushed to exchange. this one is asking a bit much - but > was hoping someone had run into this problem before. Fetchmail doesn't work in this way; it just grabs mail from the source and dumps it to the sink account that you specify. Regardless of whether you're using POP or IMAP, I believe you can configure fetchmail to delete the message from the server after it's been fetched so you don't have duplicate mails. You might want to look at imapsync instead for what you're trying to do: http://directory.fsf.org/imapsync.html It says it supports MS Exchange 5.5 and Courier-IMAP. Cheers, -- Johnny Lam From pete at nomadlogic.org Mon Apr 23 12:11:04 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 09:11:04 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <462CD7EB.3000709@pkgsrc.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> Message-ID: <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> > Peter Wright wrote: >> 1) any known gotcha's with using fetchmail with exchanges broken IMAP >> implementation (i.e. non-standards compliant implementation) > > You can always fall back to using POP if Exchange supports it. yea i was thinking about that - although i'd like to access this account from multiple machines using various MUA's. > >> 2) ideally i'd like to be able to sync b/w my exchange mail store and my >> fetchmail/imapd store. i.e. - i delete from my fetchmail'd account and >> those changes get pushed to exchange. this one is asking a bit much - >> but >> was hoping someone had run into this problem before. > > Fetchmail doesn't work in this way; it just grabs mail from the source > and dumps it to the sink account that you specify. Regardless of > whether you're using POP or IMAP, I believe you can configure fetchmail > to delete the message from the server after it's been fetched so you > don't have duplicate mails. yea - figured as much. for testing i've been *very* careful not to delete my messages :) > > You might want to look at imapsync instead for what you're trying to do: > > http://directory.fsf.org/imapsync.html > > It says it supports MS Exchange 5.5 and Courier-IMAP. > perfect! i'll have to give this a shot and i'll report back. -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From KReiter at insidefsi.net Mon Apr 23 13:52:00 2007 From: KReiter at insidefsi.net (Kevin Reiter) Date: Mon, 23 Apr 2007 13:52:00 -0400 Subject: [nycbug-talk] FreeBSD syslog.conf quickie In-Reply-To: <184B0715C3D74243B86F872B55C340E703A9126E@fsi32.fsidp.insidefsi.com> Message-ID: <0CF59C4890F7A04AAC3B1E798E6F86F3182B64@fsi32.fsidp.insidefsi.com> kreiter at insidefsi.net wrote: : All, : : A quick question on the syntax for defining a remote system logging : to a remote syslog server. : : I have a number of routers I want to send their syslogs to a FreeBSD : system. I've checked all the firewall rules and know that not to be : a problem. My question is, in the syslog.conf file itself, how to I : define a host to log to a specific logfile? I've RTFM, checked : Google, and checked everything else I thought might have been mucking : up the waters, but can't seem to get this straight. : : I'm looking to get all syslogs for hostname "ssr" to go to : /var/log/router.log : : In my /etc/syslog.conf, I have the following entry: : : +ssr.* /var/log/router.log : :: From man 5 syslog.conf: : : A hostname specification of the form `#+hostname' or `+hostname' means : the following blocks will be applied to messages received from the : speci- : fied hostname. Alternatively, the hostname specification : `#-hostname' or `-hostname' causes the following blocks to be applied : to messages from : any host but the one specified. If the hostname is given as `@', the : local hostname will be used. As for program specifications, multiple : comma-separated values may be specified for hostname specifications. : : I've tried using +ssr, +ssr.*, and a few other variations with no : luck. : : Am I just reading the examples wrong? (In my defense, I've had a : headache for the past 3 days, which hasn't helped..) : : Thanks, : Kev FYI, I fixed it by installing syslog-ng. I was actually surprised that only 1 person replied to this (thanks, Rodrique.) This message may contain confidential or proprietary information and is intended solely for the individual(s) to whom it is addressed. If you are not a named addressee you should not disseminate, distribute or copy this e-mail or act upon the information contained herein. Please notify the sender immediately by e-mail if you have received this e-mail by mistake and delete this e-mail from your system. From pete at nomadlogic.org Mon Apr 23 14:08:33 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 11:08:33 -0700 (PDT) Subject: [nycbug-talk] sun hardware sale Message-ID: <8037.160.33.20.11.1177351713.squirrel@webmail.nomadlogic.org> sorry for the advert - although i know a couple people use sun hardware on this list. maybe this is a good chance to purchase some test Sun hardware for *BSD devs. http://www.sun.com/emrkt/25sale/index.jsp?tfa5101 we have had pretty good success with the T2000, it's a pretty nice piece of hardware for heavily muti-threaded apps. -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From njt at ayvali.org Mon Apr 23 17:02:05 2007 From: njt at ayvali.org (N.J. Thomas) Date: Mon, 23 Apr 2007 17:02:05 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> Message-ID: <20070423210204.GP1311@ayvali.org> * Peter Wright [2007-04-23 09:11:04 -0700]: > > I believe you can configure fetchmail to delete the message from the > > server after it's been fetched so you don't have duplicate mails. > > yea - figured as much. for testing i've been *very* careful not to > delete my messages :) FYI, read the fetchmail section on this page: http://people.ubuntu-in.org/~ghoseb/esr.html Choice quotes: Not that fetchmail is much to boast about in itself. A real Unix hacker has called it, among other things, an abomination before God. Fortunately, there's a simple replacement [ Now hosted at http://pyropus.ca/software/getmail/ ]. Ironically, it's written in ESR's latest craze, python. It goes on to explain why getmail is better than fetchmail. (Full disclosure: I've been using getmail for a couple of years now, but I've not used fetchmail before.) Worth a look in any case. Thomas From pete at nomadlogic.org Mon Apr 23 17:10:22 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 14:10:22 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <20070423210204.GP1311@ayvali.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> <20070423210204.GP1311@ayvali.org> Message-ID: <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> > * Peter Wright [2007-04-23 09:11:04 -0700]: >> > I believe you can configure fetchmail to delete the message from the >> > server after it's been fetched so you don't have duplicate mails. >> >> yea - figured as much. for testing i've been *very* careful not to >> delete my messages :) > > FYI, read the fetchmail section on this page: > > http://people.ubuntu-in.org/~ghoseb/esr.html > > Choice quotes: > > Not that fetchmail is much to boast about in itself. A real Unix > hacker has called it, among other things, an abomination before > God. Fortunately, there's a simple replacement [ Now hosted at > http://pyropus.ca/software/getmail/ ]. Ironically, it's written > in ESR's latest craze, python. > > It goes on to explain why getmail is better than fetchmail. > > (Full disclosure: I've been using getmail for a couple of years now, but > I've not used fetchmail before.) > > Worth a look in any case. > Thanks for the heads up Thomas! ESR was the original fetchmail author no? - so i'll have to take a serious look at getmail. -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From pete at nomadlogic.org Mon Apr 23 17:18:09 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 14:18:09 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> <20070423210204.GP1311@ayvali.org> <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> Message-ID: <35462.160.33.20.11.1177363089.squirrel@webmail.nomadlogic.org> > >> * Peter Wright [2007-04-23 09:11:04 -0700]: >>> > I believe you can configure fetchmail to delete the message from the >>> > server after it's been fetched so you don't have duplicate mails. >>> >>> yea - figured as much. for testing i've been *very* careful not to >>> delete my messages :) >> >> FYI, read the fetchmail section on this page: >> >> http://people.ubuntu-in.org/~ghoseb/esr.html >> >> Choice quotes: >> >> Not that fetchmail is much to boast about in itself. A real Unix >> hacker has called it, among other things, an abomination before >> God. Fortunately, there's a simple replacement [ Now hosted at >> http://pyropus.ca/software/getmail/ ]. Ironically, it's written >> in ESR's latest craze, python. >> >> It goes on to explain why getmail is better than fetchmail. >> >> (Full disclosure: I've been using getmail for a couple of years now, but >> I've not used fetchmail before.) >> >> Worth a look in any case. >> > > Thanks for the heads up Thomas! > > ESR was the original fetchmail author no? - so i'll have to take a serious > look at getmail. > after reading the link, now i get it. i seem to remember the wars going on with ESR and some folks on the net a while back ;) thanks again for the link - i'm checking out getmail now.. -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From okan at demirmen.com Mon Apr 23 17:20:14 2007 From: okan at demirmen.com (Okan Demirmen) Date: Mon, 23 Apr 2007 17:20:14 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> Message-ID: <20070423212014.GF8823@clam.khaoz.org> On Fri 2007.04.20 at 10:55 -0700, Peter Wright wrote: > > hi all, i've finally lost it and can no longer use my companies exchange > solution for work related email. i'm thinking of setting up a solution > using fetchmail. ideally i'd like to spool the mail in a courier-imapd > store, not local pop. i believe this is possible. > > does anyone have any pointers for this task? it looks like the fetchmail > site at berlios.de is down (you don't have permission to read index.html - > how rude)...i've installed fetchmail on my mac and am reading the man page > now. you've gotten lots of replies, but here's a few: isync (c) mailsync (c) offlineimap (python) imapsync (perl) i've never used any of them, so ymmv... From pete at nomadlogic.org Mon Apr 23 17:28:46 2007 From: pete at nomadlogic.org (Peter Wright) Date: Mon, 23 Apr 2007 14:28:46 -0700 (PDT) Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <20070423212014.GF8823@clam.khaoz.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <20070423212014.GF8823@clam.khaoz.org> Message-ID: <59282.160.33.20.11.1177363726.squirrel@webmail.nomadlogic.org> > On Fri 2007.04.20 at 10:55 -0700, Peter Wright wrote: >> >> hi all, i've finally lost it and can no longer use my companies exchange >> solution for work related email. i'm thinking of setting up a solution >> using fetchmail. ideally i'd like to spool the mail in a courier-imapd >> store, not local pop. i believe this is possible. >> >> does anyone have any pointers for this task? it looks like the >> fetchmail >> site at berlios.de is down (you don't have permission to read index.html >> - >> how rude)...i've installed fetchmail on my mac and am reading the man >> page >> now. > > you've gotten lots of replies, but here's a few: > isync (c) > mailsync (c) > offlineimap (python) > imapsync (perl) > > i've never used any of them, so ymmv... thanks okan - i checked out the imapsync perl script and it looks good. mostly because i can acutally read the perl code which will make it easier for me to patch it against our broken exchange server. going to look into isync and mailsync now... -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From scottro at nyc.rr.com Mon Apr 23 18:22:01 2007 From: scottro at nyc.rr.com (Scott Robbins) Date: Mon, 23 Apr 2007 18:22:01 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> <20070423210204.GP1311@ayvali.org> <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> Message-ID: <20070423222201.GE85003@mail.scottro.net> On Mon, Apr 23, 2007 at 02:10:22PM -0700, Peter Wright wrote: > > > > > Thanks for the heads up Thomas! > > ESR was the original fetchmail author no? - so i'll have to take a serious > look at getmail. > Getmail (more properly getmail--oh, darn, I just deleted that message from the program's author on the getmail list) is a nice program. I have a page to quickly get you up to speed, it was listed, when getmail changed from 3.x to 4.x in ports/UPDATING for awhile. :) Oh, not quite the end, I oughta give the page. http://home.nyc.rr.com/computertaijutsu/getmail.html (The reference to the email by the author was on the getmail list--someone called it GetMail and Charles posted that it should be called getmail as someone apparently took the name GetMail for a commercial MS app.) It's an excellent program, the program's author faithfully monitors the getmail mailing list and quickly answers queries. Most bugs are fixed very quickly. For example, I reported one that was fixed either the next day or the day after. -- Scott Robbins PGP keyID EB3467D6 ( 1B48 077D 66F6 9DB0 FDC2 A409 FA54 EB34 67D6 ) gpg --keyserver pgp.mit.edu --recv-keys EB3467D6 Spike: What's this? Sittin' around watching the telly while there's evil still afoot? It's not very industrious of you. I say we go out there and kick a little demon ass! What, can't go without your Buffy, is that it? Let's find her! She is the chosen one, after all. Come on! Vampires! Grrr! Nasty! Let's annihilate them, for justice, and for... the safety of puppies... and Christmas, right? Let's fight that evil! Let's kill something! Oh, come on! From njt at ayvali.org Mon Apr 23 22:03:21 2007 From: njt at ayvali.org (N.J. Thomas) Date: Mon, 23 Apr 2007 22:03:21 -0400 Subject: [nycbug-talk] fetchmail pointers In-Reply-To: <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> References: <35956.160.33.20.11.1177091758.squirrel@webmail.nomadlogic.org> <18769.160.33.20.11.1177342213.squirrel@webmail.nomadlogic.org> <462CD7EB.3000709@pkgsrc.org> <63653.160.33.20.11.1177344664.squirrel@webmail.nomadlogic.org> <20070423210204.GP1311@ayvali.org> <16841.160.33.20.11.1177362622.squirrel@webmail.nomadlogic.org> Message-ID: <20070424020321.GS1311@ayvali.org> * Peter Wright [2007-04-23 14:10:22 -0700]: > > Not that fetchmail is much to boast about in itself. A real Unix > > hacker has called it, among other things, an abomination before God. > > ESR was the original fetchmail author no? - so i'll have to take a > serious look at getmail. Well, according to most reports, he took over the popclient package, changed the name, and added some features. Some people think the features are minor enough that it didn't really warrant all the press coverage that some people (largely, esr himself) gave it. I don't know too much about that, but I do know that fetchmail has had some serious security holes surface over the years that kept me away (even as recent as 2005/2006), so I've stuck with getmail. Thomas From nycbug at cyth.net Tue Apr 24 22:15:18 2007 From: nycbug at cyth.net (Ray Lai) Date: Tue, 24 Apr 2007 22:15:18 -0400 Subject: [nycbug-talk] IPv6 source routing attack Message-ID: <20070425021541.GB20064@cybertron.cyth.net> For all the people who were at the IPv6 discussion last month: http://undeadly.org/cgi?action=article&sid=20070423145143&mode=expanded -Ray- From mspitzer at gmail.com Wed Apr 25 11:48:39 2007 From: mspitzer at gmail.com (Marc Spitzer) Date: Wed, 25 Apr 2007 11:48:39 -0400 Subject: [nycbug-talk] google enhances mysql Message-ID: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> http://www.informationweek.com/news/showArticle.jhtml?articleID=199201237 marc -- Freedom is nothing but a chance to be better. Albert Camus From chsnyder at gmail.com Wed Apr 25 12:15:22 2007 From: chsnyder at gmail.com (csnyder) Date: Wed, 25 Apr 2007 12:15:22 -0400 Subject: [nycbug-talk] google enhances mysql In-Reply-To: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> References: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> Message-ID: On 4/25/07, Marc Spitzer wrote: > http://www.informationweek.com/news/showArticle.jhtml?articleID=199201237 > Or skip the ads and see http://google-code-updates.blogspot.com/2007/04/google-releases-patches-that-enhance.html Interesting approach. They didn't fork MySQL. They didn't wait for the patches to be integrated. They just put them out there and said, "Here, this makes things better." -- Chris Snyder http://chxo.com/ From pete at nomadlogic.org Wed Apr 25 12:25:39 2007 From: pete at nomadlogic.org (Peter Wright) Date: Wed, 25 Apr 2007 09:25:39 -0700 (PDT) Subject: [nycbug-talk] google enhances mysql In-Reply-To: References: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> Message-ID: <21833.160.33.20.11.1177518339.squirrel@webmail.nomadlogic.org> > On 4/25/07, Marc Spitzer wrote: >> http://www.informationweek.com/news/showArticle.jhtml?articleID=199201237 >> > > Or skip the ads and see > http://google-code-updates.blogspot.com/2007/04/google-releases-patches-that-enhance.html > > Interesting approach. They didn't fork MySQL. They didn't wait for the > patches to be integrated. They just put them out there and said, > "Here, this makes things better." > it's funny, this is the second cool thing i've seen a commercial company do to help out an open source DB. skype as recently released a tool set for postgreSQL: https://developer.skype.com/SkypeGarage/DbProjects/SkyTools -p -- ~~oO00Oo~~ Peter Wright pete at nomadlogic.org www.nomadlogic.org/~pete 310.869.9459 From mspitzer at gmail.com Wed Apr 25 12:26:23 2007 From: mspitzer at gmail.com (Marc Spitzer) Date: Wed, 25 Apr 2007 12:26:23 -0400 Subject: [nycbug-talk] google enhances mysql In-Reply-To: References: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> Message-ID: <8c50a3c30704250926t1f6c4ffbn107021ae9c6aff89@mail.gmail.com> On 4/25/07, csnyder wrote: > On 4/25/07, Marc Spitzer wrote: > > http://www.informationweek.com/news/showArticle.jhtml?articleID=199201237 > > > > Or skip the ads and see > http://google-code-updates.blogspot.com/2007/04/google-releases-patches-that-enhance.html > > Interesting approach. They didn't fork MySQL. They didn't wait for the > patches to be integrated. They just put them out there and said, > "Here, this makes things better." Actually they did fork mysql, the patches can not be incorporated into mysqls main trunk without them loosing the ability to dual licence the code, unless google gives mysql ownership of the patch. the exact same thing happened with kde and qt. marc -- Freedom is nothing but a chance to be better. Albert Camus From carton at Ivy.NET Wed Apr 25 13:20:15 2007 From: carton at Ivy.NET (Miles Nordin) Date: Wed, 25 Apr 2007 13:20:15 -0400 Subject: [nycbug-talk] google enhances mysql In-Reply-To: <8c50a3c30704250926t1f6c4ffbn107021ae9c6aff89@mail.gmail.com> (Marc Spitzer's message of "Wed, 25 Apr 2007 12:26:23 -0400") References: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> <8c50a3c30704250926t1f6c4ffbn107021ae9c6aff89@mail.gmail.com> Message-ID: >>>>> "ms" == Marc Spitzer writes: ms> the patches can not be incorporated into mysqls main trunk ms> without them loosing the ability to dual licence the code, HA! very nice. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 304 bytes Desc: not available URL: From mspitzer at gmail.com Wed Apr 25 16:06:15 2007 From: mspitzer at gmail.com (Marc Spitzer) Date: Wed, 25 Apr 2007 16:06:15 -0400 Subject: [nycbug-talk] google enhances mysql In-Reply-To: References: <8c50a3c30704250848l1920a9edt1b0051a01fc7807@mail.gmail.com> <8c50a3c30704250926t1f6c4ffbn107021ae9c6aff89@mail.gmail.com> Message-ID: <8c50a3c30704251306r24c7f02bv7d8d37c353802722@mail.gmail.com> On 4/25/07, Miles Nordin wrote: > >>>>> "ms" == Marc Spitzer writes: > > ms> the patches can not be incorporated into mysqls main trunk > ms> without them loosing the ability to dual licence the code, > > HA! very nice. And mysql did it to them selves by releasing gpled code, I think that is irony. marc -- Freedom is nothing but a chance to be better. Albert Camus