From mcevoy.pat at gmail.com Thu May 2 07:11:35 2019 From: mcevoy.pat at gmail.com (Pat McEvoy) Date: Thu, 2 May 2019 07:11:35 -0400 Subject: [talk] Unpaid bar bill... Message-ID: <1F0709FF-010C-444A-BAD4-6A94E728780C@gmail.com> Looks like someone walked out and forgot to pay for a few drinks. You can call Suspenders and they can take a card over the phone. +1 (212) 732-5005 -------------- next part -------------- A non-text attachment was scrubbed... Name: image1.jpeg Type: image/jpeg Size: 63881 bytes Desc: not available URL: -------------- next part -------------- Patrick McEvoy From mcevoy.pat at gmail.com Wed May 1 21:35:45 2019 From: mcevoy.pat at gmail.com (Pat McEvoy) Date: Wed, 1 May 2019 21:35:45 -0400 Subject: [talk] Bar bill... Message-ID: <5821B5E8-6B01-4800-9A25-1A570AE33EF1@gmail.com> Looks like someone walked out and forgot to pay for a few drinks. -------------- next part -------------- A non-text attachment was scrubbed... Name: image1.jpeg Type: image/jpeg Size: 413449 bytes Desc: not available URL: -------------- next part -------------- Patrick McEvoy From steve.b at osfda.org Wed May 8 16:34:05 2019 From: steve.b at osfda.org (Steve) Date: Wed, 8 May 2019 16:34:05 -0400 Subject: [talk] pg_ssl_init... In-Reply-To: <3D6AD610-6B86-4F69-94A7-88D995A1C96D@gmail.com> References: <3D6AD610-6B86-4F69-94A7-88D995A1C96D@gmail.com> Message-ID: Hey, guys- I was the fellow from the "OSFDA" trade association that gave a shout-out at the last meeting, requesting testing of a utility for postgres+SSL. Our trade association: https://osfda.org Our calendar of NY open source developer events that would be of interest to anyone in finance: https://osfda.org/calendar And here are the utilities I mentioned that should get testing on BSD (similar to what letsencrypt did, they're written in python -3...): https://gitlab.com/osfda/pg_ssl_init Thanks! On 4/30/2019 9:43 AM, Pat McEvoy wrote: > Details here: > https://www.nycbug.org/index?action=view&id=10667 > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk -------------- next part -------------- An HTML attachment was scrubbed... URL: From george at ceetonetechnology.com Wed May 8 20:46:00 2019 From: george at ceetonetechnology.com (George Rosamond) Date: Thu, 09 May 2019 00:46:00 +0000 Subject: [talk] Fwd: [UPDATE] Tor Browser 8.0.9 In-Reply-To: References: Message-ID: <97224977-c468-c762-d7bb-7478e453dc39@ceetonetechnology.com> If anyone didn't hear about the Mozilla mess, this is the diff to tor browser 8.0.8 that bring it to 8.0.9 and fixes for OpenBSD. For those running an up-to-date OpenBSD snapshot, the patch is attached. Especially if you have a decently fast box. Updated ports tree required. It's high-time that we started pinging talk@ about TB work for testing. g -------- Forwarded Message -------- Subject: [UPDATE] Tor Browser 8.0.9 Date: Wed, 08 May 2019 19:41:46 -0500 From: attila Lightly tested under amd64. Pax, -A -------------- next part -------------- Index: www/tor-browser/Makefile.inc =================================================================== RCS file: /cvs/ports/www/tor-browser/Makefile.inc,v retrieving revision 1.16 diff -u -p -r1.16 Makefile.inc --- www/tor-browser/Makefile.inc 1 May 2019 17:36:30 -0000 1.16 +++ www/tor-browser/Makefile.inc 8 May 2019 19:01:18 -0000 @@ -6,7 +6,7 @@ MASTER_SITES ?= https://bits.torbsd.org PERMIT_PACKAGE_CDROM ?= Yes CATEGORIES = www BROWSER_NAME = tor-browser -TB_VERSION = 8.0.8 +TB_VERSION = 8.0.9 TB_PREFIX = tb SUBST_VARS += BROWSER_NAME TB_VERSION Index: www/tor-browser/browser/distinfo =================================================================== RCS file: /cvs/ports/www/tor-browser/browser/distinfo,v retrieving revision 1.15 diff -u -p -r1.15 distinfo --- www/tor-browser/browser/distinfo 1 May 2019 17:36:30 -0000 1.15 +++ www/tor-browser/browser/distinfo 8 May 2019 19:01:18 -0000 @@ -1,2 +1,2 @@ -SHA256 (mozilla/tor-browser-8.0.8.tar.gz) = cLUrSyd9M80aNGbtm+CnQzKL6FaQsSghQw8TnQbDv90= -SIZE (mozilla/tor-browser-8.0.8.tar.gz) = 387583620 +SHA256 (mozilla/tor-browser-8.0.9.tar.gz) = B2YOSrwnXSihuHRczush6oYeo9LPFSJGNqoTlE3cmAI= +SIZE (mozilla/tor-browser-8.0.9.tar.gz) = 387604509 Index: www/tor-browser/https-everywhere/Makefile =================================================================== RCS file: /cvs/ports/www/tor-browser/https-everywhere/Makefile,v retrieving revision 1.14 diff -u -p -r1.14 Makefile --- www/tor-browser/https-everywhere/Makefile 1 May 2019 17:36:30 -0000 1.14 +++ www/tor-browser/https-everywhere/Makefile 8 May 2019 19:01:18 -0000 @@ -2,6 +2,7 @@ ADDON_NAME = https-everywhere V = 2019.1.31 +REVISION = 0 COMMENT = Tor Browser add-on: force https where possible HOMEPAGE = https://www.eff.org/https-everywhere MASTER_SITES = https://www.eff.org/files/ Index: www/tor-browser/noscript/Makefile =================================================================== RCS file: /cvs/ports/www/tor-browser/noscript/Makefile,v retrieving revision 1.15 diff -u -p -r1.15 Makefile --- www/tor-browser/noscript/Makefile 1 May 2019 17:36:30 -0000 1.15 +++ www/tor-browser/noscript/Makefile 8 May 2019 19:01:18 -0000 @@ -1,7 +1,7 @@ # $OpenBSD: Makefile,v 1.15 2019/05/01 17:36:30 landry Exp $ ADDON_NAME = noscript -V = 10.2.4 +V = 10.6.1 COMMENT = Tor Browser add-on: flexible JS blocker HOMEPAGE = http://noscript.net MASTER_SITES = https://secure.informaction.com/download/releases/ Index: www/tor-browser/noscript/distinfo =================================================================== RCS file: /cvs/ports/www/tor-browser/noscript/distinfo,v retrieving revision 1.12 diff -u -p -r1.12 distinfo --- www/tor-browser/noscript/distinfo 1 May 2019 17:36:30 -0000 1.12 +++ www/tor-browser/noscript/distinfo 8 May 2019 19:01:18 -0000 @@ -1,2 +1,2 @@ -SHA256 (noscript-10.2.4.xpi) = n1W+zeIzA0jINUusEs4pXcHlg49vxhYUpHbHeF9dQ/Q= -SIZE (noscript-10.2.4.xpi) = 502546 +SHA256 (noscript-10.6.1.xpi) = sVBH0ARdEvKLLh5ES9uGgAJXtcpsyPTIAisgxVDNlyc= +SIZE (noscript-10.6.1.xpi) = 519725 Index: www/tor-browser/tor-launcher/Makefile =================================================================== RCS file: /cvs/ports/www/tor-browser/tor-launcher/Makefile,v retrieving revision 1.14 diff -u -p -r1.14 Makefile --- www/tor-browser/tor-launcher/Makefile 1 May 2019 17:36:30 -0000 1.14 +++ www/tor-browser/tor-launcher/Makefile 8 May 2019 19:01:18 -0000 @@ -2,7 +2,7 @@ ADDON_NAME = tor-launcher V = 0.2.16.6 -REVISION = 0 +REVISION = 1 COMMENT = Tor Browser add-on to manage tor instance GUID = tor-launcher at torproject.org PKGNAME = ${TB_NAME} Index: www/tor-browser/torbutton/Makefile =================================================================== RCS file: /cvs/ports/www/tor-browser/torbutton/Makefile,v retrieving revision 1.14 diff -u -p -r1.14 Makefile --- www/tor-browser/torbutton/Makefile 1 May 2019 17:36:30 -0000 1.14 +++ www/tor-browser/torbutton/Makefile 8 May 2019 19:01:18 -0000 @@ -1,7 +1,7 @@ # $OpenBSD: Makefile,v 1.14 2019/05/01 17:36:30 landry Exp $ ADDON_NAME = torbutton -V = 2.0.11 +V = 2.0.13 COMMENT = Tor Browser add-on for configuring Tor Browser settings GUID = torbutton at torproject.org PKGNAME = ${TB_NAME} Index: www/tor-browser/torbutton/distinfo =================================================================== RCS file: /cvs/ports/www/tor-browser/torbutton/distinfo,v retrieving revision 1.13 diff -u -p -r1.13 distinfo --- www/tor-browser/torbutton/distinfo 1 May 2019 17:36:30 -0000 1.13 +++ www/tor-browser/torbutton/distinfo 8 May 2019 19:01:18 -0000 @@ -1,2 +1,2 @@ -SHA256 (torbutton-2.0.11.tar.gz) = gDISXcp7ySxiIoMeztczI334baQxUFIXzqdwUymix0w= -SIZE (torbutton-2.0.11.tar.gz) = 671438 +SHA256 (torbutton-2.0.13.tar.gz) = g5tTfy3LpE7MekZeCVZ8/4a6DDZs/nb4aIJzQMq8DR0= +SIZE (torbutton-2.0.13.tar.gz) = 669174 Index: meta/tor-browser/Makefile =================================================================== RCS file: /cvs/ports/meta/tor-browser/Makefile,v retrieving revision 1.16 diff -u -p -r1.16 Makefile --- meta/tor-browser/Makefile 1 May 2019 17:36:30 -0000 1.16 +++ meta/tor-browser/Makefile 8 May 2019 19:01:18 -0000 @@ -4,12 +4,13 @@ COMMENT= Tor Browser meta package MAINTAINER= Sean Levy -PKGNAME= tor-browser-8.0.8 +PKGNAME= tor-browser-8.0.9 -RUN_DEPENDS= www/tor-browser/browser>=8.0.8 \ - www/tor-browser/torbutton>=2.0.11 \ - www/tor-browser/tor-launcher>=0.2.16.6p0 \ - www/tor-browser/noscript>=10.2.4 \ - www/tor-browser/https-everywhere>=2019.1.31 +RUN_DEPENDS= www/tor-browser/browser>=8.0.9 \ + www/tor-browser/torbutton>=2.0.13 \ + www/tor-browser/tor-launcher>=0.2.16.6p1 \ + www/tor-browser/noscript>=10.6.1 \ + www/tor-browser/https-everywhere>=2019.1.31p0 \ + net/tor>=0.3.5.8 .include From george at ceetonetechnology.com Fri May 10 09:18:00 2019 From: george at ceetonetechnology.com (George Rosamond) Date: Fri, 10 May 2019 13:18:00 +0000 Subject: [talk] BSDCan Updates Message-ID: <4362221d-d5f9-de43-7d82-5397102a7edb@ceetonetechnology.com> Dan says... > Hello BSDCan enthusiast, > > I write with updates for BSDCan 2019. > > 1 - Newcomers orientation and mentorship > 2 - Social event & menu > 3 - Live streaming > 4 - Goat BOF > > > The schedule can be found here: https://www.bsdcan.org/2019/schedule/ > > 1 - Newcomers orientation and mentorship > > Newcomers orientation and mentorship: If this is your first time at BSDCan, we highly recommend this session on Thursday night at 6PM in DMS 1140. Experienced attendees are encouraged to mentor first-timers and help out. See https://www.bsdcan.org/2019/schedule/events/1140.en.html > > 2 - Social event & menu > > The menu details for the social event on Saturday night have been posted. Saturday 18 May from 6:45 PM at Sens House - 73 York Street, Ottawa > > More details and the menu outline are at https://www.bsdcan.org/2019/schedule/events/1157.en.html > > The menu will be updated with a vegan dinner and vegan canapes. I will post to this list once those details are available. > > If you have dietary needs which are not met by that menu, please let us know via email to info at bsdcan.org with subject "Meals" as soon as you can. The venue is happy to help but we need to let them know by end of Monday 13th. This applies both to the social event and the tutorials/talks. If you need something, please let us know. > > The venue will not have music. Conversation should be easy, not shouted over music. > > Non-alcoholic beverages will be available free of charge. > > > 3 - Live streaming > > There will be live-streaming available. Please watch our Twitter account for details: https://twitter.com/BSDCan > > > 4 - Goat BOF > > If you arrive before the tutorials start, be sure to attend the world-famous Goat BOF on 14 May at the Royal Oak at 161 Laurier Ave East. If case you are wondering what a Goat BOF is, this is how it got started: https://twitter.com/dlangille/status/462732293686452224 > > > Thank you > > Have a great time at BSDCan 2019. From raulcuza at gmail.com Fri May 10 14:00:54 2019 From: raulcuza at gmail.com (Raul Cuza) Date: Fri, 10 May 2019 14:00:54 -0400 Subject: [talk] kill -0 Message-ID: Hola People Who Use Shells, I just had someone submit a shell script using `kill -0 ${PID}` to test if a process is running. It exits non-zero if ${PID} isn't a running process and zero otherwise. But... http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. The ubuntu man page mentions it poorly with "Particularly useful signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill --list` there is no signal 0. And stackoverflow says https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do matching what the person who submitted the PR says. Am I wrong to think this should be accomplished in a way with better documentation? I feel like I am nit picking. Ra?l From jschauma at netmeister.org Fri May 10 14:09:37 2019 From: jschauma at netmeister.org (Jan Schaumann) Date: Fri, 10 May 2019 14:09:37 -0400 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: <20190510180936.GU13229@netmeister.org> Raul Cuza wrote: > I just had someone submit a shell script using `kill -0 ${PID}` to > test if a process is running. It exits non-zero if ${PID} isn't a > running process and zero otherwise. But... > > http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. kill(2) describes this behavior: "sig may be one of the signals specified in sigaction(2) or it may be 0, in which case error checking is performed but no signal is actually sent. This can be used to check the validity of pid." -Jan From raulcuza at gmail.com Fri May 10 14:13:27 2019 From: raulcuza at gmail.com (Raul Cuza) Date: Fri, 10 May 2019 14:13:27 -0400 Subject: [talk] kill -0 In-Reply-To: <20190510180936.GU13229@netmeister.org> References: <20190510180936.GU13229@netmeister.org> Message-ID: On Fri, May 10, 2019 at 2:09 PM Jan Schaumann wrote: > > Raul Cuza wrote: > > > I just had someone submit a shell script using `kill -0 ${PID}` to > > test if a process is running. It exits non-zero if ${PID} isn't a > > running process and zero otherwise. But... > > > > http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. > > kill(2) describes this behavior: > > "sig may be one of the signals specified in > sigaction(2) or it may be 0, in which case error > checking is performed but no signal is actually sent. > This can be used to check the validity of pid." > > -Jan I will update my brain and not be "that PR reviewer". Thank you! R. From pete at nomadlogic.org Fri May 10 14:59:41 2019 From: pete at nomadlogic.org (Pete Wright) Date: Fri, 10 May 2019 11:59:41 -0700 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: On 2019-05-10 11:00, Raul Cuza wrote: > Hola People Who Use Shells, > > I just had someone submit a shell script using `kill -0 ${PID}` to > test if a process is running. It exits non-zero if ${PID} isn't a > running process and zero otherwise. But... > > http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. > > The ubuntu man page mentions it poorly with "Particularly useful > signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill > --list` there is no signal 0. > > And stackoverflow says > https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do > matching what the person who submitted the PR says. > > Am I wrong to think this should be accomplished in a way with better > documentation? I feel like I am nit picking. this seems like a bash'ism - from the bash manpage: SIGNALS ?????? When bash is interactive, in the absence of any traps, it ignores ?????? SIGTERM (so that kill 0 does not kill an interactive shell), and SIGINT ?????? is caught and handled (so that the wait builtin is interruptible).? In ?????? all cases, bash ignores SIGQUIT.? If job control is in effect, bash ?????? ignores SIGTTIN, SIGTTOU, and SIGTSTP. -p -- Pete Wright pete at nomadlogic.org @nomadlogicLA From pete at nomadlogic.org Fri May 10 15:05:13 2019 From: pete at nomadlogic.org (Pete Wright) Date: Fri, 10 May 2019 12:05:13 -0700 Subject: [talk] historical unix video for friday afternoon Message-ID: <111cf3b6-4cc3-3161-1bcd-e739fdf526e7@nomadlogic.org> stumbled across this video today on youtube.? might be a good way to spend "don't touch prod fridays" :) https://www.youtube.com/watch?v=XvDZLjaCJuw&pbjreload=10 -pete -- Pete Wright pete at nomadlogic.org @nomadlogicLA From pete at nomadlogic.org Fri May 10 15:06:27 2019 From: pete at nomadlogic.org (Pete Wright) Date: Fri, 10 May 2019 12:06:27 -0700 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: On 2019-05-10 11:59, Pete Wright wrote: > > > On 2019-05-10 11:00, Raul Cuza wrote: >> Hola People Who Use Shells, >> >> I just had someone submit a shell script using `kill -0 ${PID}` to >> test if a process is running. It exits non-zero if ${PID} isn't a >> running process and zero otherwise. But... >> >> http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. >> >> The ubuntu man page mentions it poorly with "Particularly useful >> signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill >> --list` there is no signal 0. >> >> And stackoverflow says >> https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do >> >> matching what the person who submitted the PR says. >> >> Am I wrong to think this should be accomplished in a way with better >> documentation? I feel like I am nit picking. > > this seems like a bash'ism - from the bash manpage: > > SIGNALS > ?????? When bash is interactive, in the absence of any traps, it ignores > ?????? SIGTERM (so that kill 0 does not kill an interactive shell), > and SIGINT > ?????? is caught and handled (so that the wait builtin is > interruptible).? In > ?????? all cases, bash ignores SIGQUIT.? If job control is in effect, > bash > ?????? ignores SIGTTIN, SIGTTOU, and SIGTSTP. > > heh wish i saw Jan's post before i wrote this as his reply seems to answer the real question... -p -- Pete Wright pete at nomadlogic.org @nomadlogicLA From edlinuxguru at gmail.com Fri May 10 15:31:45 2019 From: edlinuxguru at gmail.com (Edward Capriolo) Date: Fri, 10 May 2019 15:31:45 -0400 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: On Fri, May 10, 2019 at 3:06 PM Pete Wright wrote: > > > On 2019-05-10 11:59, Pete Wright wrote: > > > > > > On 2019-05-10 11:00, Raul Cuza wrote: > >> Hola People Who Use Shells, > >> > >> I just had someone submit a shell script using `kill -0 ${PID}` to > >> test if a process is running. It exits non-zero if ${PID} isn't a > >> running process and zero otherwise. But... > >> > >> http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. > >> > >> The ubuntu man page mentions it poorly with "Particularly useful > >> signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill > >> --list` there is no signal 0. > >> > >> And stackoverflow says > >> > https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do > >> > >> matching what the person who submitted the PR says. > >> > >> Am I wrong to think this should be accomplished in a way with better > >> documentation? I feel like I am nit picking. > > > > this seems like a bash'ism - from the bash manpage: > > > > SIGNALS > > When bash is interactive, in the absence of any traps, it ignores > > SIGTERM (so that kill 0 does not kill an interactive shell), > > and SIGINT > > is caught and handled (so that the wait builtin is > > interruptible). In > > all cases, bash ignores SIGQUIT. If job control is in effect, > > bash > > ignores SIGTTIN, SIGTTOU, and SIGTSTP. > > > > > > heh wish i saw Jan's post before i wrote this as his reply seems to > answer the real question... > > -p > > -- > Pete Wright > pete at nomadlogic.org > @nomadlogicLA > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk > It feels safer to me to use ps route. Even if you are counting the os to preserve the 0 behaviour I feel I would fat finger it and send a kill -1 or 2 -------------- next part -------------- An HTML attachment was scrubbed... URL: From fire at firecrow.com Fri May 10 15:37:17 2019 From: fire at firecrow.com (fire crow) Date: Fri, 10 May 2019 15:37:17 -0400 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: On Fri, May 10, 2019, 15:32 Edward Capriolo wrote: > > > On Fri, May 10, 2019 at 3:06 PM Pete Wright wrote: > >> >> >> On 2019-05-10 11:59, Pete Wright wrote: >> > >> > >> > On 2019-05-10 11:00, Raul Cuza wrote: >> >> Hola People Who Use Shells, >> >> >> >> I just had someone submit a shell script using `kill -0 ${PID}` to >> >> test if a process is running. It exits non-zero if ${PID} isn't a >> >> running process and zero otherwise. But... >> >> >> >> http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. >> >> >> >> The ubuntu man page mentions it poorly with "Particularly useful >> >> signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill >> >> --list` there is no signal 0. >> >> >> >> And stackoverflow says >> >> >> https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do >> >> >> >> matching what the person who submitted the PR says. >> >> >> >> Am I wrong to think this should be accomplished in a way with better >> >> documentation? I feel like I am nit picking. >> > >> > this seems like a bash'ism - from the bash manpage: >> > >> > SIGNALS >> > When bash is interactive, in the absence of any traps, it ignores >> > SIGTERM (so that kill 0 does not kill an interactive shell), >> > and SIGINT >> > is caught and handled (so that the wait builtin is >> > interruptible). In >> > all cases, bash ignores SIGQUIT. If job control is in effect, >> > bash >> > ignores SIGTTIN, SIGTTOU, and SIGTSTP. >> > >> > >> >> heh wish i saw Jan's post before i wrote this as his reply seems to >> answer the real question... >> >> -p >> >> -- >> Pete Wright >> pete at nomadlogic.org >> @nomadlogicLA >> >> _______________________________________________ >> talk mailing list >> talk at lists.nycbug.org >> http://lists.nycbug.org:8080/mailman/listinfo/talk >> > > > It feels safer to me to use ps route. Even if you are counting the os to > preserve the 0 behaviour I feel I would fat finger it and send a kill -1 or > 2 > or more likely fat finger to 9 which is next to 0 on a keyboard ~fire > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From steve.b at osfda.org Fri May 10 16:03:14 2019 From: steve.b at osfda.org (Steve) Date: Fri, 10 May 2019 16:03:14 -0400 Subject: [talk] historical unix video for friday afternoon In-Reply-To: <111cf3b6-4cc3-3161-1bcd-e739fdf526e7@nomadlogic.org> References: <111cf3b6-4cc3-3161-1bcd-e739fdf526e7@nomadlogic.org> Message-ID: <9e6bddb9-2ba1-6b84-d563-fcf01a712a20@osfda.org> Love the "Tube Man" graphic at the end. Another computer archeology video: /*APL*/ (1975...) On 5/10/2019 3:05 PM, Pete Wright wrote: > stumbled across this video today on youtube.? might be a good way to > spend "don't touch prod fridays" :) > > https://www.youtube.com/watch?v=XvDZLjaCJuw&pbjreload=10 > > -pete > -------------- next part -------------- An HTML attachment was scrubbed... URL: From raulcuza at gmail.com Fri May 10 16:42:04 2019 From: raulcuza at gmail.com (Raul Cuza) Date: Fri, 10 May 2019 16:42:04 -0400 Subject: [talk] kill -0 In-Reply-To: References: Message-ID: On Fri, May 10, 2019 at 3:37 PM fire crow wrote: > > > > On Fri, May 10, 2019, 15:32 Edward Capriolo wrote: >> >> >> >> On Fri, May 10, 2019 at 3:06 PM Pete Wright wrote: >>> >>> >>> >>> On 2019-05-10 11:59, Pete Wright wrote: >>> > >>> > >>> > On 2019-05-10 11:00, Raul Cuza wrote: >>> >> Hola People Who Use Shells, >>> >> >>> >> I just had someone submit a shell script using `kill -0 ${PID}` to >>> >> test if a process is running. It exits non-zero if ${PID} isn't a >>> >> running process and zero otherwise. But... >>> >> >>> >> http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. >>> >> >>> >> The ubuntu man page mentions it poorly with "Particularly useful >>> >> signals include HUP, INT, KILL, STOP, CONT, and 0." Looking at `kill >>> >> --list` there is no signal 0. >>> >> >>> >> And stackoverflow says >>> >> https://stackoverflow.com/questions/11012527/what-does-kill-0-pid-in-a-shell-script-do >>> >> >>> >> matching what the person who submitted the PR says. >>> >> >>> >> Am I wrong to think this should be accomplished in a way with better >>> >> documentation? I feel like I am nit picking. >>> > >>> > this seems like a bash'ism - from the bash manpage: >>> > >>> > SIGNALS >>> > When bash is interactive, in the absence of any traps, it ignores >>> > SIGTERM (so that kill 0 does not kill an interactive shell), >>> > and SIGINT >>> > is caught and handled (so that the wait builtin is >>> > interruptible). In >>> > all cases, bash ignores SIGQUIT. If job control is in effect, >>> > bash >>> > ignores SIGTTIN, SIGTTOU, and SIGTSTP. >>> > >>> > >>> >>> heh wish i saw Jan's post before i wrote this as his reply seems to >>> answer the real question... >>> >>> -p >>> >>> -- >>> Pete Wright >>> pete at nomadlogic.org >>> @nomadlogicLA >>> >>> _______________________________________________ >>> talk mailing list >>> talk at lists.nycbug.org >>> http://lists.nycbug.org:8080/mailman/listinfo/talk >> >> >> >> It feels safer to me to use ps route. Even if you are counting the os to preserve the 0 behaviour I feel I would fat finger it and send a kill -1 or 2 > > > or more likely fat finger to 9 which is next to 0 on a keyboard > ~fire >> That's why I only run `sudo` with the partial differential equation solver challenge plugin. If I can't solve the problem, I gotta go home. From njt at ayvali.org Fri May 10 16:49:49 2019 From: njt at ayvali.org (N.J. Thomas) Date: Fri, 10 May 2019 13:49:49 -0700 Subject: [talk] historical unix video for friday afternoon In-Reply-To: <111cf3b6-4cc3-3161-1bcd-e739fdf526e7@nomadlogic.org> References: <111cf3b6-4cc3-3161-1bcd-e739fdf526e7@nomadlogic.org> Message-ID: <20190510204949.GA91241@ayvali.org> * Pete Wright [2019-05-10 12:05:13-0700]: > stumbled across this video today on youtube.? might be a good way to > spend "don't touch prod fridays" :) > > https://www.youtube.com/watch?v=XvDZLjaCJuw&pbjreload=10 If you guys like stuff like this, join the TUHS mailing list: https://www.tuhs.org/ They have really cool threads from many of the folks who were there. Thomas From kmsujit at gmail.com Sat May 11 07:42:39 2019 From: kmsujit at gmail.com (Sujit K M) Date: Sat, 11 May 2019 17:12:39 +0530 Subject: [talk] AWS vs Azure Message-ID: On a recent discussion I was rightly pointed to a lot of FreeBSD work that went into FreeBSD for AWS. I might not be that correct saying AWS is there for FreeBSD 11. Now on parallel, I checked a kernel built install for Linux Kernel 5.0 and it really went smoothly. What I want to know is how does AWS/Azure look up kernel. I thought we have some FreeBSD experts who would like to let us know what exactly goes into the image to make it a smooth transition from one version of OS to another. Though I could have googled and got the information. I would like to concrete implementation details I can rely on. From pete at nomadlogic.org Sat May 11 12:43:54 2019 From: pete at nomadlogic.org (Pete Wright) Date: Sat, 11 May 2019 09:43:54 -0700 Subject: [talk] AWS vs Azure In-Reply-To: References: Message-ID: <5c63e41e-b817-2e6f-a415-a8c6136b9851@nomadlogic.org> On 5/11/19 4:42 AM, Sujit K M wrote: > On a recent discussion I was rightly pointed to a lot of FreeBSD work > that went into FreeBSD for AWS. I might not be that correct saying AWS > is there for FreeBSD 11. There are AMI's available from Colin Percival going back to 8.x iirc - I believe the release-engineering team (and thus official AMI images) started being available with 11.x or 12.x. > > Now on parallel, I checked a kernel built install for Linux Kernel 5.0 and > it really went smoothly. > > What I want to know is how does AWS/Azure look up kernel. I thought > we have some FreeBSD experts who would like to let us know what exactly > goes into the image to make it a smooth transition from one version of OS > to another. Though I could have googled and got the information. I would like > to concrete implementation details I can rely on. I am not really sure what you are asking.? i've never had issues running make buildworld to do source updates of freebsd on AWS, nor using freebsd-update. as far as generating AMI images I believe the current release process is based on this initial work done by Colin: http://www.daemonology.net/blog/2015-11-21-FreeBSD-AMI-builder-AMI.html this might be out of date but hopefully is a good starting point for how the whole process looks. -pete -- Pete Wright pete at nomadlogic.org @nomadlogicLA From kmsujit at gmail.com Sat May 11 23:45:08 2019 From: kmsujit at gmail.com (Sujit K M) Date: Sun, 12 May 2019 09:15:08 +0530 Subject: [talk] AWS vs Azure In-Reply-To: <5c63e41e-b817-2e6f-a415-a8c6136b9851@nomadlogic.org> References: <5c63e41e-b817-2e6f-a415-a8c6136b9851@nomadlogic.org> Message-ID: > > Now on parallel, I checked a kernel built install for Linux Kernel 5.0 and > > it really went smoothly. > > > > What I want to know is how does AWS/Azure look up kernel. I thought > > we have some FreeBSD experts who would like to let us know what exactly > > goes into the image to make it a smooth transition from one version of OS > > to another. Though I could have googled and got the information. I would like > > to concrete implementation details I can rely on. > I am not really sure what you are asking. i've never had issues running > make buildworld to do source updates of freebsd on AWS, nor using > freebsd-update. I am more interested in knowing the how the for example operating environment is detected for example using AWS. https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AMIs.html I would like to know more than what is given in the above link as certainly I should be able to configure an instance. https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/UserProvidedKernels.html The above link gives how a custom linux kernel can be used. But my question more to anything other is what are dependencies for example I move from Linux Kernel version 3.0 to Linux Kernel version 5.0 what is the guarantee that the Image generated is correct. Does AWS imply that unless Amazon Certify the 5.0 Kernel It cannot be done. From pete at nomadlogic.org Sun May 12 13:28:37 2019 From: pete at nomadlogic.org (Pete Wright) Date: Sun, 12 May 2019 10:28:37 -0700 Subject: [talk] AWS vs Azure In-Reply-To: Message-ID: <716f7816-b276-49f5-a1e6-906bfc4350c0@email.android.com> An HTML attachment was scrubbed... URL: From okan at demirmen.com Mon May 13 08:53:15 2019 From: okan at demirmen.com (Okan Demirmen) Date: Mon, 13 May 2019 08:53:15 -0400 Subject: [talk] kill -0 In-Reply-To: References: <20190510180936.GU13229@netmeister.org> Message-ID: <20190513125315.GA29018@carbon.khaoz.org> On Fri 2019.05.10 at 14:13 -0400, Raul Cuza wrote: > On Fri, May 10, 2019 at 2:09 PM Jan Schaumann wrote: > > > > Raul Cuza wrote: > > > > > I just had someone submit a shell script using `kill -0 ${PID}` to > > > test if a process is running. It exits non-zero if ${PID} isn't a > > > running process and zero otherwise. But... > > > > > > http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. > > > > kill(2) describes this behavior: > > > > "sig may be one of the signals specified in > > sigaction(2) or it may be 0, in which case error > > checking is performed but no signal is actually sent. > > This can be used to check the validity of pid." > > > > -Jan > > I will update my brain and not be "that PR reviewer". Thank you! It's documented in both kill(2) and kill(1) on OpenBSD at least. From christos at zoulas.com Mon May 13 08:55:31 2019 From: christos at zoulas.com (Christos Zoulas) Date: Mon, 13 May 2019 08:55:31 -0400 Subject: [talk] kill -0 In-Reply-To: <20190513125315.GA29018@carbon.khaoz.org> References: <20190510180936.GU13229@netmeister.org> <20190513125315.GA29018@carbon.khaoz.org> Message-ID: <7B557795-E036-4C99-B7C5-48F3B0ADC7DF@zoulas.com> https://pubs.opengroup.org/onlinepubs/009695399/functions/kill.html christos > On May 13, 2019, at 8:53 AM, Okan Demirmen wrote: > > On Fri 2019.05.10 at 14:13 -0400, Raul Cuza wrote: >> On Fri, May 10, 2019 at 2:09 PM Jan Schaumann wrote: >>> >>> Raul Cuza wrote: >>> >>>> I just had someone submit a shell script using `kill -0 ${PID}` to >>>> test if a process is running. It exits non-zero if ${PID} isn't a >>>> running process and zero otherwise. But... >>>> >>>> http://nixdoc.net/man-pages/FreeBSD/kill.1.html says nothing about it. >>> >>> kill(2) describes this behavior: >>> >>> "sig may be one of the signals specified in >>> sigaction(2) or it may be 0, in which case error >>> checking is performed but no signal is actually sent. >>> This can be used to check the validity of pid." >>> >>> -Jan >> >> I will update my brain and not be "that PR reviewer". Thank you! > > It's documented in both kill(2) and kill(1) on OpenBSD at least. > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk -------------- next part -------------- An HTML attachment was scrubbed... URL: From raulcuza at gmail.com Mon May 13 09:18:22 2019 From: raulcuza at gmail.com (Raul Cuza) Date: Mon, 13 May 2019 09:18:22 -0400 Subject: [talk] AWS vs Azure In-Reply-To: <5c63e41e-b817-2e6f-a415-a8c6136b9851@nomadlogic.org> References: <5c63e41e-b817-2e6f-a415-a8c6136b9851@nomadlogic.org> Message-ID: On Sat, May 11, 2019 at 12:44 PM Pete Wright wrote: > > > > On 5/11/19 4:42 AM, Sujit K M wrote: > > On a recent discussion I was rightly pointed to a lot of FreeBSD work > > that went into FreeBSD for AWS. I might not be that correct saying AWS > > is there for FreeBSD 11. > > There are AMI's available from Colin Percival going back to 8.x iirc - I > believe the release-engineering team (and thus official AMI images) > started being available with 11.x or 12.x. Are any of the release-engineering team in or going to be in the NYC area? This would make a great talk at our meetings. RC From raulcuza at gmail.com Mon May 13 09:21:49 2019 From: raulcuza at gmail.com (Raul Cuza) Date: Mon, 13 May 2019 09:21:49 -0400 Subject: [talk] kill -0 In-Reply-To: <7B557795-E036-4C99-B7C5-48F3B0ADC7DF@zoulas.com> References: <20190510180936.GU13229@netmeister.org> <20190513125315.GA29018@carbon.khaoz.org> <7B557795-E036-4C99-B7C5-48F3B0ADC7DF@zoulas.com> Message-ID: On Mon, May 13, 2019 at 8:55 AM Christos Zoulas wrote: > > https://pubs.opengroup.org/onlinepubs/009695399/functions/kill.html > That's the best man of the lot. FreeBSD's needs a pull request/patch. RC From george at ceetonetechnology.com Fri May 17 10:11:00 2019 From: george at ceetonetechnology.com (George Rosamond) Date: Fri, 17 May 2019 14:11:00 +0000 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD Message-ID: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> Greetings. I've generally run OpenBSD on my swarm of APU2s, but I'm going to run FreeBSD 12.0-RELEASE on one. I've tended to shy away from ZFS since I'm not in large data mode, and with a single disk in this case, there's really no huge benefit. But I decided to go with ZFS on this msata EVO Sony with geli encryption. Does it make sense? Does anyone else have experiences with ZFS v UFS on APU2s? g From ike at blackskyresearch.net Fri May 17 10:23:20 2019 From: ike at blackskyresearch.net (Isaac (.ike) Levy) Date: Fri, 17 May 2019 10:23:20 -0400 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> Message-ID: <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> Word, On Fri, May 17, 2019, at 10:11 AM, George Rosamond wrote: > Greetings. > > I've generally run OpenBSD on my swarm of APU2s, but I'm going to run > FreeBSD 12.0-RELEASE on one. > > I've tended to shy away from ZFS since I'm not in large data mode, and > with a single disk in this case, there's really no huge benefit. But there is, depending... :) Just to be pedantic, top of head reasons why ZFS is not just great for "big disks", but is advantageous on this sort of board/media: - No fsck, (design is better than journaling even). - Block level checksumming (find what would otherwise be silent corruption) - Even if not multi-disk zpool, (which can auto-replicate bad blocks), one can preempt major problems by knowing their data is good. - Snapshots and User Features (way better than hard partitioning in practice, because all can be changed in live system runtime): - New logical partitions any time you want them - Quotas - Backup/Restore via snapshots/incrementals (awesome for embedded app deploys) Etc... etc... > > But I decided to go with ZFS on this msata EVO Sony with geli encryption. This should be excellent, I'd be curious to hear your experience here- (since I know you to have years of embedded BSD practical experience, but new to using ZFS on it...) > > Does it make sense? Does anyone else have experiences with ZFS v UFS on APU2s? It's a real delightful combo, IMHO, and great timing- because ZFS memory/implementation is really reliable, (used to *require* so much more RAM and tweaking), nowadays it just works- properly boring. Again, would love to hear your good/bad/ugly as you go. Best, .ike > > g > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk > From kmsujit at gmail.com Fri May 17 10:51:21 2019 From: kmsujit at gmail.com (Sujit K M) Date: Fri, 17 May 2019 20:21:21 +0530 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> Message-ID: On Fri, May 17, 2019, 7:42 PM George Rosamond wrote: > Greetings. > > I've generally run OpenBSD on my swarm of APU2s, but I'm going to run > FreeBSD 12.0-RELEASE on one. > > I've tended to shy away from ZFS since I'm not in large data mode, and > with a single disk in this case, there's really no huge benefit. > > But I decided to go with ZFS on this msata EVO Sony with geli encryption. > > Does it make sense? Does anyone else have experiences with ZFS v UFS on > APU2s? > I am yet to rate a fs on Linux. But I do rate zfs highly. Simply because of tools like zpool. I find disk management in zfs is much better. > > g > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk > -------------- next part -------------- An HTML attachment was scrubbed... URL: From spork at bway.net Fri May 17 13:36:14 2019 From: spork at bway.net (Charles Sprickman) Date: Fri, 17 May 2019 13:36:14 -0400 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> Message-ID: <195C32F7-6A69-4483-B083-690D41538064@bway.net> > On May 17, 2019, at 10:23 AM, Isaac (.ike) Levy wrote: > > Word, > > On Fri, May 17, 2019, at 10:11 AM, George Rosamond wrote: >> Greetings. >> >> I've generally run OpenBSD on my swarm of APU2s, but I'm going to run >> FreeBSD 12.0-RELEASE on one. >> >> I've tended to shy away from ZFS since I'm not in large data mode, and >> with a single disk in this case, there's really no huge benefit. > > But there is, depending... :) > > Just to be pedantic, top of head reasons why ZFS is not just great for "big disks", but is advantageous on this sort of board/media: > > - No fsck, (design is better than journaling even). > - Block level checksumming (find what would otherwise be silent corruption) > - Even if not multi-disk zpool, (which can auto-replicate bad blocks), one can preempt major problems by knowing their data is good. - Maybe not useful, and perhaps even bad on an SSD (DIY write amplification), setting the ?copies? parameter lets you take advantage of all the free space you probably have on something like a firewall where you?re using like 5% of the disk space? I know on a spinny drive copies can save your life, not sure how that intersects with the more common SSD failure modes tho... > - Snapshots and User Features (way better than hard partitioning in practice, because all can be changed in live system runtime): > - New logical partitions any time you want them > - Quotas > - Backup/Restore via snapshots/incrementals (awesome for embedded app deploys) > > Etc... etc... > >> >> But I decided to go with ZFS on this msata EVO Sony with geli encryption. > > This should be excellent, I'd be curious to hear your experience here- (since I know you to have years of embedded BSD practical experience, but new to using ZFS on it...) > >> >> Does it make sense? Does anyone else have experiences with ZFS v UFS on APU2s? > > It's a real delightful combo, IMHO, and great timing- because ZFS memory/implementation is really reliable, (used to *require* so much more RAM and tweaking), nowadays it just works- properly boring. Tell me more? I avoid ZFS on anything without 8GB or more of RAM, so I?ve never used it on 512MB/1GB/2GB VPS hosts I have out there. Should I be using it? Thanks, C > > Again, would love to hear your good/bad/ugly as you go. > > Best, > .ike > > >> >> g >> >> _______________________________________________ >> talk mailing list >> talk at lists.nycbug.org >> http://lists.nycbug.org:8080/mailman/listinfo/talk >> > > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk From ike at blackskyresearch.net Fri May 17 15:09:23 2019 From: ike at blackskyresearch.net (Isaac (.ike) Levy) Date: Fri, 17 May 2019 15:09:23 -0400 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <195C32F7-6A69-4483-B083-690D41538064@bway.net> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> <195C32F7-6A69-4483-B083-690D41538064@bway.net> Message-ID: <3c7ca11c-e8cb-40fa-9fbf-0d3a03fd8027@www.fastmail.com> On Fri, May 17, 2019, at 1:37 PM, Charles Sprickman wrote: > > > On May 17, 2019, at 10:23 AM, Isaac (.ike) Levy wrote: > > > > Word, > > > > On Fri, May 17, 2019, at 10:11 AM, George Rosamond wrote: > >> Greetings. > >> > >> I've generally run OpenBSD on my swarm of APU2s, but I'm going to run > >> FreeBSD 12.0-RELEASE on one. > >> > >> I've tended to shy away from ZFS since I'm not in large data mode, and > >> with a single disk in this case, there's really no huge benefit. > > > > But there is, depending... :) > > > > Just to be pedantic, top of head reasons why ZFS is not just great for "big disks", but is advantageous on this sort of board/media: > > > > - No fsck, (design is better than journaling even). > > - Block level checksumming (find what would otherwise be silent corruption) > > - Even if not multi-disk zpool, (which can auto-replicate bad blocks), one can preempt major problems by knowing their data is good. > > - Maybe not useful, and perhaps even bad on an SSD (DIY write > amplification), setting the ?copies? parameter lets you take advantage > of all the free space you probably have on something like a firewall > where you?re using like 5% of the disk space? I know on a spinny drive > copies can save your life, not sure how that intersects with the more > common SSD failure modes tho... SSD's die just like any disk! Spork, thanks for blowing my mind again today it never even occured to me that block "copies" could be used to increase chance of recovery as bitrot sets in on single disks... (e.g. when the ZFS scrubs start reporting lots of dead blocks [and copying duplicates], it's time to buy a new disk...) Years of using ZFS and never even thought of that awesome little gem. Best, .ike > > > - Snapshots and User Features (way better than hard partitioning in practice, because all can be changed in live system runtime): > > - New logical partitions any time you want them > > - Quotas > > - Backup/Restore via snapshots/incrementals (awesome for embedded app deploys) > > > > Etc... etc... > > > >> > >> But I decided to go with ZFS on this msata EVO Sony with geli encryption. > > > > This should be excellent, I'd be curious to hear your experience here- (since I know you to have years of embedded BSD practical experience, but new to using ZFS on it...) > > > >> > >> Does it make sense? Does anyone else have experiences with ZFS v UFS on APU2s? > > > > It's a real delightful combo, IMHO, and great timing- because ZFS memory/implementation is really reliable, (used to *require* so much more RAM and tweaking), nowadays it just works- properly boring. > > Tell me more? I avoid ZFS on anything without 8GB or more of RAM, so > I?ve never used it on 512MB/1GB/2GB VPS hosts I have out there. Should > I be using it? > > Thanks, > > C > > > > > Again, would love to hear your good/bad/ugly as you go. > > > > Best, > > .ike > > > > > >> > >> g > >> > >> _______________________________________________ > >> talk mailing list > >> talk at lists.nycbug.org > >> http://lists.nycbug.org:8080/mailman/listinfo/talk > >> > > > > _______________________________________________ > > talk mailing list > > talk at lists.nycbug.org > > http://lists.nycbug.org:8080/mailman/listinfo/talk > > From nycbug-talk at reynolds.users.panix.com Fri May 17 16:43:51 2019 From: nycbug-talk at reynolds.users.panix.com (Brian Reynolds) Date: Fri, 17 May 2019 16:43:51 -0400 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <195C32F7-6A69-4483-B083-690D41538064@bway.net> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> <195C32F7-6A69-4483-B083-690D41538064@bway.net> Message-ID: <20190517204351.GA6358@panix.com> Charles Sprickman wrote: > > On May 17, 2019, at 10:23 AM, Isaac (.ike) Levy > > wrote: > > On Fri, May 17, 2019, at 10:11 AM, George Rosamond wrote: > >> > >> Does it make sense? Does anyone else have experiences with ZFS > >> v UFS on APU2s? So long as you're running 64 bit, I think it makes perfect sense. Of course it would be even better if you could squeeze in two SSDs and mirror them. > > It's a real delightful combo, IMHO, and great timing- because ZFS > > memory/implementation is really reliable, (used to *require* so > > much more RAM and tweaking), nowadays it just works- properly > > boring. > > Tell me more??? I avoid ZFS on anything without 8GB or more of RAM, > so I???ve never used it on 512MB/1GB/2GB VPS hosts I have out > there. Should I be using it? First off, my background is as a Solaris sysadmin who now concentrates on FreeBSD. I have used ZFS on Sun, and other hardware, under Solaris 10, and other OSes, for a number of years. That said, the memory "requirements" seen in the post-Sun implementations of ZFS appear to me to be the result of interpreting the worst case from Sun's already conservative documentation. The highest of the memory usage requirements (~1GB RAM per TB of storage) assumed you were running de-dupe, something, as I recall, Sun eventually recommended people not use. The other big user of memory is caching. If the ZFS sub-system sees that there is available (i.e., unused) memory it will request all of it. If it sees that other parts of the system are requesting memory it will release some of the cache to ease the system's burdens. So far as tweaking ZFS goes, the most comprehensive document for fiddling with ZFS internals was known as the "ZFS Evil Tuning Guide". It's name was chosen to discourage people from fiddling. The tl;dr is "Don't do it." There are better ways to improve system performance. In addition to bigger machines, I routinely ran Solaris 10 and ZFS on Sun Blade 100s with at most 4GB RAM and a couple of first generation IDE hard drives. I also ran ZFS with FreeBSD 9 & 10 on Dell desktops with more memory, but room for only one drive. With that little memory (< 4GB once the OS takes its chunk) some features will be automatically turned off, but I never missed them. Even without the data redundancy, ZFS was still a win due to snapshots, which in turn allow boot environments (under Solaris and FreeBSD). Boot environments saved me from bad OS and package updates on several occasions. -- Brian Reynolds -- reynolds at panix.com "Long ago, it was the way of my people to travel to the Moon. My people travelled to the Moon in a ship called 'Apollo.' I tell the story of Apollo in the hope that some day, my people will again travel to the Moon." P. Alway From ike at blackskyresearch.net Tue May 21 10:02:50 2019 From: ike at blackskyresearch.net (Isaac (.ike) Levy) Date: Tue, 21 May 2019 10:02:50 -0400 Subject: [talk] RTC on Beaglebone? Message-ID: <60c9aa21-fe73-4eaf-8e86-fd2210c3eeba@www.fastmail.com> Hi All, FreeBSD or OpenBSD on ARM/embedded question: I?ve got a timer project, and was wondering if anyone had experience with RTC and BeagleBone boards? I don?t need a battery-backed RTC, I just need to keep good wall-clock time when the machine is on. (I think the beaglebone SOC has an RTC in there somewhere?) I?ve found *one* post from someone tinkering with a BeagleBone and RTC on FreeBSD, https://forums.freebsd.org/threads/rtc-on-beaglebone.64802/ ?- Last project I did like this, I used a PcEngines APU, because the on-board RTC/battery works without any fuss. For this project, I need to power this from a battery so a PcEngines board is way too much of a power hog... Thoughts? -- Anyone have experience with other hardware/boards? Thee project I?m doing needs: - needs to run FreeBSD or OpenBSD - needs to keep decent wall-clock time - needs to have wifi (can do cheap/slow/weak USB plug type) - needs to be powerful enough to run a *very basic* webserver and some light CGI (older beaglebone or rPi is totally punchy enough here) - needs to have power specs similar to a beaglebone or RaspPi - needs to have at least a hand full of GPIO pins - Bonus: one or more i2c headers - Bonus: hardware GPIO interrupts would be nice I think a beaglebone is my best bet, but there are so many cool little boards out there (?)... Best, .ike -------------- next part -------------- An HTML attachment was scrubbed... URL: From jpb at jimby.name Tue May 21 10:27:11 2019 From: jpb at jimby.name (jpb) Date: Tue, 21 May 2019 14:27:11 +0000 Subject: [talk] RTC on Beaglebone? In-Reply-To: <60c9aa21-fe73-4eaf-8e86-fd2210c3eeba@www.fastmail.com> References: <60c9aa21-fe73-4eaf-8e86-fd2210c3eeba@www.fastmail.com> Message-ID: <20190521142711.5uyhbl5yliouh745@jimby.name> On Tue, May 21, 2019 at 10:02:50AM -0400, Isaac (.ike) Levy wrote: > Hi All, > > FreeBSD or OpenBSD on ARM/embedded question: > > I?ve got a timer project, and was wondering if anyone had experience > with RTC and BeagleBone boards? > > I don?t need a battery-backed RTC, I just need to keep good wall-clock > time when the machine is on. (I think the beaglebone SOC has an RTC in > there somewhere?) > > I?ve found *one* post from someone tinkering with a BeagleBone and RTC > on FreeBSD, > > https://forums.freebsd.org/threads/rtc-on-beaglebone.64802/ > > ?- > > Last project I did like this, I used a PcEngines APU, because the > on-board RTC/battery works without any fuss. For this project, I need > to power this from a battery so a PcEngines board is way too much of a > power hog... > > Thoughts? > > -- > > Anyone have experience with other hardware/boards? > > Thee project I?m doing needs: > > - needs to run FreeBSD or OpenBSD > > - needs to keep decent wall-clock time > > - needs to have wifi (can do cheap/slow/weak USB plug type) > > - needs to be powerful enough to run a *very basic* webserver and some > light CGI (older beaglebone or rPi is totally punchy enough here) > > - needs to have power specs similar to a beaglebone or RaspPi > > - needs to have at least a hand full of GPIO pins > > - Bonus: one or more i2c headers > > - Bonus: hardware GPIO interrupts would be nice > > I think a beaglebone is my best bet, but there are so many cool little > boards out there (?)... > > Best, > > .ike Hey Ike, I did a BB Black project using FreeBSD 11.x a couple years ago using GPIO, and used NTP for keeping time. Worked great. The BB board was coupled to an external system that controlled water flow that had to be turned on and off at specific times. GPIO pins controlled the external board that controlled water flow. Used the onboard ethernet controller, not wifi, but I could have used that if needed (I have a bunch of mini usb wireless adapters). Hit me up offlist if you want more info. Cheers, Jim B. From mark.saad at ymail.com Tue May 21 20:34:29 2019 From: mark.saad at ymail.com (Mark Saad) Date: Tue, 21 May 2019 20:34:29 -0400 Subject: [talk] Iwm in openbsd 6.5-current Message-ID: <41A67A46-80DC-4511-8A3A-1522686D48C7@ymail.com> All In openbsd current , I am fighting to figure out how to make my iwn wireless card properly associate with my iPhones personal hotspot . I am able to get the wpa2 authentication working however when I start dhclient it immediately downs the nic and sleeps . Now to be clear with 6.5 release the iwn had issues and would not work consistently; current works better and I can associate to cisco ap?s and run dhclient with out issues . This is current pulled down today ; for amd64 . Anyone have any ideas ? --- Mark Saad | mark.saad at ymail.com From nonesuch at longcount.org Wed May 22 09:11:43 2019 From: nonesuch at longcount.org (Mark Saad) Date: Wed, 22 May 2019 09:11:43 -0400 Subject: [talk] New email for talk and same email of questions Message-ID: Hi everybody I am starting to retire my yahoo email . It?s become more trouble then it?s worth. I sent this last night and it?s still just a repost . All In openbsd current , I am fighting to figure out how to make my iwm wireless card properly associate with my iPhones personal hotspot . I am able to get the wpa2 authentication working however when I start dhclient it immediately downs the nic and sleeps . Now to be clear with 6.5 release the iwm had issues and would not work consistently; current works better and I can associate to cisco ap?s and run dhclient with out issues . This is current pulled down today ; for amd64 generic #36. Anyone have any ideas ? --- Mark Saad | nonesuch at longcount.org From mcevoy.pat at gmail.com Wed May 22 17:10:07 2019 From: mcevoy.pat at gmail.com (Pat McEvoy) Date: Wed, 22 May 2019 17:10:07 -0400 Subject: [talk] Next NYC*BUG: Message-ID: <6C4B9466-4AF3-46E9-BCE8-A21D979634FD@gmail.com> Hey Folks, Does anyone have any topics they would like to talk about for the June NYC*BUG? We have a few things slated for the coming months, but nothing firm for June. Please hit this list or email me directly if you have ideas. Also, I grabbed some swag for the group from BSDCan and will be sure to bring to the meeting. Be well, Patrick From nycbug-talk at reynolds.users.panix.com Wed May 22 21:52:12 2019 From: nycbug-talk at reynolds.users.panix.com (Brian Reynolds) Date: Wed, 22 May 2019 21:52:12 -0400 Subject: [talk] ZFS v UFS on APU2 msata SSD with FreeBSD In-Reply-To: <20190517204351.GA6358@panix.com> References: <3f63064f-a8f7-d75d-9059-7b83329917c1@ceetonetechnology.com> <081bc86b-462c-4b81-a6bb-a433dd39ac75@www.fastmail.com> <195C32F7-6A69-4483-B083-690D41538064@bway.net> <20190517204351.GA6358@panix.com> Message-ID: <20190523015212.GC11206@panix.com> I wrote: > > In addition to bigger machines, I routinely ran Solaris 10 and ZFS on > Sun Blade 100s with at most 4GB RAM and a couple of first generation > IDE hard drives. I also ran ZFS with FreeBSD 9 & 10 on Dell desktops > with more memory, but room for only one drive. Oops. A bit of a mistake there. The Sun Blade 100 has a maximum of 2GB of RAM, and was fine running ZFS. The Dell desktops had 4GB, and were also OK running ZFS. -- Brian Reynolds -- reynolds at panix.com "Long ago, it was the way of my people to travel to the Moon. My people travelled to the Moon in a ship called 'Apollo.' I tell the story of Apollo in the hope that some day, my people will again travel to the Moon." P. Alway From kmsujit at gmail.com Thu May 23 08:19:01 2019 From: kmsujit at gmail.com (Sujit K M) Date: Thu, 23 May 2019 17:49:01 +0530 Subject: [talk] New email for talk and same email of questions In-Reply-To: References: Message-ID: On Wed, May 22, 2019, 6:42 PM Mark Saad wrote: > Hi everybody > I am starting to retire my yahoo email . It?s become more trouble then > it?s worth. > > I sent this last night and it?s still just a repost . > > All > In openbsd current , I am fighting to figure out how to make my iwm > wireless card properly associate with my iPhones personal hotspot . I am > able to get the wpa2 authentication working however when I start dhclient > it immediately downs the nic and sleeps . Now to be clear with 6.5 release > the iwm had issues and would not work consistently; current works better > and I can associate to cisco ap?s and run dhclient with out issues . This > is current pulled down today ; for amd64 generic #36. Anyone have any > ideas ? > Looks strange. Could you try Android. Might be some default settings. I see in my Android most of the network are accessible. But in iPhone I have issues connecting with dlink repeaters. Also could you try with VoLTE disabled. --- > Mark Saad | nonesuch at longcount.org > _______________________________________________ > talk mailing list > talk at lists.nycbug.org > http://lists.nycbug.org:8080/mailman/listinfo/talk > -------------- next part -------------- An HTML attachment was scrubbed... URL: