[talk] dmesgd plugged

Charles Sprickman spork at bway.net
Sun Oct 7 14:44:37 EDT 2018



> On Oct 7, 2018, at 1:23 AM, Warner Losh <imp at bsdimp.com> wrote:
> 
> 
> 
> On Sat, Oct 6, 2018 at 11:02 PM Charles Sprickman <spork at bway.net <mailto:spork at bway.net>> wrote:
> 
>> On Oct 6, 2018, at 5:48 PM, Brian Callahan <bcallah at devio.us <mailto:bcallah at devio.us>> wrote:
>> 
>> 
>> On 10/06/18 17:44, Warner Losh wrote:
>>> 
>>> 
>>> On Sat, Oct 6, 2018 at 2:26 PM George Rosamond <george at ceetonetechnology.com <mailto:george at ceetonetechnology.com>> wrote:
>>> 
>>> 
>>> Warner Losh:
>>> > Greetings,
>>> > 
>>> > Just thought I'd mentioned that I plugged the dmesgd submission site on
>>> > FreeBSD's mailing list, so you might see an up-tick in submissions from
>>> 
>>> 
>>> Yes... to echo brian... more is better.
>>> 
>>> Glad to hear it.
>>> 
>>> BTW, there's a small niggle... FreeBSD just went to a reproducible build by default, so there's strings like
>>> FreeBSD 12.0-ALPHA6  r338698+59ff9aa72cd4(master) amd64
>>> now that  lack the #xxx in them. This is for someone that's using git-svn to build the release (which is why it has both the r# and the git hash). So their submission was rejected.
>>> 
>>> mine that was built with different options looked like:
>>> 
>>> FreeBSD 12.0-ALPHA2 #40+8d55b8a51f5(freze-newer)-dirty: Wed Aug 22 15:34:50 MDT 2018
>>> 
>>> the person that contacted me was clueful enough to puzzle it  out, but I thought I'd pass it along.
>>> 
>>> Also, I was going to bang together a quick python script that automatically submits the dmesg.boot file since the web page is kinda klunky if you are doing a bunch of servers... Any objections? Anybody beaten me to the punch?
>>> 
>> 
>> There's a Perl script by afresh1@ that does this for OpenBSD; I assume it can be quickly made to support any other *BSD.
>> https://gist.github.com/afresh1/99cdd481184147f0e8c0 <https://gist.github.com/afresh1/99cdd481184147f0e8c0>
> 
> Curl works as a one-liner:
> 
> curl -v -d "nickname=spork" -d "email=spork at bway.net <mailto:email=spork at bway.net>" -d 0 v2" -d "do=addd" --data-urlencode 'dmesg@/var/run/dmesg.boot' http://dmesgd.nycbug.org/index.cgi <http://dmesgd.nycbug.org/index.cgi>
> 
> I think the third -d should be -d "description=name of machine v2" no?
> 
> Warner
> 
> http://dmesgd.nycbug.org/index.cgi?do=view&id=3851 <http://dmesgd.nycbug.org/index.cgi?do=view&id=3851>
> 
> I just did http://dmesgd.nycbug.org/index.cgi?do=view&id=3852 <http://dmesgd.nycbug.org/index.cgi?do=view&id=3852> this way (though it's a legit entry too).

All - just curious if anyone else uses dmidecode (sysutils/dmidecode)?

More specifically, for anyone looking to script dmesg collection, dmidecode pulls some decent info about the make/model of the mainboard (or in some cases, the server as a whole) that would be useful info to have in the “description” field.

For example, part of the summary on this HP server in my garage:

Handle 0x0100, DMI type 1, 27 bytes
System Information
        Manufacturer: HP
        Product Name: ProLiant ML10 v2
        Version: Not Specified
        Serial Number: CN65310J2D
        UUID: 32353338-3636-4e43-3635-3331304a3244
        Wake-up Type: Power Switch
        SKU Number: 835266-P01
        Family: ProLiant

I’m wondering if there’s any way to grab this info without directing people to download yet another utility (I’m sure someone will grumble about ‘curl’). Of note, the first line of dmidecode’s output:

# dmidecode 3.2
Scanning /dev/mem for entry point.  <<<—  Hmmm?
SMBIOS 2.8 present.
78 structures occupying 2568 bytes.
Table at 0xF3FCB000.

Charles


> 
> Warner
>  
> Charles
> 
>> 
>> ~Brian
>> _______________________________________________
>> talk mailing list
>> talk at lists.nycbug.org <mailto:talk at lists.nycbug.org>
>> http://lists.nycbug.org/mailman/listinfo/talk <http://lists.nycbug.org/mailman/listinfo/talk>
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.nycbug.org/pipermail/talk/attachments/20181007/25d1873b/attachment.html>


More information about the talk mailing list