IRC chat logs for #ltsp on irc.freenode.net (webchat)


Channel log from 27 March 2016   (all times are UTC)

00:34vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
03:01GodFodder has left IRC (GodFodder!~rcccole@2600:1007:b123:5971:ec4f:ece9:8881:af91, Ping timeout: 248 seconds)
03:21GodFodder has joined IRC (GodFodder!~rcccole@2600:1007:b123:5971:ec4f:ece9:8881:af91)
03:22GodFodder has joined IRC (GodFodder!~rcccole@2600:1007:b123:5971:ec4f:ece9:8881:af91)
03:22GodFodder has left IRC (GodFodder!~rcccole@2600:1007:b123:5971:ec4f:ece9:8881:af91, Max SendQ exceeded)
03:25GodFodder has joined IRC (GodFodder!~rcccole@2600:1007:b123:5971:ec4f:ece9:8881:af91)
04:13vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
07:15vagrantc has left IRC (vagrantc!~vagrant@unaffiliated/vagrantc, Quit: leaving)
07:26
<alkisg>
!learn todo-alkisg as put panic=60 in the kernel cmdline
07:26
<ltsp>
The operation succeeded.
07:26
<alkisg>
!forget todo-alkisg
07:26
<ltsp>
The operation succeeded.
07:26
<alkisg>
!learn alkisg-todo as put panic=60 in the kernel cmdline
07:26
<ltsp>
The operation succeeded.
07:26
<alkisg>
!alkisg-todo
07:26
<ltsp>
alkisg-todo: (#1) support xnbd-proxy for local caching: https://bitbucket.org/hirofuchi/xnbd/wiki/Home#!scenario-2-simple-proxy-server-distributed-copy-on-write, or (#2) Support UEFI, or (#3) support for per-user login commands in lts.conf, or (#4) put panic=60 in the kernel cmdline
10:26adrianorg has left IRC (adrianorg!~adrianorg@187.113.250.186, Ping timeout: 244 seconds)
10:28adrianorg has joined IRC (adrianorg!~adrianorg@177.134.63.189)
11:27robb_nl has joined IRC (robb_nl!~robb_nl@ip-83-134-22-171.dsl.scarlet.be)
12:08robb_nl has left IRC (robb_nl!~robb_nl@ip-83-134-22-171.dsl.scarlet.be, Ping timeout: 250 seconds)
12:22robb_nl has joined IRC (robb_nl!~robb_nl@ip-83-134-22-241.dsl.scarlet.be)
12:32ricotz has joined IRC (ricotz!~ricotz@ubuntu/member/ricotz)
12:54F-GT has joined IRC (F-GT!~phantom@ppp121-44-91-206.lns20.syd4.internode.on.net)
14:27robb_nl has left IRC (robb_nl!~robb_nl@ip-83-134-22-241.dsl.scarlet.be, Remote host closed the connection)
15:22vagrantc has joined IRC (vagrantc!~vagrant@unaffiliated/vagrantc)
16:35
<alkisg>
vagrantc: https://www.kernel.org/doc/Documentation/kernel-parameters.txt ==> panic=60 reboots after 1 minute
16:36
`echo 60 > /sys/module/kernel/parameters/panic` does that without having to pass it in the cmdline
16:36
So, how about a new lts.conf option, KERNEL_PANIC=<integer, default 60> ?
16:37
If it's 0 (the kernel default), then we exit that init-ltsp.d script and do nothing
16:37
<vagrantc>
sounds fine.
16:37
<alkisg>
The idea is that ltsp clients that get kernel panics could maybe reboot instead of just hanging there, and that 1 minute is enough for someone around to read what it says
16:37
<vagrantc>
and possibly useful for other projects i'm working on right now
16:38
<alkisg>
init-ltsp.d/50-kernel-panic is ok as a name?
16:38
<vagrantc>
alkisg: although a default value seems not so great.
16:38
<alkisg>
Ah, you think it should only be active if set by the user in lts.conf?
16:39
<vagrantc>
maybe, or set to a much higher value by default
16:39
<alkisg>
OK, how much higher? 10 minutes? More?
16:39
<vagrantc>
really depends on the use-case
16:40
e.g. a lab that uses the terminals almost constantly would want it different than a lab which uses the terminals sporadically
16:40
<alkisg>
We can't autodetect that
16:41
<vagrantc>
a location where someone may stay logged in all day but leave for hours at a time would be different
16:41
alkisg: exactly!
16:41
i'm not sure there's a reasonable default.
16:41
<alkisg>
I'm not sure it's better for either case not to have a default though
16:41
<vagrantc>
not having a default means it hangs indefinitely, no?
16:41
<alkisg>
I.e. what good would it do to have a hanged client (usually under xorg) without the panic message being displayed for e.g. 1 hour?
16:42
Usually we don't even see kernel panics because they happen after the clients boot
16:42
<vagrantc>
as opposed to one which silently reboots?
16:42
<alkisg>
One that silently reboots is again accessible, so the sysadmin can access it remotely
16:43
There's one advantage, vs 0 in the other case
16:43
I think the panic timeout is only useful in text mode
16:43
I.e. when the client boots
16:43
(or in SCREEN_07=shell, ssh etc)
16:44
I.e. it will allow the person there to read the message, get a photo of the error etc
16:44
Ah
16:44
OK, here's an idea, let's set it later on then
16:45
...although usually if it panics while the client is booting, it's reproducible
16:45
So I'm not sure how much it helps if you don't reboot, vs rebooting and being able to remotely change pxelinux.cfg/default or lts.conf and fix the issue that caused the panic without having physical access to the client
16:46
OK after all that, I vote for a default timeout of 1-10 minutes :)
16:46
Set at init-ltsp.d, nope, not later on
17:18GodFather has joined IRC (GodFather!~rcc@71-82-75-229.dhcp.bycy.mi.charter.com)
17:49GodFather has left IRC (GodFather!~rcc@71-82-75-229.dhcp.bycy.mi.charter.com, Ping timeout: 252 seconds)
17:53GodFather has joined IRC (GodFather!~rcc@71-82-75-229.dhcp.bycy.mi.charter.com)
18:04GodFather has left IRC (GodFather!~rcc@71-82-75-229.dhcp.bycy.mi.charter.com, Remote host closed the connection)
18:04solarnetone has joined IRC (solarnetone!~scott@rrcs-67-79-182-229.se.biz.rr.com)
18:05GodFather has joined IRC (GodFather!~rcc@71-82-75-229.dhcp.bycy.mi.charter.com)
20:06robb_nl has joined IRC (robb_nl!~robb_nl@ip-83-134-22-241.dsl.scarlet.be)
21:19robb_nl has left IRC (robb_nl!~robb_nl@ip-83-134-22-241.dsl.scarlet.be, Quit: I'm gone, bye bye)
21:23pppingme has left IRC (pppingme!~pppingme@unaffiliated/pppingme, Ping timeout: 260 seconds)
21:28pppingme has joined IRC (pppingme!~pppingme@unaffiliated/pppingme)
21:57ricotz has left IRC (ricotz!~ricotz@ubuntu/member/ricotz, Quit: Leaving)