Re: Figure out what killed an app (rhbz#2253099)

On Mon, 2024-02-05 at 16:20 +0200, Yanko Kaneti wrote: > On Mon, 2024-02-05 at 14:31 +0100, Florian Weimer wrote: > > > > > > Running with rtkit-daemon stoped and masked prevents the main evo > > > process being prioritized and consequently killed with higher load. > > > > > > The fact that the k

Re: Figure out what killed an app (rhbz#2253099)

On Mon, 2024-02-05 at 14:31 +0100, Florian Weimer wrote: > * Yanko Kaneti: > > > On Sat, 2024-02-03 at 13:29 +0100, Ondrej Mosnáček wrote: > > > On Fri, 2 Feb 2024 at 17:52, Yanko Kaneti wrote: > > > > > > > > On Thu, 2024-02-01 at 09:44 +0100, Ondrej Mosnáček wrote: > > > > > On Thu, 1 Feb 2024

Re: Figure out what killed an app (rhbz#2253099)

* Yanko Kaneti: > On Sat, 2024-02-03 at 13:29 +0100, Ondrej Mosnáček wrote: >> On Fri, 2 Feb 2024 at 17:52, Yanko Kaneti wrote: >> > >> > On Thu, 2024-02-01 at 09:44 +0100, Ondrej Mosnáček wrote: >> > > On Thu, 1 Feb 2024 at 09:13, Milan Crha wrote: >> > > > The kernel tracing log for sig==9 sh

Re: Figure out what killed an app (rhbz#2253099)

On Sun, 2024-02-04 at 14:03 +0200, Yanko Kaneti wrote: > The kernel killing the main evolution process which was set as having > rt priority by some webkit coincidence. Hi, there it is then. I see those: > Jan 31 10:49:22 localhost.localdomain rtkit-daemon[826]: > Successfully made th

Re: Figure out what killed an app (rhbz#2253099)

On Sat, 2024-02-03 at 13:29 +0100, Ondrej Mosnáček wrote: > On Fri, 2 Feb 2024 at 17:52, Yanko Kaneti wrote: > > > > On Thu, 2024-02-01 at 09:44 +0100, Ondrej Mosnáček wrote: > > > On Thu, 1 Feb 2024 at 09:13, Milan Crha wrote: > > > > The kernel tracing log for sig==9 shows: > > > > > > > > gn

Re: Figure out what killed an app (rhbz#2253099)

On Fri, 2 Feb 2024 at 17:52, Yanko Kaneti wrote: > > On Thu, 2024-02-01 at 09:44 +0100, Ondrej Mosnáček wrote: > > On Thu, 1 Feb 2024 at 09:13, Milan Crha wrote: > > > The kernel tracing log for sig==9 shows: > > > > > > gnome-terminal--2924[002] dN.2. 2520.462889: signal_generate: > > > sig

Re: Figure out what killed an app (rhbz#2253099)

On Thu, 2024-02-01 at 09:44 +0100, Ondrej Mosnáček wrote: > On Thu, 1 Feb 2024 at 09:13, Milan Crha wrote: > > The kernel tracing log for sig==9 shows: > > > > gnome-terminal--2924[002] dN.2. 2520.462889: signal_generate: > > sig=9 errno=0 code=128 comm=alloc-too-much pid=3502 grp=1 res=0 >

Re: Figure out what killed an app (rhbz#2253099)

On Thu, 1 Feb 2024 at 09:13, Milan Crha wrote: > The kernel tracing log for sig==9 shows: > > gnome-terminal--2924[002] dN.2. 2520.462889: signal_generate: > sig=9 errno=0 code=128 comm=alloc-too-much pid=3502 grp=1 res=0 > > There is no such thing (apart of the tracing log) when Evolution is

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 2024-01-31 at 13:40 -0500, Paul Grosu wrote: > 2) Or you can completely disable it: Hi, I do not want to disable the oom service. Remember, it does that on user machines, not only on mine. Telling people: "you want to use app a, b, c, then disable oom" as a new Fedora 40 feature is

Re: Figure out what killed an app (rhbz#2253099)

On Wed, Jan 31, 2024, at 11:40 AM, Paul Grosu wrote: > > > https://github.com/facebookincubator/oomd/ > > 2) Or you can completely disable it: > > https://www.cjjackson.dev/posts/what-is-systemd-oomd-how-to-disable-it/ We need bug reports to get it fixed rather than disabling it, if it's cau

Re: Figure out what killed an app (rhbz#2253099)

I think your options are two: 1) Write a plugin for the oom service to capture the state (and log it) before it kills the process. Here's the source code to learn more about that: https://github.com/facebookincubator/oomd/ 2) Or you can completely disable it: https://www.cjjackson.dev/posts/wh

Re: Figure out what killed an app (rhbz#2253099)

On Wed, Jan 31 2024 at 06:53:25 PM +01:00:00, Milan Crha wrote: Evo itself doesn't use any seccomp or such, these things can be used by the WebKitGTK. A quick grep revealed: https://github.com/WebKit/WebKit/blob/main/Source/WebKit/UIProcess/Launcher/glib/ProcessLauncherGLib.cpp#L258 but that

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 2024-01-31 at 16:24 +0100, Petr Pisar wrote: > A possible explanation is that the signals are procecessed > asychnously and evolution manages to dispatch the signal to bwrap > before kernel > termites evolution because of the first signal. Or I misinterpret the > log. Hi, there had

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 2024-01-31 at 08:31 -0600, Michael Catanzaro wrote: > Maybe it could also be sent by mutter if a program > is unresponsive? Hi, the app is perfectly responsive. I click on a widget and the app is killed immediately. There is no freeze of the app. > WebKitGTK doesn't use SIGKILL t

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 31 Jan 2024 at 12:00, Michael Catanzaro wrote: > On Wed, Jan 31 2024 at 04:42:08 PM +01:00:00, Clemens Lang > wrote: > > Throwing some ideas out there, is it possible that evolution runs > > with a seccomp filter or other BPF program configured to kill the > > process on violation, and t

Re: Figure out what killed an app (rhbz#2253099)

On Wed, Jan 31 2024 at 04:42:08 PM +01:00:00, Clemens Lang wrote: Throwing some ideas out there, is it possible that evolution runs with a seccomp filter or other BPF program configured to kill the process on violation, and that’s what’s happening here? I don't think so. flatpak does use secc

Re: Figure out what killed an app (rhbz#2253099)

Hey, > On 31. Jan 2024, at 16:24, Petr Pisar wrote: > > Key information is code=128. That code is probably si_code member described in > sigaction(2). The manual lists a lot of values as constants. 128 value is > SI_KERNEL according to /usr/include/asm-generic/siginfo.h. It is documented > as "

Re: Figure out what killed an app (rhbz#2253099)

V Wed, Jan 31, 2024 at 01:07:42PM +0100, Milan Crha napsal(a): > evolution-3211 [003] d..1. 355.904404: signal_generate: sig=9 errno=0 > code=128 comm=evolution pid=3211 grp=1 res=0 > evolution-3211 [003] d..2. 355.904450: signal_generate: sig=9 errno=0 > code=0 comm=bwrap pid=3257 grp=1

Re: Figure out what killed an app (rhbz#2253099)

SIGKILL is almost always sent by systemd-oomd (or the kernel OOM killer). That's the most likely explanation. Theoretically it could also be sent by systemd if a service didn't quit quickly enough following a SIGTERM. Maybe it could also be sent by mutter if a program is unresponsive? WebKitG

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 2024-01-31 at 13:07 +0100, Milan Crha wrote: > In such case, I might be able to catch this in gdb, right? Maybe with > a breakpoint in the `kill` function, and any other? Okay, I tried with the following (more variants, just in case): gdb evolution \ --ex "b kill" \ --e

Re: Figure out what killed an app (rhbz#2253099)

On Wed, 2024-01-31 at 12:19 +0100, Petr Pisar wrote: > This procedure works for me > . > The tracefs file system has a nice log. Hi, that works pretty well, thank you. Funny enough, if I read it correctly, it says evo

Re: Figure out what killed an app (rhbz#2253099)

On Wed, Jan 31, 2024 at 11:06:02AM +, Tom Hughes via devel wrote: > On 31/01/2024 10:08, Milan Crha wrote: > > > I tried to investigate a rawhide bug: > > https://bugzilla.redhat.com/show_bug.cgi?id=2253099 > > which is about Evolution being killed "by something". That's the thing, > > I do no

Re: Figure out what killed an app (rhbz#2253099)

V Wed, Jan 31, 2024 at 11:08:16AM +0100, Milan Crha napsal(a): > Hi, > I tried to investigate a rawhide bug: > https://bugzilla.redhat.com/show_bug.cgi?id=2253099 > which is about Evolution being killed "by something". That's the thing, > I do not know what killed it, thus even why it had bee

Re: Figure out what killed an app (rhbz#2253099)

On 31/01/2024 10:08, Milan Crha wrote: I tried to investigate a rawhide bug: https://bugzilla.redhat.com/show_bug.cgi?id=2253099 which is about Evolution being killed "by something". That's the thing, I do not know what killed it, thus even why it had been killed. It's even not killed after cert

Figure out what killed an app (rhbz#2253099)

Hi, I tried to investigate a rawhide bug: https://bugzilla.redhat.com/show_bug.cgi?id=2253099 which is about Evolution being killed "by something". That's the thing, I do not know what killed it, thus even why it had been killed. It's even not killed after certain steps, it's killed "random