GNU bug report logs - #51793
FAIL: tests/misc/env-signal-handler

Previous Next

Full log

Message #8 received at 51793 <at> debbugs.gnu.org (full text, mbox):

From: Pádraig Brady <P <at> draigBrady.com>
To: Andreas Schwab <schwab <at> linux-m68k.org>, 51793 <at> debbugs.gnu.org
Subject: Re: bug#51793: FAIL: tests/misc/env-signal-handler
Date: Fri, 12 Nov 2021 22:33:17 +0000

On 12/11/2021 19:25, Andreas Schwab wrote:
> --- exp-err6   2021-11-11 22:58:04.360716802 +0000
> +++ err6       2021-11-11 22:58:04.752716821 +0000
> @@ -1,2 +1 @@
>   timeout: sending signal INT to command 'env'
> -timeout: sending signal KILL to command 'env'

The above is for the previous command not the following.

> ./tests/misc/env-signal-handler.sh: line 127: 26396 Killed                  env timeout --verbose --kill-after=.1 --signal=INT .1 env --ignore-signal sleep 10 > /dev/null 2> err7t
> FAIL tests/misc/env-signal-handler.sh (exit status: 1)

Could you paste a bit more context from the previous command.
Specifically I'm wondering if the "Killed" message was displayed
(i.e. if the SIGKILL was actually sent).

I presume this is a race that you're seeing rarely.
I've never seen this, and I can't see the race.

What might be happening is the "sending signal KILL..."
message sent to stderr, is in a buffer somewhere,
which is dropped upon processing of SIGKILL.

What file system, kernel, shell are you using.

cheers,
Pádraig

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.