Re: s6-rc-init hangs during stage2

From: Vasil Yordanov <vasil.yordanov_at_gmail.com>
Date: Tue, 7 Jun 2016 14:23:13 +0300

Hello,

Unfortunately I'm not able to reproduce the problem.
I decided to upgrade the Docker from 1.9 to 1.10, because 1.9 does not
support mount of tmpfs.
After that upgrade the s6-rc-init does not hang. I even wrote a script to
continuously test it.
For now its works for me, but if in future this happens I will create a
more detailed strace to trace the forks.

Regards,

Vasil.

On Mon, Jun 6, 2016 at 8:28 PM, Laurent Bercot <ska-supervision_at_skarnet.org>
wrote:

> On 06/06/2016 17:02, Colin Booth wrote:
>
> When trying to bring s6rc-fdholder up directly I noticed it was dying
>> with a SIGFPE, here's the strace for that (./run modified to exec into
>> strace <opts> s6-fdholderd which is why the s6-rc-fdholder-filler
>> calls are missing):
>> http://paste.debian.net/718429/
>>
>
> Thanks.
> This is very probably unrelated. *Nothing* in s6-fdholderd uses floating
> point, and the SIGFPE... just can't happen. I'm looking suspiciously at
> regcomp(), which is a complex function that is called right when you're
> getting the SIGFPE, but I can't for the life of me figure out why even
> glibc would make floating point calls in regcomp().
> Could you try with another libc (or another version of glibc) and tell me
> if your s6-fdholderd works correctly?
>
> I'll be working on the s6-rc-init issue.
>
> --
> Laurent
>
Received on Tue Jun 07 2016 - 11:23:13 UTC

This archive was generated by hypermail 2.3.0 : Sun May 09 2021 - 19:44:19 UTC