#860648 gap-radiroot: FTBFS on i386: not enough memory during build on i386

#860648#5
Date:
2017-04-19 07:37:57 UTC
From:
To:
Hi,

During a rebuild of all packages in stretch (in a stretch chroot, not a
sid chroot), your package failed to build on i386.

Relevant part (hopefully):
http://aws-logs.debian.net/2017/04/18/gap-radiroot_2.7-2_testing-i386.log

A list of current common problems and possible solutions is available at
http://wiki.debian.org/qa.debian.org/FTBFS . You're welcome to contribute!

About the archive rebuild: The rebuild was done on EC2 VM instances from
Amazon Web Services, using a clean, minimal and up-to-date chroot. Every
failed build was retried once to eliminate random failures.

#860648#10
Date:
2017-04-27 09:11:22 UTC
From:
To:
severity 860648 normal
quit

Hello Lucas,
Sorry for the delay, for some reason my ISP blocked this email.
Do you know how much memory is there on this system ?
The test suite need 128M which does not seems unreasonable.

I did a test build using cowbuilder for stretch i386 and it worked fine.
It seems to me your system was experiencing a memory crunch during the
build.

Cheers,
Bill.

#860648#17
Date:
2017-04-27 12:53:07 UTC
From:
To:
256 GB of RAM on the amd64 host, of which only 4GB is available per
process since it's i386.

Lucas

#860648#22
Date:
2017-04-27 13:22:41 UTC
From:
To:
How many threads
getconf _NPROCESSORS_CONF
report ?

I think the problem is that it needs 128M by threads.

Cheers,

#860648#27
Date:
2017-04-27 15:18:38 UTC
From:
To:
64, so yes, that's probably the issue

Lucas

#860648#32
Date:
2017-04-30 14:47:52 UTC
From:
To:
There is only 64MB of virtual memory by threads.
It is not a very realistic setup.
No real i386 box is going to have some much threads.

Honestly I do not really know how to fix this unless I hardcode
a limit for the number of threads in the test suite.

Cheers,