Re: [RFS] jalview

["James Procter (Staff)" <J.Procter@dundee.ac.uk>]

Hi Nilesh/Pierre

It looks like the malloc-fail is due to jalview trying to grab as much memory as possible on start up. That might need to be controlled for testing, since the tests need at most 512MB.  @Pierre - should be simply a case of adding -

-jvmmemmax=512M 

for all test runs in https://salsa.debian.org/java-team/jalview/-/blob/master/debian/tests/someCommandLineInvocations

The errors about no X11 display being set are non-fatal and look like our problem upstream (well, Ben's perhaps 🙂 )  rather than Pierre's.

Jim.

---

From: Pierre Gruet <pgtdebian@free.fr>
Sent: 07 February 2021 14:45
To: Debian Med Project List <debian-med@lists.debian.org>
Subject: Re: [RFS] jalview

 

Hi Nilesh,

Le 07/02/2021 à 15:30, Nilesh Patra a écrit :
> Hi Pierre,
>
> Thanks for your work on jalview. However, it seems that the autopkgtest
> fails on i386, which will prevent migration to testing.
> It looks like a memalloc problem so maybe this can be simply skipped. Or
> if I'm mistaken, please consider fixing this and uploading.

Yes, thanks for checking! Indeed the memory allocation appears to be
problematic on i386. But I found some ways to fix it in a sensible way
-- I am performing some checks and uploading then.

>
> PS: Please upload with urgency:high so it hits bullseye

Seems relevant. Although the autopkgtest will reduce the delay to 2 days
in any case.

>
> [1]:
> https://ci.debian.net/data/autopkgtest/testing/i386/j/jalview/10322903/log.gz
> <https://ci.debian.net/data/autopkgtest/testing/i386/j/jalview/10322903/log.gz>
>
> Nilesh

Bye,
Pierre

The University of Dundee is a registered Scottish Charity, No: SC015096