Hi John, works great, thanks for your effort!
> I believe the following changeset fixes this problem. > > https://bitbucket.org/jmchilton/galaxy-central-lwr/commits/3b6c209d2d078bad2fa9957fe989345d3d2533ae > > I have incorporated the fix into pull request 138 that has other job > runner fixes and enhancements. > > -John > > P.S. > > Peter's point about the tasked job runner being frequently broken is a > good one, it has been broken more often than not over the last 6 > months. Digging through Trello it is clear the same applies to > composite datatypes. If only there was a big important feature that > depended on these that we could push through to ensure these > enhancements that are important to the community got more attention > :). Maybe its possible to run a few selected toolshed tools in an automatic test suite before each release or after such large changes. We have really complex tools in the toolshed that uses mainly all features available. Have a nice weekend! Bjoern > On Sun, Mar 17, 2013 at 11:00 AM, Peter Cock <p.j.a.c...@googlemail.com> > wrote: > > On Sun, Mar 17, 2013 at 3:40 PM, Björn Grüning > > <bjoern.gruen...@pharmazie.uni-freiburg.de> wrote: > >> Am Sonntag, den 17.03.2013, 15:26 +0000 schrieb Peter Cock: > >>> On Sun, Mar 17, 2013 at 3:13 PM, Björn Grüning wrote: > >>> > Hi all, > >>> > > >>> > can anyone confirm that the parallelism features is not working in > >>> > galaxy-central? Parallelism is among others used by the blast+ wrappers. > >>> > > >>> > The only error message I get is: > >>> > > >>> > galaxy.jobs DEBUG 2013-03-17 16:08:27,729 (64174) Working directory for > >>> > job > >>> > is: > >>> > /media/data/web/galaxy-central/database/job_working_directory/064/64174 > >>> > galaxy.jobs.handler DEBUG 2013-03-17 16:08:27,750 (64174) Dispatching to > >>> > tasks runner > >>> > galaxy.jobs.handler ERROR 2013-03-17 16:08:27,750 put(): (64174) Invalid > >>> > job runner: tasks > >>> > galaxy.datatypes.metadata DEBUG 2013-03-17 16:08:28,079 Cleaning up > >>> > external metadata files > >>> > galaxy.jobs.handler INFO 2013-03-17 16:08:28,120 (64174) Job dispatched > >>> > > >>> > It's working fine for me with galaxy-dist. > >>> > Thanks! > >>> > Bjoern > >>> > >>> Which hg revision are you on from galaxy-central? It was > >>> working for me last time I tested, but my setup isn't fully > >>> up-to-date with the development repository. > >> > >> I'm on 9088:1aa6224b32c2. Which hg revision are you using? Maybe I can > >> track down the commit. handler.py changed a lot recently :( > > > > So yours is very up to date, just 20 hours ago according to bitbucket: > > https://bitbucket.org/galaxy/galaxy-central/commits/1aa6224b32c2 > > > > For the default branch (which I merge into my working branch), > > I last updated 24 days ago - over three weeks: > > https://bitbucket.org/galaxy/galaxy-central/commits/45f1d93124ad > > > > $ hg id > > 45f1d93124ad > > > > $ hg log -l 1 -b default > > changeset: 9548:45f1d93124ad > > parent: 9546:1adf6fdd9c49 > > user: InitHello <inithe...@gmail.com> > > date: Thu Feb 21 11:06:11 2013 -0500 > > summary: Changed to_safe_string to use markupsafe.escape for > > unsafe characters. > > > > Assuming I'm right and parallelism was working fine with that > > commit, we've only narrowed this down to the last 24 days. > > If I update mine and it breaks, I'll let you know. > > > > Peter > > > > ___________________________________________________________ > > Please keep all replies on the list by using "reply all" > > in your mail client. To manage your subscriptions to this > > and other Galaxy lists, please use the interface at: > > > > http://lists.bx.psu.edu/ ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/