Dear galaxy maintainers, we have the problem that killing jobs running on drmaa before they run leading to a deleted_new state, kills our handlers (sometimes?). The job gets sent to the handler, but the handler never acknowledges the jobs (local and drmaa) and all submitted jobs afterwards stay in submitted state. Normally submission works fine, running also. Is such a problem known to anybody else? in my universe_wsgi.ini: enable_job_recovery = False I don't know which other settings could affect this. Could it be an old drmaa library, maybe because the SGE was updated after galaxy installation? Other ideas? thank you very much, ido
Hi Ido, I just found you post, and I think we are having the similar issue as your. (I posted it in the mailing yesterday, you can find it for the detail of my problem). I am wondering if you are able to fix your problem? Cheers, Derrick On Tue, Dec 11, 2012 at 12:13 AM, Ido Tamir <tamir@imp.ac.at> wrote:
Dear galaxy maintainers,
we have the problem that killing jobs running on drmaa before they run leading to a deleted_new state, kills our handlers (sometimes?). The job gets sent to the handler, but the handler never acknowledges the jobs (local and drmaa) and all submitted jobs afterwards stay in submitted state.
Normally submission works fine, running also.
Is such a problem known to anybody else?
in my universe_wsgi.ini: enable_job_recovery = False I don't know which other settings could affect this.
Could it be an old drmaa library, maybe because the SGE was updated after galaxy installation? Other ideas?
thank you very much, ido
___________________________________________________________ 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:
No the problem is still there, but we have not updated yet since beginning of december. maybe this is related: http://gmod.827538.n3.nabble.com/segfault-in-libdrmaa-gt-galaxy-front-end-fa... best, ido On Jan 21, 2013, at 5:45 AM, Derrick Lin wrote:
Hi Ido,
I just found you post, and I think we are having the similar issue as your. (I posted it in the mailing yesterday, you can find it for the detail of my problem).
I am wondering if you are able to fix your problem?
Cheers, Derrick
On Tue, Dec 11, 2012 at 12:13 AM, Ido Tamir <tamir@imp.ac.at> wrote: Dear galaxy maintainers,
we have the problem that killing jobs running on drmaa before they run leading to a deleted_new state, kills our handlers (sometimes?). The job gets sent to the handler, but the handler never acknowledges the jobs (local and drmaa) and all submitted jobs afterwards stay in submitted state.
Normally submission works fine, running also.
Is such a problem known to anybody else?
in my universe_wsgi.ini: enable_job_recovery = False I don't know which other settings could affect this.
Could it be an old drmaa library, maybe because the SGE was updated after galaxy installation? Other ideas?
thank you very much, ido
___________________________________________________________ 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:
participants (2)
-
Derrick Lin -
Ido Tamir