I posted last week about this, and was told it would be slow. Files will trim but then wait for RNA star (some for 5 days) without loading it. No error messages. I deleted them and tried again, trying different trimming programs. I included 1 that had previously successfully trimmed and aligned. Is this just a long waiting time or is there a bug. If it is a long wait, how long should I plan for?
Thanks
Dear @jlocker,
Please read this post, maybe it helps. Long waiting times with RNA STAR can be the case. Nevertheless, 5 days seem quite long.
The only case a problematic link with a trimming tool and RNA STAR exists is, when you have paired-end sequencing data and you forgot to keep both mates synchronized, i.e., you just trimmed and filter one of the mates (files) and the second has read IDs that do not exist in the first file anymore.
If there would be a serious bug, then RNA STAR would abort or result in an error that is clearly visible by a red box.
I hope I could help and best wishes,
Florian
Hi Florian,
2 of 3 paired libraries (trimmed on Fri) ran this AM; the 3rd is still pending. Is 3 days the now the expected wait, or is there a system problem that will get better? I completely processed 3 other libraries last Monday and they finished on the same day.
Thanks,
Joe
3 days is still quite long, well, depending on how big your files are. I dont know of a system error. I will ask in the Galaxy meeting. But maybe try refresh the history and maybe restart the job.
Cheers,
Florian
Hi Florian and Joe,
I was the originator of that earlier RNA STAR post. My sense is that something is still amiss. Jobs that always in the past took a day or two now take weeks. It doesn’t appear to be a problem with the job parameters themselves because sometimes one file in a collection will complete successfully, whereas the other files will stay queued indefinitely (>2 weeks). After this week or more of the job being queued, sometimes the job information appears corrupted (for instance, it will say the job was initiated in 2015), but there is never an explicit error–it just remains queued. This occurs whether or not I use trimmomatic prior to RNA STAR. I hope these observations might be helpful.
Best,
Phillip
Dear @pad59,
Thank you very much for your information. I will forward it to the Galaxy community and we will investigate what goes wrong.
Sorry for the inconvenience at the moment.
Cheers,
Florian
Dear @pad59,
Are you operating on usegalaxy.eu or usegalaxy.org? Or a total different galaxy instance?
Hi @Flow
I encountered these issues on usegalaxy.org. I couldn’t successfully complete an RNA STAR job for several months, even after my note here, so I thought it must be something on my end that was the problem. Therefore on Sep 30 I deleted my usegalaxy.org account and started a usegalaxy.eu account. RNA STAR behaves totally normally on the latter. Therefore I think the problem is somehow at the usegalaxy.org server. Please let me know if I can help in any way.
Phillip
Dear @pad59,
I informed the community. I think this is maybe a resource problem or something odd with the queuing/ setup of the tool.
I am not sure how long it will take to fix this. But I hope it will soon be fixed.
Thank you very much for your patience.
Best wishes,
Florian
Hi all – The server has a banner up about job resources. Once it clears, then all will be back to normal. Certain tools are limited to one active job at a time. Those same tools compete with other tools for cluster resources.
The best strategy is to queue up jobs (ideally with a workflow) then allow the work to process. It is fine to have one active account at each public Galaxy server. Each hosts different tools, resources, etc.