Job execution delays due to input dataset states

Question: Job execution delays due to input dataset states

12 months ago by

lippsdd • 0 wrote:

I began running several ht-seq count analyses on Monday, October 16th. Since then they have appeared in my history but still are grey and show a small clock in the corner which I'm assuming means they have not began to run. I have encountered several problems like this in the past where I submitted a task and it never ran (even after waiting for several weeks to a month). Any assistance/insight into this issue would be greatly appreciated!

queue tophat jobs metadata hisat2 • 362 views

ADD COMMENT • link •

modified 12 months ago by Jennifer Hillman Jackson ♦ 25k • written 12 months ago by lippsdd • 0

12 months ago by

Jennifer Hillman Jackson ♦ 25k

United States

Jennifer Hillman Jackson ♦ 25k wrote:

Hello,

The input Tophat jobs each have a metadata problem that must be corrected before used as input to downstream tools. Click on the link in the yellow box of the expanded datasets to reset the metadata. Tophat is known to produce datasets in this state, which can complicate using it. Tophat is considered deprecated with HISAT2 being the replacement.

That advise is for the current jobs that are queued. For prior delays, especially if that long, I suspect there was also some type of input problem but cannot confirm that. However, if it comes up again and you are waiting more than 2 days, please let us know and we can help check to find out what is going on.

Galaxy tutorials, several of which include HISAT2 with example settings for different use cases: https://galaxyproject.org/learn/

Thanks, Jen, Galaxy team

ADD COMMENT • link written 12 months ago by Jennifer Hillman Jackson ♦ 25k

Hi Jen, thanks for the answer, I'm having the same problem but with the Map with BWA-MEM tool. Could you be a little more specific on which yellow box to click (maybe a screenshot)? I can't find any yellow box.

ADD REPLY • link written 12 months ago by sarahstainbrook2018 • 0

Hi Sarah,

Your mapping jobs are probably just queued. The queue is busy right now. Allow these to stay queued and avoid deleting and restarting. We talked about this at Gitter but I wanted to leave a comment in case others run into problems with job delays.

lippsdd had a specific problem where the inputs (already completed inputs) to a tool were in a problematic state, causing downstream tools to stay queued (until that state was corrected).

This was due to a server problem that was just now fixed for many mapping tools - we are still testing, our jobs are also queued - and when done and pass, this ticket will be updated. Ps: I added a graphic for what the metadata warning looks like to help, good idea, thank you! https://github.com/galaxyproject/usegalaxy-playbook/issues/62

Thanks! Jen, Galaxy team

ADD REPLY • link modified 12 months ago • written 12 months ago by Jennifer Hillman Jackson ♦ 25k

Similar posts • Search »