Question: "An error occurred setting the metadata for this dataset"
1
gravatar for eurioste
5 months ago by
eurioste10
eurioste10 wrote:

Hello, I'm just reporting I'm getting "An error occurred setting the metadata for this dataset" for almost all jobs, including simple jobs that previously worked with the data in my workflow, like FastQ groomer jobs. Because of this many jobs are getting stalled or having errors.

software error metadata • 334 views
ADD COMMENTlink modified 5 months ago by Jennifer Hillman Jackson23k • written 5 months ago by eurioste10

I also have jobs just spinning or waiting. I think there's some trouble going on - you're not alone.

ADD REPLYlink written 5 months ago by louisa.pyle10

I'm also having this same or very similar error message with with programs like FASTQC and Trimmomatic. I have workflows I've used repeatedly in the past quite successfully - now they're failing but it seems that random steps in the process are failing. For example, one workflow has two FASTQC steps - somtimes one fails, sometimes both fail but the fact that sometimes one or the other succeeds tells me the problem isn't in my data but in the processing.

ADD REPLYlink modified 5 months ago • written 5 months ago by zachary.gaber80

Hello - We would like to track down this problem. Would one of you be able to send in a shared history link with the problematic datasets left intact?

Generate the history share link and include it in an email to galaxy-bugs@lists.galaxyproject.org (private list). Note the dataset numbers for input/output and also please include a link to this post so we can associate the two. How to generate a history share link: https://galaxyproject.org/learn/share/

Thanks for reporting the problems and our admin will look at this ASAP! Jen, Galaxy team

ADD REPLYlink written 5 months ago by Jennifer Hillman Jackson23k

Thanks Jen! email sent just now. Louisa

ADD REPLYlink written 5 months ago by louisa.pyle10

Thanks Louisa! We believe we know what is going on, fixed the problem, and I am going to post a reply. I'll update if reviewing your history changes anything.

ADD REPLYlink written 5 months ago by Jennifer Hillman Jackson23k

Hello Jennifer, I did what you asked and sent an email containing links to this thread and my shared history with the bug. It's been happening very consistently (more than 75% of the time I try to run these operations I get a bug) for me for at least the past 24 hours so hopefully you guys won't have trouble reproducing the bug and then figuring out what's going on. Thanks for your help and please contact me if you need any information beyond what I put in the email!

Zachary

ADD REPLYlink written 5 months ago by zachary.gaber80

Thanks Zachary for sending this in. I will also look at your history to make sure more is not going on, however also try the advice below now as it is the most likely problem/resolution. Thanks! Jen

ADD REPLYlink written 5 months ago by Jennifer Hillman Jackson23k

Hello Jennifer - I just reran the same workflow on the same data that's been getting bugged - seems to work now! Thanks for fixing this!

ADD REPLYlink written 5 months ago by zachary.gaber80

Oh, great news! Thanks for the update!

ADD REPLYlink written 5 months ago by Jennifer Hillman Jackson23k
2
gravatar for Jennifer Hillman Jackson
5 months ago by
United States
Jennifer Hillman Jackson23k wrote:

Hello all

If you have had job execution problems at Galaxy Main (http:/usegalaxy.org) over the last few days, please do the following:

For successful jobs with metadata problems: Click on the "autodetect metadata" link within the expanded dataset box to reset. These datasets should be fully intact. However, rerunning is also an option.

For jobs that failed, these will have an error similar to: IOError: [Errno 2] No such file or directory: '/galaxy-repl/main/scratch2/XXXX' or Remote job server indicated a problem running or monitoring this job. Rerun the just jobs that failed or the complete workflow, your choice.

Note: If the job errors again with the Remote job server error, the problem could be with your inputs or tool form usage. Troubleshooting help is here. If you cannot determine the problem after reviewing, a bug report can be sent in.

For either: If a workflow was used and the history is complicated with success, failed, and metadata-problem jobs, permanently deleting the prior run and rerunning from the start may be the simplest solution.

Many will be rerunning jobs, so the queues may be longer than normal for certain job types. Please be patient and allow queued jobs to stay queued until executed.

Thanks for reporting the issues and please let us know if the reruns are problematic in this same post as a comment to this reply.

Jen, Galaxy team

ADD COMMENTlink modified 5 months ago • written 5 months ago by Jennifer Hillman Jackson23k
1

Thanks so much Jen and Galaxy team! Foolishly, I started with my longest job, so it's still running. We'll see!

ADD REPLYlink written 5 months ago by louisa.pyle10
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 16.09
Traffic: 103 users visited in the last hour