Dear Galaxy,
I left the software running for 5 days doing a cufflinks job on some TopHat RNA seq results. The history items just stayed grey with an exclamation mark in the top left corner for the entire time. Why won't my Cufflinks work on this data? Surely 5 days is way too long for the queueing issues with the server?
Thanks Richard
What Galaxy are the jobs running on? http://usegalaxy.org or ??
yes, usegalaxy.org
I just filed another question as I went back to the accepted alignments TopHat results and I get the following error when I click on them;
"An error occurred setting the metadata for this dataset Set it manually or retry auto-detection"
But when I do autodetect I get the following message;
"This dataset is currently being used as input or output. You cannot change metadata until the jobs have completed or you have canceled them."
(I have cancelled the Cufflinks jobs, so there shouldn't be anything programs using the datasets)
Ok, so the jobs were queued, and now have run. For those, my guess is that you had several jobs running at the same time (same history, or different histories) causing the queue delays. Waiting for these to finish was the correct choice.
Try "Purging deleted datasets" for the Cufflinks jobs (in the History menu - gear icon - at the top of the History panel). Next refresh the history panel with the double arrow icon at the top. Then try to re-set the metadata again for the Tophat jobs. These Cufflinks jobs need to clear out of the database before inputs can be modified. It may take an hour or so.
There is a known issue with Tophat on some clusters where the accepted hits output will sometimes need to have metadata reset before moving forward with other jobs. This is related to Jetstream and our administrator is working on it with priority.
Sorry for the complications you ran into, seems like you hit a batch of them all at once! Jen, Galaxy team
Thanks Jen,
I have purged the files and still the error is coming up when I autodetect on the accepted hits. I'll wait an hour or two, then try again.
Ok. I can let you know that some users have decided to re-run the Tophat jobs, allow them to complete, re-set metadata, then start downstream jobs that use them. I don't think this will be necessary. Instead, wait, as you are doing, for the other started jobs to fully exit the system.
So I waited until this morning for the files to be purged, but I still got the error message. As a result I deleted the TopHat results and reran them, but now they all fail? The history items go red saying "tool error, Remote job server indicated a problem running or monitoring this job." Is there something wrong with the servers at this time?
