search

veg / hyphy

HyPhy: Hypothesis testing using Phylogenies
http://www.hyphy.org
Other
217 stars 69 forks source link

Issues trying to run GARD #1129

Closed Frances-Ratcliffe closed 4 years ago

Frances-Ratcliffe commented 4 years ago

Hi, I am having persistent issues trying to run GARD, any help would be greatly appreciated. My data is a clustalW multiple alignment in fasta format, it is DNA nucleotides. I have checked that the sequence translates to protein and is in the correct reading frame. The error I get doesn't give any clues as to how to solve the problem: "We're sorry, there was an error processing your job. Please try again, or visit our GitHub issues and create an issue if the issue persists" Many thanks, Frances

stevenweaver commented 4 years ago

Dear @Frances-Ratcliffe,

Thank you for reporting this issue. It appears two of our nodes is having trouble allocating enough memory. We will remove them from the available nodes to submit GARD jobs to.

Best, Steven

stevenweaver commented 4 years ago

Dear @Frances-Ratcliffe,

A remedy has been put in place, and we will actively monitor to ensure the issue has been resolved in full throughout the day.

Best, Steven

Frances-Ratcliffe commented 4 years ago

Brilliant, thank you very much. I will try to run it again now and let you know if there are any continuing problems. Is there a maximum number of sequences I can run? I have 230pm fragment and 131 different variants. Best, Frances

Frances-Ratcliffe commented 4 years ago

Dear @stevenweaver It works when I submit 20 sequences but when I try 50, 100 or my whole data set (131 seqs) I still get the same "We're sorry, there was an error processing your job. Please try again, or visit our GitHub issues and create an issue if the issue persists" error message.

Is there a limit to the number of sequences that can be submitted or am I missing an option/setting?

Thanks, Frances

stevenweaver commented 4 years ago

Dear @Frances-Ratcliffe,

The error you are seeing is typically only seen once a dataset has passed validation and has been sent to the cluster for processing. Unfortunately, all GARD jobs, besides one that was rightfully terminated, has completed successfully since I've posted the last comment. Do you have an id that is associated with these failing errors? (It is the datamonkey url at the time of failure).

If so, can you email it to me at sweaver@temple.edu?

Best, Steven

stevenweaver commented 4 years ago

Issue remedied and further correspondence via email. Related to #765 and veg/datamonkey-js#197.