Over disk quota and not going back

Question: Over disk quota and not going back

21 months ago by

Sweden

dcf1007 • 40 wrote:

Hi, I was loading some list containing SRA datasets and I went over quota. I tried to solve by purging everything I had in my history and datasets. Now the datasets and the history I retained are less than 1mb but the system still says I am using 788GB. I have logged out, waited several minutes and logged in again but the quota continues at 788GB.

Am I missing anything to free the space? I have deleted, deleted permanently the files and purged the histories.

Thank you in advance, David

over disk quota • 606 views

ADD COMMENT • link •

modified 20 months ago • written 21 months ago by dcf1007 • 40

20 months ago by

Dannon Baker ♦ 3.7k

United States

Dannon Baker ♦ 3.7k wrote:

Sorry for the trouble, I've recalculated your quota and you should be good to go. We've overhauled the logout-on-recalculate logic and it should work once the new code is deployed. Let me know if you have this problem again before that, and we'll get it taken care of.

ADD COMMENT • link written 20 months ago by Dannon Baker ♦ 3.7k

Thank you very much!

Best regards, David

ADD REPLY • link written 20 months ago by dcf1007 • 40

Dear Dannon,

I am having problems again. First of all, according to Galaxy I am using -35.2GB of space (-14%). At the same time, I am trying to run BOWTIE2 in a small dataset (500mb more or less) but I keep on getting empty SAM/BAM and empty stats files. The stderr shows the following message:

Building a SMALL index close failed in file object destructor: IOError: [Errno 28] No space left on device /jetstream/scratch0/main/jobs/15310938/command.sh: line 82: /jetstream/scratch0/main/jobs/15310938/return_code: No space left on device /jetstream/scratch0/main/jobs/15310938/command.sh: line 83: /jetstream/scratch0/main/jobs/15310938/metadata/__instrument_core_epoch_end: No space left on device

Obviously I am not over my quota (I have no data saved at all) so I don't really know what is going on. Also, sending the task to Stampede seems to solve the issue.

Best regards, David

ADD REPLY • link modified 20 months ago • written 20 months ago by dcf1007 • 40

Similar posts • Search »