Hello all, The memory usage issue is happening again. After 14 days at 544MB, one galaxy python process (which servers web pages, not running jobs) jumps to 9.7GB. The second python process (another web-server) jumps from 1550MB to 13.0GB. Both these increases happen within one hour. Between the galaxy lob, the apache log and the "events" table, I can't find a specific reason for that. It seems that at the time of the memory increase, the only operation was history refresh. Again, any tips on how to deal with it (or even find the cause) would be appreciated. Thanks, -gordon Assaf Gordon wrote, On 07/07/2010 12:16 PM:
Hello all,
After upgrading to a recent version (47411bb09df4), memory usage reaches 12GB per python process after less then 48 hours (with normal usage by users). It only happens with the web-side processes, the job-runner process and the reports process are at their usual usage level.
Debug is off and Interactive is off, The major change is that I've started using SGE (with DRMAA).
Any tips would be very appreciated. -gordon