 |
 |
Unrelated, but shouldn't setting the IO priority of the xrootd process help out here? Seems like bumping the priority of xrootd and doing an ionice on logrotate would help.
You may also want to look into decreasing your VM dirty threshold - if the logrotate process does enough dirtying to trigger writeback, the kernel should start to throttle logrotate; if it doesn't trigger the VM dirty threshold before it exits, the writeback probably has a better chance of affecting Xrootd.
Just some thoughts for helping survival.
—
Reply to this email directly or view it on GitHub.
Use REPLY-ALL to reply to list
To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1