Description of problem: sosreport should collect /var/run/glusterfs also as it has geo-replication related log/files Version-Release number of selected component (if applicable): How reproducible: Steps to Reproduce: 1. 2. 3. Actual results: sosreport does not pick /var/run/gluster* Expected results: Additional info:
What sort of size are we looking at for these files? Depending on the answer it may be worth enabling either default or optional size-limits for this collection.
I believe /var/run is meant for daemon state file like pid. Why logs go into /var/run/gluster* directory?
(In reply to Bala.FA from comment #3) > I believe /var/run is meant for daemon state file like pid. Why logs go > into /var/run/gluster* directory? Geo-Replication uses /var/run/gluster/ as a working directory which hold processed changelogs. There is one log file generated by geo-replication (libgfchangelog.so to be precise). I'll change that to generate log file in the standard log file directory. Collecting /var/run/gluster/ will be a good idea anyway in order to debug geo-replication changelog issues.
/var/run/gluster is for keeping any 'state' information. Ie, anything used by the process while its running. (pid is one of such state info). log here is not logging of msgs, but the 'changelog' as Venky rightly said. So please collect /var/log/gluster
I don't see a corresponding RHEL bug for this RFE. What's the status? It's easy enough for us to add this to sos but I'd like to see some examples of the files / directory structure that we should expect.
I updated the working dir location in bz 1022523. Closing this bug as duplicate. *** This bug has been marked as a duplicate of bug 1022523 ***