Hi Francis, We are not using activesync. Because we thought about activesync at first, we removed it explicitly. I wrote the line in the config, but restarting the service shows "/etc/default/sogo: line 5: SoGoDebugRequests: command not found“ (I tried different spellings as well)
We have PREFORK=10 in our sogo.conf. Which is also the only setting we modified (and the only entry in our sogo.conf). `ps aux | grep sogo` shows 11 Processes. Which is expected. The machine has 6GB of RAM, which should be plenty for 10 Processes with 384MB each (this is, what sogo says, when its starting up). Thanks for your help! ~Filias > Am 05.03.2015 um 03:20 schrieb Francis Lachapelle <[email protected]>: > > Hi Filias > > On Mar 2, 2015, at 4:33 AM, Filias Heidt <[email protected]> wrote: > >> I have a strange problem, where SoGo starts to use up all memory (currently >> 6GB) which then crashes the whole server. I can’t find any directly related >> entry in the logs, however I see a lot PROPFIND from the same IP, just >> before it stops. >> >> The last log entry before the crash is always just some random operation, >> the next entry is >> Mar 02 09:34:16 sogod [1295]: version 2.2.16 (build @shiva.inverse >> 201502121302) — starting >> >> This happens roughly once per day, at seemingly random times. The only thing >> left to do, if this happens, is power-cycling the server. >> >> Any tips, how I can debug this further? Or has anybody experienced the same >> problem? > > Are you using ActiveSync? If so, what is the average number of clients that > connect to your server simultaneously? > > Do you have any slow queries (LDAP/SQL)? You can set SOGoDebugRequests = YES > to get some timings. > > How many sogod processes do you have? Can you share your sogo.conf? > > > Francis-- > [email protected] > https://inverse.ca/sogo/lists
signature.asc
Description: Message signed with OpenPGP using GPGMail
