I am using the settings below without problems. Notice the very large memory allocated ( 6GB ). Also note that I was also experiencing the behavior you mentioned. In my case it was because of a contact duplicating bug that kept on re-importing the contacts and it would thus take a very long time to process in MySQL. Once I imported the contacts via the web interface the bug went away and the sogod processes would die when expected. WOWorkersCount = 120; SOGoMaximumPingInterval = 3540; SOGoMaximumSyncInterval = 3540; SOGoInternalSyncInterval = 60; SOGoMaximumSyncWindowSize = 30; SOGoMaximumSyncResponseSize= 2048; WOWatchDogRequestTimeout=60; SxVMemLimit = 6100; Make sure to also make the proper adjustments in your apache configuration file: ProxyPass /Microsoft-Server-ActiveSync http://127.0.0.1:20000/SOGo/Microsoft-Server-ActiveSync retry=60 connectiontimeout=5 timeout=3600 On Tue, Apr 12, 2016 at 06:24 AM, Florian Unger wrote: Hi,
all the topics I found on the net spoke about setting small values for sync to solve issues so I tried these settings: SOGoMaximumSyncResponseSize = 512; SOGoMaximumSyncWindowSize = 256; which did not not help. I will try your configuration with a large response size and see if that is doing the trick. Thanks Florian On 11.04.2016 15:51, Christian Jensen (cmjscrip...@outlook.dk (mailto:cmjscrip...@outlook.dk)) wrote: Hi. have you tried setting 'SOGoMaximumSync****' i had almost the same issue and after setting them Office 2013 synchronized just fine. SOGoMaximumSyncResponseSize = 10240; SOGoMaximumSyncWindowSize = 256; Regards Christian Den 11-04-2016 kl. 15:35 skrev Florian Unger (florian.un...@intego.de (mailto:florian.un...@intego.de)): Hi, I am having some trouble with a small scale SOGo installation (less than 10 users) using SOGo 3.0.2 with ActiveSync and Outlook 2013 clients. I have SOGo configured for 15 workers and these are the ActiveSync specific settings I am using: WOWorkersCount = 15; SOGoMaximumPingInterval = 600; SOGoMaximumSyncInterval = 600; SOGoInternalSyncInterval = 30; WOWatchDogRequestTimeout = 6; SxVMemLimit = 768; /* Debug */ SOGoEASDebugEnabled = YES; After some time processes keep hanging in the same request until all workers are occupied and I get error "[WOWatchDog]> No child available to handle incoming request!" At the end is an example of the log file of process 6808 and the last requests it got. I also made backtraces with gdb from that process (see attachment) - the process was either in function NGDescriptorRecv or in a sleep from SOGoActiveSyncDispatcher. I tried increasing timeouts up to 1 hour as suggested in the Active Sync tuning hints from the manual, but that did not help. Processes were sitting just longer until they got killed by the watch dog. Any hint how to resolve this issue would be great. Best regards Florian -- users@sogo.nu (mailto:users@sogo.nu) https://inverse.ca/sogo/lists (https://inverse.ca/sogo/lists) -- users@sogo.nu https://inverse.ca/sogo/lists
