| Kerio Connect RAM leaking [message #136627] |
Fri, 01 September 2017 13:30  |
gvn_kerio
Messages: 7
Registered: September 2017
|
|
|
|
Hi forums
Have anyone else seen same behaviour on their Kerio Connect setups?
Kerio Connect version 9.2.3 (2929) on Windows Server 2012 R2
RAM will slowly crawl up towards 18 GB of RAM, and the server becomes unusable (Server has 24 GB of RAM)
At around 15-16 GB of RAM the server starts getting slow (web interface, and delayed emails)
The only way to fix this is to reboot the entire server, if we try to restart the kerio connect service, it will just stand in the "stopping" state forever.
Here you can see the RAM usage on this Kerio server, the sudden fall in the graph was our last restart of the server.
[img]./fa/4712/0/[/img]
Any help with this would be hugely appreciated, this is quite a big problem for us, we have many customers on this server and we must have 99% uptime.
Regards.
Gert
[Updated on: Fri, 01 September 2017 13:31] Report message to a moderator |
|
|
|
|
|
| Re: Kerio Connect RAM leaking [message #136641 is a reply to message #136639] |
Mon, 04 September 2017 15:53  |
gvn_kerio
Messages: 7
Registered: September 2017
|
|
|
|
It is the Kerio Connect Engine program under the Kerio Connect service.
Nothing else is using the memory, only Kerio
We restarted the Server this weekend, and we are back to 1.7 GB (was around 500 MB just after reboot)
|
|
|
|
|
|
| Re: Kerio Connect RAM leaking [message #136651 is a reply to message #136644] |
Tue, 05 September 2017 10:30 |
gvn_kerio
Messages: 7
Registered: September 2017
|
|
|
|
I've checked Windows Event Viewer, no errors there.
Note I've removed the email address and IP.
The kerio log on the admin panel is spamming this:
[05/Sep/2017 09:32:11] MeetingResponseCommand.cpp: Unable to update the event [uid=0x000000b0] in Calendar. [User: <user<_at_>company.com>, Device: (60ONM6KG2H1UN6UV53B5IK4A44, iPhone, Apple-iPhone8C1/1407.60), IP address: <IP-address>]
I'm seeing a few of these (no near as many as the first one i wrote above)
[05/Sep/2017 09:59:13] ASyncEngineBridgeMessageManager.cpp: ActiveSyncResolveRecipients::ResolveRecipient: Unable to open contact: 0x0000009E from folder '~<user@company.com>/Public Folders/Contacts'. [User: <user<_at_>company.com>, Device: (DHHC3UUBED10L3TFHR7F034L0C, iPhone, Apple-iPhone7C2/1407.60), IP address: <IP-address>]
[Updated on: Tue, 05 September 2017 10:31] Report message to a moderator |
|
|
|
| Re: Kerio Connect RAM leaking [message #136652 is a reply to message #136651] |
Tue, 05 September 2017 10:33 |
gvn_kerio
Messages: 7
Registered: September 2017
|
|
|
|
The warning logs has few of these (10 domains)
[05/Sep/2017 09:49:36] XMPP Server [external]: [773] Failed to process domain <domain-name> (DomainFullSync$ExtDomainsHandler.handle): com.kerio.im.xmpp.api.XmppException: Failed to check database F:\Mailstore\xmpp\database\<domain-name>at com.kerio.im.tigase.derby.DerbyChecker.check(DerbyChecker.ja va:65) at com.kerio.im.tigase.DomainServiceImpl.activate(DomainService Impl.java:165) at com.kerio.im.core.domain.DomainControl.add(DomainControl.jav a:67) at com.kerio.im.core.domain.DomainFullSync$ExtDomainsHandler.ha ndle(DomainFullSync.java:87) at com.kerio.im.connect.AdminApi.getAllDomains(AdminApi.java:21 6) at com.kerio.im.connect.ExtServiceImpl.getAllDomains(ExtService Impl.java:99) at com.kerio.im.core.domain.DomainFullSync.sync(DomainFullSync. java:44) at com.kerio.im.core.domain.DomainFullSync.run(DomainFullSync.j ava:35) at java.util.concurrent.Executors$RunnableAdapter.call(Executor s.java:511) at java.util.concurrent.FutureTask.runAndReset(FutureTask.java: 308) at java.util.concurrent.ScheduledThreadPoolExecutor$S
|
|
|
|
|
|
|
|
|
|
| Re: Kerio Connect RAM leaking [message #148959 is a reply to message #136680] |
Fri, 30 October 2020 08:23 |
ZZZKOT
Messages: 27
Registered: September 2019
|
|
|
|
Hello!
We have exactly the same picture. When I looked at the attached screenshot, I saw a picture as on our server.
We on Kerio Connect 9.2.10 x64 Windows Server, 32 GB RAM, 2-Xeon 2620 2.4 Gz
350 User boxes.
Over time (for us it is about 2 months) the server starts to work slower and slower. Memory consumption in the system reaches 15-16 GB (the kerioconnect.exe process itself is about 6 GB). After that, in 1-2 days the situation deteriorates sharply, the server practically ceases to function.The send queue collects hundreds of emails and is slow. Administration Console pages are not loading. Inside the operating system itself, strong slowdowns are also observed, starting from the login to the server.
Stopping and starting the process does not help (it takes about 30 minutes) - only a complete reboot of the server.
We thought there was a problem with the version, but I see that it has been dragging on for a very long time.
Does anyone have the same server speed issues? Periodically restarting the equipment is not an option for us.
|
|
|
|
| Re: Kerio Connect RAM leaking [message #148961 is a reply to message #148959] |
Fri, 30 October 2020 09:42 |
HOF
Messages: 2
Registered: October 2020
Location: Czech Republic, EU
|
|
|
|
Hello,
we have exacly the same problem: (ver. 9.3.0, Windows server 2016 Std., RAM 12 GB, 4x xeon 3.0, 55 users).
2 weeks ago we had only 8 GB RAM and Connect was stable, but after update Connect (9.3.0) I had to restart 3-times per day,
and after upgrade RAM (12 GB now) I have to restart the whole server every one or two days.
Now I am seriously thinkink about downgrade to older Connect 
|
|
|
|
|
|
| Re: Kerio Connect RAM leaking [message #148963 is a reply to message #148962] |
Fri, 30 October 2020 11:13 |
HOF
Messages: 2
Registered: October 2020
Location: Czech Republic, EU
|
|
|
|
We have the same problem with 9.3.0 - total slowdown of server, we have to restart the whole server every 2nd day.
Windows server 2016, 12 GB RAM, 55 users.
|
|
|
|
| Re: Kerio Connect RAM leaking [message #148966 is a reply to message #148962] |
Fri, 30 October 2020 12:20 |
ZZZKOT
Messages: 27
Registered: September 2019
|
|
|
|
Thanks Martin!
I did not find a description of fixing this problem in patch notes 9.3.0+
Initially it seemed to me that this is purely our problem, until I found this old topic.
But in any case, I'll wait with the installation of 9.3.0 for now, reading the adjacent forum topics 
|
|
|
|
|
|
Members
Search
Help
Register
Login
Home
