Connect. Communicate. Collaborate. Securely.

Home » Kerio User Forums » Kerio Connect » Problem with 7.3 Message Storage
  •  
hbianchi

Messages: 121
Karma: 8
Send a private message to this user
Since I installed the 7.3 version in our production environment, I received some unexplained claims from users about:

- old messages (very old) appearing in their mailboxes without any recovering procedure.
- more space used and quota usage increased or even exhausted without any special reason.

I noted the the "warning log" has showing, permanently, messages as I copy below

[09/Nov/2011 16:33:52] Inconsistent message index detected in folder ~juangonzalez<_at_>perfil.com/Sent Items.
[09/Nov/2011 16:33:52] Restoring index in folder ~joseperez<_at_>perfil.com/Sent Items. Flags and status information may be lost

It is like a process is following the database and making some kind of process in it that is casing this "problems".

  •  
Pavel Dobry (Kerio)

Messages: 5245
Karma: 251
Send a private message to this user
Starting with version 7.3.0 Kerio Connect periodically starts background message store checking. This can reveal bugs in message index consistency. Ie. if some messages were missing in the index file (due to some bug, error or administrator manipulation) then the messages are accessible now and also user quota is recalculated. So, these messages were in user folders but they were not accessible.
  •  
hbianchi

Messages: 121
Karma: 8
Send a private message to this user
Thank you for the explanation and I suggest to put it as a warning in the release notes.

After several years of operation, a big installation with more than 600 users, has lots of "missing" meesages. Such automatic index rebuilding process, results in several users claiming for "older messes recovered without any action" or "starage quota exhausted withour any reason" and we, the IT crew, were not aware of that at all.

Regards.
  •  
id t

Messages: 50
Karma: 0
Send a private message to this user
7.3 completely brought my server (180 users) to his knees. I upgraded at 1.30am and everything seemed fine, but this morning, when people started to come in the office and open their mail, all hell break loose; I understad it's background checking all folders for inconsistencies, but with mailboxes of over 10GB this is getting ridiculous, message are bounced in out of queue because :


[10/Nov/2011 16:37:37] folder_manager.cpp: Cannot open folder ~alessandro<_at_>xxxx.nl/INBOX: locked after 20 retries.

presumably because that folder is locked by the check process (which I cannot stop). If I still have a job tomorrow, I'm gonna be very angry with someone.
  •  
giobbi

Messages: 90
Karma: 0
Send a private message to this user
10GB per mailbox??? Omg.
  •  
id t

Messages: 50
Karma: 0
Send a private message to this user
ehm..names are hidden to protect the guilty

[img]http://www.ams0.org/huge_mboxes.png[/img]
  •  
hbianchi

Messages: 121
Karma: 8
Send a private message to this user
In my case, with 600 users with a media of 500 MB to 1 GB mailboxes, I don't have performance problems, but I had quota problems, because the rebuilding engine "recovered" all the "lost" messages that were in the mailboxes and many users claimed because they stop to receive mails. Rebuilding process should have some king of configuration to select the day/time when it would run, and also it should be possible to stop.
  •  
id t

Messages: 50
Karma: 0
Send a private message to this user
Can we downgrade? It's 24hrs and I can't see any other solution
  •  
id t

Messages: 50
Karma: 0
Send a private message to this user
CPU utilization still high, but at least I've found a way to empty the cache: increasing the queue size to 64 (now at 128) delivered the >1200 message in the queue in less than 5 minutes. There are still many for a specific user whose index is being rebuilt now, but seems like it's working (we'll see tomorrow when the people come in).
  •  
giobbi

Messages: 90
Karma: 0
Send a private message to this user
id t wrote on Fri, 11 November 2011 00:19
ehm..names are hidden to protect the guilty

[img][url]http://www.ams0.org/huge_mboxes.png[/url][/img]


Curious about one thing; How does the client affects with this size of mailboxes? And on iPhone? Speed? Whats the message count? : )
  •  
id t

Messages: 50
Karma: 0
Send a private message to this user
The speed was perfectly fine under 7.2.1, the mailstore is on a RAID10, 8-disks, SAS, 15k rpm 600GB. 90% Apple Mail+iCal, 15 laptops with Outlook, many many iphones, blackberries and androids.

Previous Topic: Performance tweaks
Next Topic: Outlook 2010 SPA with Linux server?
Goto Forum:
  


Disclaimer:
Kerio discussion forums are intended for open communication between forum members and may contain information and material posted by members which may be useful in learning about Kerio products. The discussion forums are not intended to provide technical support for any specific product. Any information implied or expressed in the discussion forums is that of the posting member. Kerio is in no way responsible for the information posted in the forums, or its accuracy. Kerio employees may participate in the discussions, but their postings do not represent an offical position of the company on any issues raised or discussed. Kerio reserves the right to monitor and maintain the forums to promote free and accurate exchange of information.

Current Time: Sat Nov 18 09:32:37 CET 2017

Total time taken to generate the page: 0.00559 seconds
.:: Contact :: Home ::.
Powered by: FUDforum 3.0.4.