Connect. Communicate. Collaborate. Securely.

Home » Kerio User Forums » Kerio Connect » KMS 6.7.3 and Mail going missing! (Huge number of index.fld going corrupt!)
  •  
marook

Messages: 520

Karma: 3
Send a private message to this user
Hi All,

We now have over 10 users that have their Inbox gone messed up by a corrupt 'index.fld' file.

Senario:
Upgrade from 6.7.2 -> 6.7.3
Within 24 hours first user reports 'months of midding mail' in the Inbox.
Last two weeks, we have seen 10+ users (including my own) reporting a broken Inbox.

Resolutions:
Renaming the 'index.fld' file in the indox and restarting the server seems to solve the issue. Mails are stil in the '#msgs' folder.
Started to happen Huge Time after 6.7.3 upgrade.
Webmail show the same issue.


Just moved a cutomer over from CommuniGate, and we Never saw this on CGP. Not in 10+ years of servicing 300+ users. Kerio hossed first user within 6 hours.. Sad

Anyone else see this??????

Server: OS X 10.5.8 & 10.6.2
Client: IMAP (OS X Mail & few Entourage)

Open bugreport: GIL-320964

Regards,

Jakob Peterhänsel
Consultant - Humac A/S

Apple Certified Support Professional (ACSP)
Apple Certified Technical Coordinator (ACTC)
AppleSeed/CQF member since 1998
Kerio Messaging Partner
  •  
My IT Indy

Messages: 1262
Karma: 40
Send a private message to this user
I've seen that with Mac servers, in fact if you search the forum pretty much every case of this happening, happened on a Mac server.

Something is wonky with HFS+ and KMS I think. We host all larger installs on CentOS for this very reason.

-
My IT Indy
Kerio Certified Reseller and Hosted Provider
http://www.myitindy.com
  •  
marook

Messages: 520

Karma: 3
Send a private message to this user
Well, if CommuniGate Pro can run more users on the same OS with absolutely NO corruption, I have a hard time blaming the filesystem!
So unless it's pinpointed to, say, the hardware RAID card messing it up, I'm pointing fingers at Kerio.. :-/

Regards,

Jakob Peterhänsel
Consultant - Humac A/S

Apple Certified Support Professional (ACSP)
Apple Certified Technical Coordinator (ACTC)
AppleSeed/CQF member since 1998
Kerio Messaging Partner
  •  
j.a.duke

Messages: 356
Karma: 14
Send a private message to this user
marook wrote on Mon, 28 December 2009 18:03
Hi All,

We now have over 10 users that have their Inbox gone messed up by a corrupt 'index.fld' file.

Senario:
Upgrade from 6.7.2 -> 6.7.3
Within 24 hours first user reports 'months of midding mail' in the Inbox.
Last two weeks, we have seen 10+ users (including my own) reporting a broken Inbox.

Resolutions:
Renaming the 'index.fld' file in the indox and restarting the server seems to solve the issue. Mails are stil in the '#msgs' folder.
Started to happen Huge Time after 6.7.3 upgrade.
Webmail show the same issue.


Just moved a cutomer over from CommuniGate, and we Never saw this on CGP. Not in 10+ years of servicing 300+ users. Kerio hossed first user within 6 hours.. Sad

Anyone else see this??????

Server: OS X 10.5.8 & 10.6.2
Client: IMAP (OS X Mail & few Entourage)

Open bugreport: GIL-320964


I've been running 6.x -> 6.7.2 on Macs for 3+ years without any problems. No lost mail, no index file corruption.

I'd suggest making sure that the store directory (or drive) is excluded from Spotlight indexing.

I think once you do that, most, if not all, your issues will go away.

Cheers,
Jon
  •  
marook

Messages: 520

Karma: 3
Send a private message to this user
If that is the case, more people should see it. Default install is on the OS drive, and that is pr. default always indexed by Spotlight.
But we could try!

Regards,

Jakob Peterhänsel
Consultant - Humac A/S

Apple Certified Support Professional (ACSP)
Apple Certified Technical Coordinator (ACTC)
AppleSeed/CQF member since 1998
Kerio Messaging Partner
  •  
My IT Indy

Messages: 1262
Karma: 40
Send a private message to this user
marook wrote on Mon, 28 December 2009 21:20
Well, if CommuniGate Pro can run more users on the same OS with absolutely NO corruption, I have a hard time blaming the filesystem!
So unless it's pinpointed to, say, the hardware RAID card messing it up, I'm pointing fingers at Kerio.. :-/



Maybe, maybe not. Does CGP store it's email in a nested filesystem or in a db of some kind?

All I'm saying is that I've seen this on KMS installs on OS X where the mailstore gets to a certain size or to a certain amount of items contained within it.

-
My IT Indy
Kerio Certified Reseller and Hosted Provider
http://www.myitindy.com
  •  
marook

Messages: 520

Karma: 3
Send a private message to this user
CGP stores the mail just as Kerio does... :-/

Regards,

Jakob Peterhänsel
Consultant - Humac A/S

Apple Certified Support Professional (ACSP)
Apple Certified Technical Coordinator (ACTC)
AppleSeed/CQF member since 1998
Kerio Messaging Partner
  •  
Pavel Dobry (Kerio)

Messages: 5245
Karma: 251
Send a private message to this user
The corruption of folder indexes does not happen when KMS is running. It can happen when the KMS is being stopped (killed) and there are still clients connected to the server. It is closely connected to Mac and HFS+ filesystem.
Simple change in the startup/shutdown script (KerioMailServer) can provide a good workaround.
  •  
My IT Indy

Messages: 1262
Karma: 40
Send a private message to this user
That's it! It always happens when I stopped and restarted the KMS process. Can you fix it for us?

-
My IT Indy
Kerio Certified Reseller and Hosted Provider
http://www.myitindy.com
  •  
marook

Messages: 520

Karma: 3
Send a private message to this user
Ahh, ok!

Might be a GREAT idea to push that into the normal distribution!
And you even KNOW a solution???

Could you please enlighten us then??

Regards,

Jakob Peterhänsel
Consultant - Humac A/S

Apple Certified Support Professional (ACSP)
Apple Certified Technical Coordinator (ACTC)
AppleSeed/CQF member since 1998
Kerio Messaging Partner
  •  
giobbi

Messages: 90
Karma: 0
Send a private message to this user
I have been struggling with this for over a year now. I pretty sure our backup messed things up on kerio, specially when kerios own backup was running simultaniously. Kerio is seems to be extremely sensitive about this, after i turned of all all backup (only backing upp kerios own backup files) the problem stopped.

Store: 250Gb+
  •  
BLTomato

Messages: 57
Karma: 1
Send a private message to this user
Kerio_pdobry wrote on Tue, 29 December 2009 14:52
The corruption of folder indexes does not happen when KMS is running. It can happen when the KMS is being stopped (killed) and there are still clients connected to the server. It is closely connected to Mac and HFS+ filesystem.
Simple change in the startup/shutdown script (KerioMailServer) can provide a good workaround.


Totally false about the "It only happens when KMS is killed)". We experience this issue around once a month where a random user will be missing emails and it takes a reindex to restore them. It happened to 3 users last night is why I'm bumping this thread. KMS has been up the whole time, and there is nothing in the system.log to think otherwise. The mail store is not backed up by any 3rd party means, or spotlight indexed and is on an SAS direct attached RAID.

Now I have to stop KMS, reindex these users and pray like hell that it doesn't corrupt anymore indexes while I do this.

Also, I would love to know what is the "Simple change in the startup/shutdown script (KerioMailServer) can provide a good workaround."? Don't just tease us, tell us how to fix it!
  •  
nbgrant

Messages: 49
Karma: 0
Send a private message to this user
HoosierMac wrote on Mon, 28 December 2009 19:57
I've seen that with Mac servers, in fact if you search the forum pretty much every case of this happening, happened on a Mac server.

Something is wonky with HFS+ and KMS I think. We host all larger installs on CentOS for this very reason.


Agreed Hoosier.


Double check through a shell that the mail is gone. I had an index file (Mac 10.6 Xserve hosting KMS) go bad for a user and suddenly all her mail was gone back until April 09.

A quick count of her #msg folder showed a full 7K files though.

Stop the server, move the index.fld to index.bad and restart. Log in as that user through the web first and it all came streaming back into her Apple Mail right away.

Do an
ls | wc -l
in the #msg folder to double check.

Also, we had to downgrade from 6.7.3 to 7.2 ....6.7.3 patch 1 should be out shortly is what I've heard.

[Updated on: Thu, 07 January 2010 22:17]

  •  
monsterfreecity

Messages: 60
Karma: 2
Send a private message to this user
Pavel, are you going to post the "simple" safe shutdown procedure or not?

This problem has become twice as bad for me on 140 users since they are all now using the calendar, another point of failure for index.fld files.

I've disabled Spotlight, Retrospect, SuperDuper and I have personally logged out all known users only to have randomly corrupt indexes every single time I have to restart the server (like, when I'm trying to fix corrupt indexes, for example).

Extremely frustrating.
BLTomato

Messages: 57
Karma: 1
Send a private message to this user
Please post that safe shutdown, I had a not so fun time with this in the morning. Had to shutdown mail which corrupted a users index, then shutting down again to fix the first index caused a different index to become corrupt. Not a good loop to be caught in during business with over 200 users connected.
Previous Topic: Some help recovering mail please!
Next Topic: Kerio PAM
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: Fri Oct 20 10:52:00 CEST 2017

Total time taken to generate the page: 0.00555 seconds
.:: Contact :: Home ::.
Powered by: FUDforum 3.0.4.