asked on

Trenmicro scanmail crashing Dominos server

every month or so, my Dominos server (2003) crashes I typically find what I think is the culperite, outlined below:
I've been advized to disable transaction logging for the smvlog.nsf (the scan mail) database, do you think it would help?

-- nsd_all_W32I_MyServer_04_22@22_40.log contents --

------ Shared OS Fields -------

      Start Time = 04/15/2005 07:51:51 PM
      Crash Time = 04/22/2005 10:41:43 PM
      Error Message = PANIC: Logging SubSystem Failure (D:\Lotus\Domino\Data\smvlog.nsf)
      SharedDPoolSize = 1000000
      FaultRecovery = 0x00010013
      Thread [nRepScan:0d68:0002]/[nRepScan:0d68:0d6c] (d68/2/d6c)
            caused Static Hang to be set
      ConfigFileSem = ( SEM:#0:0x010d) n=0, wcnt=-1, Users=-1, Owner=[]
      FDSem = ( RWSEM:#12:0x410f) rdcnt=-1, refcnt=0 Writer=[] n=12, wcnt=-1, Users=0, Owner=[]

sravikiran

Do you have your transaction logs on the same physical drive as the data folder? If yes, try move it to a different physical drive. In the worst case scenario, atleast move your data & translogs to a different drive letter.

This error may happen due to the high disk I/O on a single disk.

sravikiran

After moving the transaction logs to a different physical drive, set "TRANSLOG_Path" NOTE.INI setting to point to the trans log location.

Bozzie4

No ! Don't use the notes.ini setting, use the Translog stuff in the server document.

But to solve your problem, you may need to apply a trend micro patch - contact them !

cheers,

Tom

Bozzie4

Because I also read on notes.net, that other people mention similar problems.

cheers,

Tom

Marketing_Insists

ASKER

sravikiran
> This error may happen due to the high disk I/O on a single disk.

The transaction logs are indeed on the same physical drive as the OS (a mirrored raid), but on it's own partition.

One thing though, these crashses are happening during a automated backup with backup exec 9.

BTW, happned again. About the same time as before.

------ Shared OS Fields -------

      Start Time = 04/30/2005 11:40:46 AM
      Crash Time = 05/03/2005 10:36:32 PM
      Error Message = PANIC: Transaction log is damaged!
      SharedDPoolSize = 1000000
      FaultRecovery = 0x00010013
      Thread [ nSERVER:073c:0055]/[ nSERVER:073c:0a00] (73c/55/a00)
            caused Static Hang to be set
      ConfigFileSem = ( SEM:#0:0x010d) n=0, wcnt=-1, Users=-1, Owner=[]
      FDSem = ( RWSEM:#12:0x410f) rdcnt=-1, refcnt=0 Writer=[] n=12, wcnt=-1, Users=0, Owner=[]

ASKER CERTIFIED SOLUTION

Bozzie4

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

qwaletee

WHAT?! Stop the scanner? Run! Hide! Not unless you are also turning off the SMTP listender, and even then...

If you are using transaction logs, then I would hope you are using a backup agent that makes use of those logs, i.e., one that backs up the logs and can use the logs to do transaction-level/point-in-time restores.

Is that the case?

Bozzie4

Stop the DATABASE scanner, Q, that's not really the same as the MAIL scanner in Trend Micro Scanmail terms.

Bozzie4

And what you definitely must not do, is run a scheduled scan during backup hours. Also, avoid all other activity on the server during backup (like running updall)

cheers,

Tom

qwaletee

Bozzie4,

> not really the same as the MAIL scanner
Sorry, missed that, I so rarely pay any attention to REPSCAN, I guess I just assume TMMSCAN without bothering to read the adectives ;)

- qwaletee

Bozzie4

And the Trend micro task names changed, too :-) It's SMDreal now :-)

cheers

Tom

Marketing_Insists

ASKER

OK thanks, I've gotten some good hints here.

It seems to be a issue with trendmicro and perhaps conflicting agents.

qwaletee

There was also a recent problem with a Trend apttern file that woudl chew CPU cycles, but you are unlikely to have experienced it.