The default notification buffer size for the TSM Journal Service is too
small. Look for this setting in the TSMJBBD.INI file and increase it.
You will find this file in the BACLIENT directory. I have been increasing it
to 0x00300000 and my problems in this area have gone away. After increasing
it stop/start the journaling service.
Take care,
Al
;
;
; Notification Buffer Size
;
NotifyBufferSize=0x00050000
;
Alan Davenport
Senior Storage Administrator
Selective Insurance Co. of America
alan.davenport AT selective DOT com
(973) 948-1306
-----Original Message-----
From: Kamp, Bruce [mailto:bkamp AT MHS DOT NET]
Sent: Tuesday, March 04, 2003 11:30 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Journal Service Question
Now I'm getting this in the jbberror.log:
03/03/2003 10:58:38 psFsMonitorThread(tid 18584): Notification buffer
overrun for monitored FS 'E:\'.
03/03/2003 10:58:38 psFsMonitorThread(tid 18584): Reallocating 0x04cccccc
byte notification buffer.
03/03/2003 10:58:39 psFsMonitorThread(tid 18584): Journal DB will be
restarted for FS 'E:\'.
03/03/2003 10:58:45 psFsMonitorThread(tid 18584): Notification buffer
overrun for monitored FS 'E:\'.
03/03/2003 10:58:45 psFsMonitorThread(tid 18584): Reallocating 0x05c28f5b
byte notification buffer.
03/03/2003 10:58:46 psFsMonitorThread(tid 18584): Journal DB will be
restarted for FS 'E:\'.
03/03/2003 10:58:51 psFsMonitorThread(tid 18584): Notification buffer
overrun for monitored FS 'E:\'.
03/03/2003 10:58:51 psFsMonitorThread(tid 18584): Reallocating 0x06e978d3
byte notification buffer.
03/03/2003 10:58:52 psFsMonitorThread(tid 18584): Journal DB will be
restarted for FS 'E:\'.
03/03/2003 12:01:44 psFsMonitorThread(tid 18584): Notification buffer
overrun for monitored FS 'E:\'.
03/03/2003 12:01:44 psFsMonitorThread(tid 18584): Reallocating 0x084b5dca
byte notification buffer.
03/03/2003 12:01:45 psFsMonitorThread(tid 18584): Journal DB will be
restarted for FS 'E:\'.
03/03/2003 15:00:42 psFsMonitorThread(tid 18584): Notification buffer
overrun for monitored FS 'E:\'.
03/03/2003 15:00:42 psFsMonitorThread(tid 18584): Reallocating 0x09f40a25
byte notification buffer.
03/03/2003 15:00:43 psFsMonitorThread(tid 18584): Journal DB will be
restarted for FS 'E:\'.
03/03/2003 18:01:32 psFsMonitorThread(tid 18584): The maximum number of
buffer overun restarts for JFS 'E:\' exceeded,monitor thread terminating.
03/03/2003 18:01:33 JbbMonitorThread(): FS Monitor thread, tid 18584 ended
with return code 122
Is this because the incrimental backup had not run yet?
Thanks,
--------------------------------------
Bruce Kamp
Midrange Systems Analyst II
Memorial Healthcare System
E: bkamp AT mhs DOT net
P: (954) 987-2020 x4597
F: (954) 985-1404
---------------------------------------
-----Original Message-----
From: Andrew Raibeck [mailto:storman AT US.IBM DOT COM]
Sent: Monday, March 03, 2003 11:08 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Journal Service Question
OK... I'd say that this is a good candidate for jbb.
Regards,
Andy
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
Internet e-mail: storman AT us.eyebm DOT com (change eye to i to reply)
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
"Kamp, Bruce" <bkamp AT MHS DOT NET>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> 03/03/2003
08:58
Please respond to "ADSM: Dist Stor Manager"
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: Re: Journal Service Question
I made all the changes & restarted the Journal Service.
This is what the stats look like usually for all the file systems on the
server (4)!
03/03/2003 02:56:08 --- SCHEDULEREC STATUS BEGIN
03/03/2003 02:56:08 Total number of objects inspected: 1,055,640
03/03/2003 02:56:08 Total number of objects backed up: 1,950
03/03/2003 02:56:08 Total number of objects updated: 0
03/03/2003 02:56:08 Total number of objects rebound: 0
03/03/2003 02:56:08 Total number of objects deleted: 0
03/03/2003 02:56:08 Total number of objects expired: 0
03/03/2003 02:56:08 Total number of objects failed: 0
03/03/2003 02:56:08 Total number of bytes transferred: 388.76 MB
03/03/2003 02:56:08 Data transfer time: 32.30 sec
03/03/2003 02:56:08 Network data transfer rate: 12,324.08 KB/sec
03/03/2003 02:56:08 Aggregate data transfer rate: 88.24 KB/sec
03/03/2003 02:56:08 Objects compressed by: 0%
03/03/2003 02:56:08 Elapsed processing time: 01:15:11
--------------------------------------
Bruce Kamp
Midrange Systems Analyst II
Memorial Healthcare System
E: bkamp AT mhs DOT net
P: (954) 987-2020 x4597
F: (954) 985-1404
---------------------------------------
-----Original Message-----
From: Andrew Raibeck [mailto:storman AT US.IBM DOT COM]
Sent: Monday, March 03, 2003 10:33 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: Journal Service Question
You're welcome! :-)
My recommendations:
1) I think the 2 GB file is symptomatic, as the journal db will "break" if
the file exceeds 2 GB in size. This is suspicious, and if the behavior
repeats itself, please open a call with support so that we can do more
in-depth diagnostics.
2) Make sure you following *all* the recommendations in the README file with
regard to NAV, including the NotifyFilter settings.
3) Sorry, I meant to add this in my prior response, but forgot: Change the
NotifyBufferSize from the default value of 0x01000000 to 0x04000000.
3) Yes, you should shut down the journal service, delete the *jdb* files,
then restart the journal service. Again, the first backup will be a full,
but then let's see how it goes from there. Also keep an eye on the *jdb*
files. Unless you have a *huge* number of files that change on a daily
basis, the journal database should not grow so large. And if you have that
many files changing daily, then journal-based backup is probably not a
viable solution.
Best regards,
Andy
Andy Raibeck
IBM Software Group
Tivoli Storage Manager Client Development
Internal Notes e-mail: Andrew Raibeck/Tucson/IBM@IBMUS
Internet e-mail: storman AT us.eyebm DOT com (change eye to i to reply)
The only dumb question is the one that goes unasked.
The command line is your friend.
"Good enough" is the enemy of excellence.
"Kamp, Bruce" <bkamp AT MHS DOT NET>
Sent by: "ADSM: Dist Stor Manager" <ADSM-L AT VM.MARIST DOT EDU> 03/03/2003
08:17
Please respond to "ADSM: Dist Stor Manager"
To: ADSM-L AT VM.MARIST DOT EDU
cc:
Subject: Re: Journal Service Question
FIRST THANKS!!
How large are the journal db files (named like *jdb*)?
Between 1k & over 2 Gig
I am using NAV 7.60 & made the registry change
Also I added the PreserveDbOnExit=1 per the readme!
Do I need to restart any services?
|
