TSM Server: 4.2.3.0 (Library manager and client) for Solaris (2.8)
3494 Library with SCSI attached 3590 drives
IBMtape 4.0.4.5
At what appears to be random times, we have mount failures where we should
be in a media wait state... We see mount failures for scratch also, with
essentially the same sequence without the volume name specified or the
backup failure message.
TSM Library manager will get
ANR8779E Unable to open drive {blah}, error number=16
Immediately followed by dismount failure
AN8469E Dismount of 3590 volume {blah} from drive {blah} in library 3494lib
failed.
And the client will see
ANR9779E Unable to acquire a drive for volume {blah} from library manager.
ANR1410W Mount request denied for volume {blah} - mount failed.
ANR1229W Volume {blah} cannot be backed up - volume is offline or access
mode is "unavailable" or "destroyed".
I will do a move data on the volumes in question later in the day, and I
encounter no errors.
I have had IBM hardware look at the 3494 and the library logs show a normal
mount then dismount of the volumes when the errors occur.
Anyone else seeing this happening in a shared library configuration?
-----Original Message-----
From: Magura, Curtis [mailto:curtis.magura AT LMCO DOT COM]
Sent: Monday, March 24, 2003 9:13 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: File scan takes ages during backup
>From the tsmjbbd.ini file where the client is installed:
; NotifyFilter
;
; Specifies what types of filesystem activity to monitor for
; a journaled filesystem.
;
; Multiple activities may be monitored by combining (logical OR)
; values.
;
; The default value is 0x117 (File and Dir name changes,
; last write changes, attribute changes, size changes, and
; security changes).
;
;
; Notification Type Filter Value
;
; File name changes, including create, 0x00000001
; delete and rename
;
; Dir name changes, including create, 0x00000002
; delete, and rename
;
; Attribute changes 0x00000004
;
; Size changes, notification is deferred 0x00000008
; until cache is flushed
;
; Last written time changes, notification 0x00000010
; is defered until cache is flushed
;
; Last access time changes 0x00000020
;
; Creation time changes 0x00000040
;
; Security changes 0x00000100
Curt Magura
Lockheed Martin EIS
Orlando Fla.
321-235-1203
-----Original Message-----
From: Richard Foster [mailto:Richard.Foster AT HYDRO DOT COM]
Sent: Monday, March 24, 2003 10:03 AM
To: ADSM-L AT VM.MARIST DOT EDU
Subject: Re: File scan takes ages during backup
Al
You asked Geoff, but I'll reply :)
> Did IBM say why the notifyfilter parameter was significant? I'm
> curious what that controls.
If I remember correctly, IBM said that it controls exactly which events get
notified to the journal service. If your journal buffer is filling up, you
want to cut out unnecessary events. But I can't remember where they define
which bit corresponds to which event.
Richard Foster
Norsk Hydro asa
|