I'm going to make a wild guess that the "CONVERSION STATE" has to do with
the node using the "archive description" table.
Try running a "clean archdir <nodename> showstats" to see what the
"conversion state" of the node is and that it actually reports either
"true" or "false".
I've been seeing odd errors with clients that initiate large numbers of
concurrent archives and I see random initial errors, that hint of lock
contention, followed by normal operation. We would see ANR0530W's on the
server.
ANR0530W Transaction failed for session session number for node node name
(client platform) - internal server error detected.
Explanation: The server ends a database update transaction for the
specified
session because an internal logic error is detected.
System Action: The server ends the specified session and server operation
continues.
We turned up this
= IC39278: ITSM Server issues ANR0530W when encountering a lock without
indicating the cause of the failure.
Affects TSM servers 5.1 and 5.2 releases, all platforms.
RECOMMENDATION: Apply fixing ptf when available. This
* problem is currently projected to be fixed
* in PTF level 5.2.3. Note that this is
* subject to change at the discretion of IBM.
Dwight E. Cook
Systems Management Integration Professional, Advanced
Integrated Storage Management
TSM Administration
(918) 925-8045
Sean M
English/Charlotte To: [EMAIL PROTECTED]
/[EMAIL PROTECTED] cc:
Sent by: "ADSM: Subject: Strange Error on TDP for
Oracle
Dist Stor
Manager"
<[EMAIL PROTECTED]
.EDU>
03/13/2004 01:51
AM
Please respond to
"ADSM: Dist Stor
Manager"
*SMers,
I wanted to find out if anyone else has encountered this error before when
running TDP for Oracle over LAN Free.
03/13/04 02:04:30 ANR9999D (Session: 3374, Origin: MAPQA1_AGNT)
smexec.c(8722): ThreadId<12> Unexpected error
(14)
obtaining the node conversion state for Session
8.
Callchain of previous message: 0x10016288
outDiagf <-
0x1040B348 CheckClientConversion <- 0x1040EA2C
HandleNodeSession <- 0x10414E74 smExecuteSession
<-
0x1051D3D8 SessionThread <- 0x10007C38
StartThread <-
0xD004B57C _pthread_body <- (SESSION: 3374)
03/13/04 02:04:30 ANR0435W (Session: 3374, Origin: MAPQA1_AGNT)
Session 8
for node MAPQA1_ORA (TDP Oracle AIX) refused -
internal
error detected. (SESSION: 3374)
When the backup sessions start, these errors are received. However, once
these errors are received, they go away and the backup continues and runs
to completion. My fear is that I might be missing some data from getting
backed up, due to this error. Here is my setup:
Client:
OS: AIX 5.2.0.0
TSM Client: 5.2.2.0
TDP for Oracle: 5.2.0.0
Storage Agent: 5.2.2.1
Atape: 8.3.6.0
TSM Server:
OS: AIX 5.1.0.0
TSM Server: 5.2.2.1
Atape: 8.3.6.0
I could just be paranoid :-). However, any help or feedback would be
greatly appreciated. Thanks in advance.
Regards,
Sean
<<inline: ecblank.gif>>
<<inline: graycol.gif>>
<<inline: pic00285.gif>>
