ADSM-L

[ADSM-L] Retry after 1855?

2009-02-10 12:34:19
Subject: [ADSM-L] Retry after 1855?
From: "Conway, Timothy" <Timothy.Conway AT JBSSA DOT COM>
To: ADSM-L AT VM.MARIST DOT EDU
Date: Tue, 10 Feb 2009 10:32:06 -0700
(or 4994, or 0342, depending on when and where you look at it).
I have a lot of data on the far end of a WAN link.  I thought the WAN
was really error-prone, but after I presented my evidence that nearly 1
in 100,000 packets arrives with the wrong checksum and was laughed at,
I've learned that that's a really good error rate.
However, as most of you are aware, 
 
 
With the BA client:
server activity log -
02/08/09   19:44:35     ANR1855E Failure validating protocol between
client and server for node SAPXIDEV on session 29795 (SESSION: 29795)
client dsmsched.log -
02/08/09   19:44:37 Normal File-->       583,009,280
/oracle/XD1/mirrlogB/log_g16m2.dbf  ** Unsuccessful **
02/08/09   19:47:34 Retry # 1  Normal File-->       583,009,280
/oracle/XD1/mirrlogB/log_g16m2.dbf [Sent]

 
With TDPO:
server activity log -
02/08/09   02:38:03     ANR1855E Failure validating protocol between
client and server for node HPRPDB02_RMAN on session 21944 (SESSION:
21944)

cleint 
02/08/09   02:38:03 ANS0342E The CRC received from the Server does not
match the CRC calculated by the client.
02/08/09   02:38:03 ANS0342E The CRC received from the Server does not
match the CRC calculated by the client.
02/08/09   02:38:03 ANS0342E The CRC received from the Server does not
match the CRC calculated by the client.
02/08/09   02:38:03 ANS0342E The CRC received from the Server does not
match the CRC calculated by the client.
02/08/09   02:38:03 ANS4994S TDP Oracle HP ANU0599 TDP for Oracle:
(18235): => (hprpdb02_rman) ANS0342E (RC236)  The CRC received from the
Server does not match th
e CRC calculated by the client.

(I don't know why it reports one CRC error 4 times)...and the backup has
failed.  To complete the backup, it must be started over from the
beginning.  I've had as many as 4 attempts fail in a row.
 
There doesn't seem to be any provision to force a retry of the
particular chunk, and I think the way RMAN uses TDP(almost like a tape
drive), that's not a relevant concept.
 
 
What do the rest of you do for your TDPO over WAN - besides put a TSM
server local to the clients?
 
 
73,
 
Tim Conway
Unix sysadmin, TSM admin
Swift & Company
9705067998
timothy.conway AT jbssa DOT com
 

<Prev in Thread] Current Thread [Next in Thread>