Hi,

I just started a restore on one of our servers -- anyone have any idea what the 'Severe' error below means ?

2011-04-20-15.04.31.920603-300 I7831335E377 LEVEL: Warning
PID : 27605 TID : 139938424375040PROC : db2agent (DBNAME)
INSTANCE: db2inst1 NODE : 000 DB : DBNAME
APPHDL : 0-7 APPID: *LOCAL.nrg1.110420200431
FUNCTION: DB2 UDB, database utilities, sqludmhv, probe:110
MESSAGE : Performing a Cross Platform Restore

2011-04-20-15.04.31.920766-300 I7831713E420 LEVEL: Warning
PID : 27605 TID : 139938424375040PROC : db2agent (DBNAME)
INSTANCE: db2inst1 NODE : 000 DB : DBNAME
APPHDL : 0-7 APPID: *LOCAL.nrg1.110420200431
FUNCTION: DB2 UDB, database utilities, sqludPrintStartingMsg, probe:1
DATA #1 : String, 51 bytes
Starting a full database restore.
Agent PID: 27605

2011-04-20-15.04.31.920845-300 I7832134E449 LEVEL: Severe
PID : 27605 TID : 139938424375040PROC : db2agent (DBNAME)
INSTANCE: db2inst1 NODE : 000 DB : DBNAME
APPHDL : 0-7 APPID: *LOCAL.nrg1.110420200431
FUNCTION: DB2 UDB, database utilities, sqluCompareDB, probe:40
MESSAGE : DiagData
DATA #1 : Hexdump, 4 bytes
0x0000000220BE0818 : DB09 0000 ....

It's not really severe, just a warning.

0x0000000220BE0818 : DB09 0000

DB09 -> 09DB (after byte reversal). This value is in hex = 2523 in dec. So, this is sql2523w.

thanks, any idea why I would have this issue too? Am so confused and baffled by this one!

Restore of backup image hanging

-1119 should not prevent restore from completing. If you're going to restore again, drop the db and delete/rename all leftover files belonging to this db and then redo it again. Then, monitor its progress using db2 list utilities show detail.

thanks a bunch.. will try it out! this problem has been driving me a little nuts today! Am just finding it a little odd because I restored the same backup image on a 32 bit installation ( on another server) and it worked just fine and was done on 1.5 hours.. not sure why restoring on a 64 bit would complain so much... seems like a system error to me...

so i just gave the command 'db2 list utilities show detail' and the completed work bytes = equal to the size of the backup image so I guess it's done restoring.. but it hasn't come to the prompt yet and I still see the 'failed to connect to database' errors in db2diag.log... any idea why this would be happening.. seems like it's stuck or something.. thanks!!!

Can you paste the output from db2 list utilities show detail and db2 list applications show detail? Do you see any logs in the logtarget location?

db2 list utilities show detail

ID = 1
Type = RESTORE
Database Name = DBNAME
Partition Number = 0
Description = db
Start Time = 04/20/2011 18:06:29.916586
Progress Monitoring:
Completed Work = 28556414976 bytes
Start Time = 04/20/2011 18:06:29.916596



CONNECT Auth Id Application Name Appl. Application Id Seq# Number of Coordinating DB Coordinator Status Status Change Time DB Name DB Path
Handle Agents partition number pid/thread
------------------------------ -------------------- ---------- ------------------------------ ---- ---------- ---------------- --------------- ------------------------------ -------------------------- -------- --------------------
NRG1 db2bp 12 *LOCAL.nrg1.110420230631 0001 1 0 742 Restoring Database Not Collected DBNAME /home/db2inst1/db2inst1/NODE0000/SQL00001/

And yes, I see the log files on the log target folder.. the timestamp on the log files is Apr 20 20:49

Possibly it got hung. Do you still see db2med/db2bm processes?

Try to do manual cleanup (db2stop + ipclean + kill all remaining processes owned by this instance + ipcrm all IPC resources owned by this instance). Then, restore the db without logtarget option.

thanks, if i give ps aux | grep db2, I don't see any db2med but see these db2bm processes --

db2inst1 750 5.2 0.2 829748 75140 ? S 18:06 14:29 db2bm.742.0
db2inst1 751 5.1 0.2 827676 71532 ? S 18:06 14:20 db2bm.742.1



also, earlier today i gave the ipclean command only and then tried the restore again.. still got the same issue.

I thought ipclean would kill all the processes owned by the instance? For ipclean, I go to /home/db2inst1/sqllib/bin and give ./ipclean db2inst1

how can I kill the other processes without ruining anything? also, couldn't find ipcrm under sqllib/bin.. is it somewhere else? what is it for?

thanks, you're the best!

do you think it would make sense to try a restore from an uncompressed image? is it possible that the COMPRLIB option is causing this problem?