I'm trying to isolate an issue with slow backup performance with HP
DataProtector 6.10 on NetWare (SP7, with latest SMS patch). As part of
this I've turned on sms debugging (tsafs /smsdebug=fffffffc
/smsdebug2=fffffffc) to try and figure out if this is a TSA problem or a
DP problem.

Unfortunately, I can't quite figure it out. After a lot of poking, I've
finally managed to isolate where in the log file the problem happens. The
problem is a very long pause between data bursts.

I'm hoping someone here knows what the format of the TSADBG.LOG file well
enough to give me some hints as to which module is pausing for so long.

a84c7620[TSA READ: 6D1466A0 : 2 ]ca0 [815 ]:Read Thread 2 : dSetID
a : cCode : 0
a84c7620[TSA READ: 6D1466A0 : 2 ]00037e60: Start
ad00b160[TSA READ: 6D1466A0 : 1 ]ca0 [815 ]:Read Thread 1 : dSetID
a : cCode : 0
ad00b160[TSA READ: 6D1466A0 : 1 ]00037e60: Start
**********PAUSE*************
674aa460[HPVBDA_P 1 ]00000490: Start
674aa460[HPVBDA_P 1 ]490 [314 ]:IN: conn=1,
handle=8001000a, toRead=fec8
674aa460[HPVBDA_P 1 ]490 [385 ]:dSet : 6733cc44 : ID :
10 : To Read : 65224
674aa460[HPVBDA_P 1 ]490 [385 ]:dSet : 6733cc44 : ID :
10 : To Read : 61179
674aa460[HPVBDA_P 1 ]490 [487 ]:Read : 65224 : ERROR :
0 : state = 1062
674aa460[HPVBDA_P 1 ]490 [502 ]:OUT: Read=fec8
ad00b160[TSA READ: 6D1466A0 : 1 ]ca0 [736 ]:Processing : dSetID : a
: readIndex : 0 : readFlags : 1062 : readOffset : 777453568

The overall problem is that one of my two DataProtector backups are
frequently slow. The first backup is 1.5TB of geospatial data, and has
never had any performance problems. The second backup is 450GB of regular
office files (just under 1 million files in that 450GB), and the full
backups run to completion in only 4 hours just 30% of the time. The other
70% of the time they'll take 4 days or longer to complete.

It isn't an open-file problem that I can tell. The files opened by the
backup connection have never overlapped with any other open files on the
volume.

--
Novell, it does a network good