Has anyone had problems with a large restore MVS to Sun? Did you find a
solution?
Server: OS 390 2.10 / TSM for MVS ver 4 rel 1 lev 3.0
Client: SunOS 5.8 / TSM ver 4 rel 1 lev 2.0
We back up a large ".tar" file nightly (12 Gbyte) and keep 3 copies.
The backups appear successful, but when we try to restore we get:
Sun side:
07/30/01 15:01:58 ANS1810E TSM session has been reestablished.
07/30/01 15:08:40 sessRecvVerb(): Invalid verb received.
07/30/01 15:08:40 sessRecvVerb(): length=750a, verb=41,magic=90
07/30/01 15:08:40 TcpFlush: Error 9 sending data on Tcp/Ip socket
4294967295.
07/30/01 15:08:40 sessRecvVerb: Error -50 from call to 'readRtn'.
07/30/01 15:08:40 ANS1809E Session is lost; initializing session reopen
procedure.
07/30/01 15:08:55 ANS1810E TSM session has been reestablished.
MVS side:
14.56.56 STC06166 ANR0406I Session 1712 started for node PRDRMS01 (SUN
SOLARIS) (BPX-Tcp/Ip
14.56.56 STC06166 172.20.1.19(-28694)).
... and around 15:08:40 ...
data movement messages... nothing about the Sun session.
The restore gets between 1 Gbyte to 5 Gbyts to the client, then the
connection
ends and the restore retries a few minutes later. We've tried different
versions
of the file and restoring to a different machine with the same results.
I've watched the connection on the MVS side and it disappears (no TimeWait,
no ClosWait, etc.). There are no TCP/IP errors on the MVS side and I
haven't
found any TCP/IP errors on the Sun side (where to look?).
The last time this file was successfully restored was 5/27/01. The only
major change
has been the switch from OS390 2.6 to OS390 2.10. Nothing on IBM Link, etc.
Riddler, Garth (HAL).vcf
