Hi All
We are running CMOD 9.0.0.6 with DB2 9.7 and TSM 6.3.5 all hosted in same AIX server.
ARSSOCKD process went down without any error message. DB2 and TSM had no issue during the time. After checking the AIX alog, we noticed that below error message thrown from TSM just before the crash. There is no error in syslog also.
0 Wed Jun 8 15:27:03 SGT 2016 arssockd (ARCHIVE): 2016-06-08 15:27:03.006064 231261 BRCHILW 2 429 ARS0429E TSM Error: ANS1033E (RC-53) An invalid TCP/IP address was specified.
, Return Code=-53, Reason=0, File=arssmsms.cpp, Line=1235 Srvr->a01goddb1a<- 0 Wed Jun 8 15:27:03 SGT 2016
0 Wed Jun 8 15:27:03 SGT 2016 arssockd (ARCHIVE): 2016-06-08 15:27:03.014272 0 ARSMSG 2 146 ARS0146E The communication layer was unable to receive data. The error number is 35
0 Wed Jun 8 15:27:03 SGT 2016 arssockd (ARCHIVE): 2016-06-08 15:27:03.032617 ARSSOCKD 4 ARSI Message Error
Also whether the below ulimit values any reason for the arssockd crash?
time(seconds) unlimited
file(blocks) unlimited
data(kbytes) unlimited
stack(kbytes) 4194304
memory(kbytes) unlimited
coredump(blocks) 0
nofiles(descriptors) 2000
threads(per process) unlimited
processes(per user) unlimited
Anyone faced similar issue before? Kindly help.
On z/OS - and I know that isn't the environment here, but - when ARSSOCKD loses TCP/IP connectivity to DB2 for too long a period of time ARSSOCKD will abend.
What happens is because ARSSOCKD abends people think there's a CMOD problem, but there really isn't. It's a TCP/IP problem and ARSSOCKD is the victime.
Ed
This looks like a configuration error. Check your dsm.sys and dsm.opt files for bad values, and it's always best to put fully qualified hostnames in configuration files.
A fully qualified domain name looks like:
tsm-cmod.dc2.example.com
rather than:
tsm-cmod
-JD.