Error 1006

GregTomkins

Active Member
10.2B Linux (WebSpeed) HPUX (db)

We have been having this problem off and on for the past few days on a system that's been stable for months. This is a remote WebSpeed agent. There are other non-WebSpeed AppServers attached to the same DB (running on a different box) that run fine, as does the DB itself (eg. no errors in the .lg file).

We update 4GL on this system regularly, but there are no obviously correlated changes here (GetSecurityPeriodInterestRates hasn't changed in >10 years) and in any case, I am not aware of any 4GL actions that would cause an remote agent to think it's DB is being shutdown when it's not.

There is a long trail of routine events prior to the problem at 7:36:05.600. This problem is not resolved by restarting WebSpeed however it is resolved by restarting all attached DB's. AFAIK the network between these machines is stable. We use VM's all over, if that matters.

Any advice?

[14/12/08@07:36:04.938-0800] P-008838 T-000000 1 WS -- (Procedure: 'gnpr/gateway.p' Line:7788) Gateway begins for: dslpgra
[14/12/08@07:36:04.953-0800] P-008838 T-000000 1 WS -- (Procedure: 'mfai/mfgapix_gs.p' Line:34) get_proc_sig_w.i 010200
[14/12/08@07:36:05.148-0800] P-008838 T-000000 1 WS -- (Procedure: 'iqai/iqgpsdx.p' Line:26116) iqgpsdx: IAWD000A ACCOUNT 12/08/14 HOUSE TD yes ALL no yes NONE yes yes
[14/12/08@07:36:05.600-0800] P-008838 T-000000 1 WS -- (Procedure: 'GetSecurityPeriodInterestRates iqai/iqgpsdx.p' Line:7905) Disconnect from server; database is being shutdown. (2659)
[14/12/08@07:36:05.605-0800] P-008838 T-000000 2 WS CONN Disconnecting from database exc2sm, user number 303. (9545)
[14/12/08@07:37:06.803-0800] P-008838 T-000000 1 WS -- (Procedure: 'iapr/ialiwdx3.p' Line:891) iapr/ialiwrx.p Database exc2sm not connected. (1006)
 

TheMadDBA

Active Member
Weird that there aren't any errors in the DB log file around that time. Can you find a stack trace (protrace) for that process?

There are few issues in the KB about temporary file locations or CONTAINS crashing and giving this error. Usually the interesting parts are going to be in the protrace and the database log.
 

Cringer

ProgressTalk.com Moderator
Staff member
First port of call would be to see what changed on the servers and the network at the time the errors started presenting themselves. Tom is right to suspect AV. Maybe also Firewall.
 
Top