Security Kernels not responding

Tom_Davidson

Tom_Davidson

VIP Member
We had an incident over the weekend where the security kernels seemed to stop responding, but were not shown as 'Zombie/Stopped'.

The symptoms were reported as UBE's in 'Waiting' status, and I had all kinds of messages in my we logs about 'Security System Failure', just a generic message.

When I investigated I had 6 security kernels running, the logs were normal, just the duplicate key on F9312.

I had to kill all the kernels and when the new ones started, all was well.

I did find this in two JDENET_N logs:

1955/12 MAIN_THREAD Fri Nov 2 17:40:52.479432 netflow.c1032
1955 (pid) in routeOutgoingQueue(), unable to send message: sd=5, msgId=1, msgHost=172.16.12.62, msgPort=6016, sendingHost=172.16.1

The sd= was different but the time was about the same.

When I checked support the above message it seemed to point to RTE, which we don't use.

Any ideas what the issue might have been? I'm loath to open a JDE Ticket as I cycled JDE services on Sunday morning, 'just in case' and finding the 80 kernel logs in my 100K+ logs is too damn hard.

Tom
 
Are you using LDAP integration? What tools release is this client on?
 
Daylight Savings Time issue?

I recall having some issues around this in years past.



We had an incident over the weekend where the security kernels seemed to stop responding, but were not shown as 'Zombie/Stopped'.

The symptoms were reported as UBE's in 'Waiting' status, and I had all kinds of messages in my we logs about 'Security System Failure', just a generic message.

When I investigated I had 6 security kernels running, the logs were normal, just the duplicate key on F9312.

I had to kill all the kernels and when the new ones started, all was well.

I did find this in two JDENET_N logs:

1955/12 MAIN_THREAD Fri Nov 2 17:40:52.479432 netflow.c1032
1955 (pid) in routeOutgoingQueue(), unable to send message: sd=5, msgId=1, msgHost=172.16.12.62, msgPort=6016, sendingHost=172.16.1

The sd= was different but the time was about the same.

When I checked support the above message it seemed to point to RTE, which we don't use.

Any ideas what the issue might have been? I'm loath to open a JDE Ticket as I cycled JDE services on Sunday morning, 'just in case' and finding the 80 kernel logs in my 100K+ logs is too damn hard.

Tom
 
Well Duh! I forgot I had multiples in my sig, sorry. This is 9.1 Tools 9.1.5.3. I am not using LDAP.

We use GMT, so don't have to deal with time changes, one of the few good things about a global install
 
Back
Top