Issue with ttcserver consuming 25% of a CPU - TimesTen Release 6.0.6
Hi !
I'm running TimesTen Release 6.0.6 on a Netra 440 with 4 CPU (SunOS 5.10)
We have a client/server architecture working decently, but some times the ttcserver seems to start consuming a lot of CPU:
PID USERNAME SIZE RSS STATE PRI NICE TIME CPU PROCESS/NLWP
29568 root 139M 138M cpu1 50 0 147:00:54 25% ttcserver/4
29916 root 1630M 1623M sleep 59 0 5:11:12 0.7% ttcserver/1
29087 root 301M 72M sleep 59 -8 4:37:16 0.6% gdproc/2
I understand this server is the one responsible of executing the DB actions on behalf of the clients.
Running ttstatus I can see the following:
TimesTen status report as of Tue Jul 20 13:17:27 2010
Daemon pid 29561 port 16001 instance tt6_0_64bit
TimesTen server pid 29568 started on port 14502
No TimesTen webserver running
------------------------------------------------------------------------
Data store /var/TimesTen/tt6_0_64bit/DemoDataStore/TpcbData
There are no connections to the data store
Replication policy : Manual
Cache agent policy : Manual
------------------------------------------------------------------------
Data store /timestendata/tt41data
There are 19 connections to the data store
Data store is in shared mode
Shared Memory KEY 0x420268fa ID 469762084 (ISM LOCKED)
Type PID Context Connection Name ConnID
Process 1573 0x0000000100165bb0 SMI 11
Process 1584 0x0000000100165bb0 SMI 12
Process 1593 0x0000000100165bb0 SMI 13
Process 1604 0x0000000100165bb0 SMI 14
Process 1605 0x0000000100165bb0 SMI 15
Process 2228 0x0000000100165bb0 TTOAMSERVER 10
Process 29270 0x000000010036fd50 DBCN64 8
Process 29270 0x0000000100376370 DBCN64 9
Process 29598 0x0000000100165bb0 ttMonitor 1
Process 29916 0x0000000100165bb0 TTOAMSERVER 2
Replication 29985 0x0000000100232670 LOGFORCE 5
Replication 29985 0x000000010024b7a0 TRANSMITTER 6
Replication 29985 0x000000010025b7c0 REPHOLD 4
Replication 29985 0x0000000100282e80 REPLISTENER 3
Replication 29985 0x00000001002ba970 RECEIVER 7
Subdaemon 29566 0x0000000100178f80 Worker 2044
Subdaemon 29566 0x00000001001d5ef0 Flusher 2045
Subdaemon 29566 0x00000001001dc510 Checkpoint 2047
Subdaemon 29566 0x0000000100224cb0 Monitor 2046
RAM residence policy: When In Use and 60 Seconds Longer
Replication policy : Manual
Replication agent is running.
Cache agent policy : Manual
------------------------------------------------------------------------
End of report
My question is if there's a way to see/trace what the ttcserver is actually doing to identify if there's an offending client causing this CPU overuse.
Thanks,
Mike