Why is cmclconf -p running??

[Dwyane Everts_1]

All,

I have a node (RP7410, 11.11) that does not have MC/SG configured or running; yet cmsnmpd and comclconf -p are running (and respawned if killed). And I'm getting syslog entries. Does anyone know why these might be running, and even better...how do I diable them from restarting?

Dwyane

[Dwyane Everts_1]

Sorry for the speeling error....

disable, not diable

D

[Dwyane Everts_1]

muuuussssstttt hhhhaaaavvvveeee mmmooorrreeee ccccooooffffffeeeeee....

spelling, not speeling

geez!!! and when I graduate preschool I will know how to spell these big words!!!!

D

[Stephen Doud · Accepted solution]

cmgetconf is spawned by other applications including:

1) Ignite/UX - make_recovery gathers cluster information

2) Legato (Networker) backup software running on the node.
- Legato modifies /etc/syslogd.conf, directing messages to /nsr/logs and
syslog.log. Check these files for coincidental Legato activity

- Practically every Legato command performs cmgetconf (commands are
located in /opt/networker/bin).

3) cron
- type crontab -l | grep
investigate each entry to exclude it as a possible contributor

4) Serviceguard Manager
- Serviceguard manager does not run cmgetconf, but it does awaken
cmclconfd.

- /var/adm/syslog/syslog.log registers activity on cmomd -
the daemon that services Serviceguard Manager connection requests.

- /var/opt/cmom/cmomd.log also captures activity requests.
Check both logs to see if cmomd was active at the time of the avalance
of cmgetconf commands.

5) Predictive HA configuration (with HA configuration enabled via
psconfig.

Predictive operates the '/opt/pred/bin/psmond' daemon constantly and
usually wakes up in the middle of the night to do diagnostic log
harvesting.

- Link to Predictive Information

- Use "ps -ef | grep psmond" to see if it's running on the server
and insure the latest Predictive patch is installed.


6) HA Tracker tool (used with CSS support and an HAO account)
Systems on a BCS or CSS software support contract are loaded with HP tools
which help to increase the support community awareness of the system's
state. Unfortunately one of those tools - HATracker, might run away.
PHSS_21271 was created to resolve this.

7) ISEE tool. ISEE replaces the HAO utility. ISEE may perform a
MAP (Motive Assistant Programs) which are diagnostic data gathering
applications. ISEE can use runner.sh, cmviewcl.sh or
cmviewconf.sh to exercise cmgetconf or other cluster commands to
gather information.

8) In one case, ovconfd has been found to run cmgetconf every 10 seconds.
See document QXCR1000193632

9) Some other program or from a script
- Requires further investigation be the system administrator.

10) It could also be run from the package control script.

11) A System Admin executed the command.

The only way to tell which if these is the culprit is by tracing the cmgetconf PID's process parentage back to the originating command.
Depending on the active duration of the cmgetconf command, it may not be possible to trace the command quickly enough. If this is the case, at least review the 'ps' list to see if any of the above utilities is operating. If that fails, open a case with the response center or research this document: UMCSGKBRC00014697
TITLE: Tracing the source of cmgetconf commands

-StephenDoud

[Dwyane Everts_1]

Stephen,

I guess since we use Ignite/UX and ISEE, they are the probable culprits!! HA Tracker is also loaded from a previous HAO install.

Thanks for the insight!!!!
Dwyane