Strange messages in SYSLOG.LOG

[aaaA_4]

We have MC/SG (11.16) used in our environment. From certain time strange messages in SYSLOG.LOG have started to appear:

May 25 08:25:45 XXXXX CM-CMD[14945]: /usr/sbin/cmgetconf -p XXXXX

These messages are related to all of our packages in MC/SG cluster and they appear regularly, with 10 sec. time period.

Any help will be appreciated.
N.

[Rajesh SB]

Hi,

cmgetconf -p means getting cluster package configuration information.
To generate the package ASCII configuration file for pkg to see, this command can be used.

You need to investigate more from where it is executing. Whether User initiated or scheduled.

Regards,
Rajesh SB

[Chauhan Amit]

Hello,

Found some docs that talk about the same problem , message in the syslog of cmgetconf -p . It's linked to LEGATO backup in SG environemt .
How do you take ur backup's ??

-Amit

[Darrel Louis]

Hi,

Can you check if you've a cron job running?

Check the following:
http://forums1.itrc.hp.com/service/forums/bizsupport/questionanswer.do?threadId=109647

Darrel

[Stephen Doud]

There are several possible sources for the cmgetconf command. This document gives a list of some:
UMCSGKBRC00014425
TITLE: cmclconfd consumes a lot of CPU time


This document provides a script to determine the source of the cmgetconf commands:

UMCSGKBRC00014697
TITLE: Tracing the source of cmgetconf commands

[aaaA_4]

Hi Stephen,

can you provide me with workarounds (if they exist) from these documents, because I do not have an access to KB.

Thank you,
N.

[aaaA_4]

Hi Chauhan,

we do not use Legato.

N.

[Chauhan Amit]

Check out this link:

http://forums1.itrc.hp.com/service/forums/bizsupport/questionanswer.do?threadId=1020285

-Amit

[Chauhan Amit]

UMCSGKBRC00014697

One method to find the source of repeated cmgetconf requests is to substitute
the cmgetconf command with a wrapper script which traces and reports the
caller of cmgetconf back to it's source, then runs the real cmgetconf.

The following command shows that cmgetconf is linked to other Serviceguard
commands:

# ll -i /usr/sbin/cm* | sort -n

This is why cmgetconf must be moved versus copied - in the following
instructions... the linkage maintains the functionality of the command.

The implement the script below:
1) mv /usr/sbin/cmgetconf to /root/bin/
(mkdir /root/bin/ if it does not exist)
2) Copy the script below to /usr/sbin/cmgetconf and give it the same
permissions as the real cmgetconf: -r-xr-xr-x

Since the multiple cmgetconf's can affect performance, the script has the
ability to send an email to immediately page/inform the designated recipient.
To use it, edit the MAILADDR definition line, removing the # symbol and
subsequent text and substituting a valid email address. NOTE that the mail
system must be working on the node in order for the message to be deliver

-Amit

* Attach Find the script . The script is not supported by HP .

[Mustafa Gulercan]

you can obtain a copy of the package's
ASCII configuration file by using the cmgetconf command,
specifying the package name.
# cmgetconf -p pkg1 pkg1.ascii

so if you see this log regularly you should check crontab.

regards;
mustafa

[Enrico P.]

Hi,
perhaps there is a monitoring of your system and the OVO agent is installed on your system.
So this message is normal.

Enrico

[aaaA_4]

Hi Enrico,

I guess, you are right. I'll check it.

Thank you,
N.