cmcld: Stopped accepting local connection requests

[W.L. Alsemgeest]

In my syslog i found the following entry.
Can someone help me to explain what this means?

==========================================
Sep 24 09:35:12 prc02b03 cmcld: Stopped accepting local connection requests because there are currently too many concurrent connections to th
e daemon (48).
Sep 24 09:35:12 prc02b03 cmcld: There are no longer too many local connections. Now accepting local connection requests.
==================================================

[Santosh Nair_1]

Looks like there might be a patch for this issue, PHSS_23660. You didn't mention which version of ServiceGuard you're running nor the OS level.

-Santosh

[Pedro Sousa]

Hi W.L.!
from another thread, Carsten Krege wrote:

"The message "cmcld: Stopped accepting local connection" is logged since SG's main process cmcld limits the number of connections to the hacl-local port that is used by all SG commands (e.g. cmviewcl, cmquerycl, etc.).

This message appears when the limit is reached (23 in your case). The limit depends from some dynamic parameters one have hardly influence on (e.g. number of dependent EMS resources configured, number of nodes in the cluster). Also the value of MAX_CONFIGURED_PACKAGES is used for the calculation. If MAX_CONFIGURED_PACKAGES is increased by 1, the max number of local connections to hacl-local is also increased by 1. To increase MAX_CONFIGURED_PACKAGES is an easy way to increase the maximum number of local connections that are accepted by cmcld.

However, doing this would also increase the amount of memory used by cmcld and is therefore not really recommended.

The normal cause for this message is an extensive use of SG commands (mostly cmviewcl running in a loop to retrieve status information). It is recommended not to use SG commands in a loop (insert a sleep command to lower the rate of SG command invocation)."

good luck!

[W.L. Alsemgeest]

Dear Santosh Nair,

We have installed MC/SG A.11.09 on UX11 patch level: 49.3.

[W.L. Alsemgeest]

Dear P.S.
Thanks for your reply but our MAX_CONFIGURED_PACKAGES 30
and thats the max value for this parameter. The cm commands we have issued were limited and there where no loops running at the time.

[Santosh Nair_1]

The only time I've see reference to this error was when there were too many SG commands being issued on the machine at the same time. Most likely what Pedro was saying, i.e. running cmviewcl in a loop without any sleeps.

When are you seeing these errors? Is it during cluster formation, package start up or just during the normal course of the day?

-Santosh

[Sanjay_6]

Hi,

Have a look at the thread below.

http://us-support.external.hp.com/cki/bin/doc.pl/sid=ca8eaa3e172d9dc340/screen=ckiDisplayDocument?docId=200000048917277

Hope this help. Also do you have the latest HP recommended SG patch on the system.

thanks

[W.L. Alsemgeest]

Thank You all.
We will look if we can reduce the SG request's (don't know how??)

Greetings,
Wim Alsemgeest