cmgetconf and cmcheckconf -timed out while query vg

[johnsonpk]

Hi ,
While executing the cmgetconf ,I am geting the following message
Timed out trying to query the following volume group(s) on node server1
/dev/vg00

Timed out trying to query the following volume group(s) on node server1
/dev/vg00
-
Timed out trying to query the following volume group(s) on node server1
/dev/vg00

Timed out trying to query the following volume group(s) on node server1
/dev/vg00
70%


And i tried to execute cmcheckconf -C cluster.ascii
i am getting the same error related to the cluster lock vg

Analysis of 304 devices should take approximately 4 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%----100%
Found 20 volume groups on node server1
Found 20 volume groups on node server2
Analysis of 40 volume groups should take approximately 1 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%---
Timed out trying to query the following volume group(s) on node freddy
/dev/lockvg01
-100%
.....
Gathering Network Configuration ........ Done


Can any one shed some light please...

Thanks Much
John

[Avinash Agarkar]

Hi john,

Have you made all the vg's cluster aware with

vgchange -c vgname

thanks,
Avinash

[Mridul Shrivastava]

Please provide the details about HPUX and MCSG version please.

[Steven E. Protter]

Shalom John,

In HP-UX 11.11 or above with SG 11.16 or above this normally occurs because of the following causes:

1) HA daemons in inetd.conf are not correctly configured according to documentation.
2) Node security is not updated to use cmnodelist functionality.

Check that and answer the other questions to get further assistance.

SEP

[Stephen Doud]

Check the the 'identd' (or 'auth') line in /etc/inetd.conf. It may be commented out, on one or more of the servers. This is a required function for Serviceguard.

[johnsonpk]

Hi All,
I apologize for the late reply..
I have bring down the cluster and did cmcheckconf /cmapplyconf and it went through successfully :)

but still a would like to know the reason behind this ..

Avinash:-- The vg00 is not part of cluster,and other Vgs are part of running cluster.

Mridul:_Hp-UX 11.11 $ SG 11.16


Steven/Stephen: i will get back to you...

[johnsonpk]

Stephen: The identd entry is not there in both nodes ..I tried to run the cmgetconf command on both node and all the time i am getting error related to the first node(ie node1)


Steven: Could you please provide me a link to the documentation (For configuring HA daemons & Node security)


Appreciate Your help.

Thanks!!
Johnson

[Stephen Doud · Accepted solution]

The Managing Serviceguard manual contains information on how to configure for identd.

Edition 15 is located at http://docs.hp.com/en/B3936-90135/B3936-90135.pdf

Page 200 starts the dicussion - configuring root-level access.

Here is what you should see in servers installed with HP-UX 11.23 and later:

# grep auth /etc/services /etc/inetd.conf
/etc/services:auth 113/tcp authentication # Authentication Service
/etc/inetd.conf:auth stream tcp6 wait root /usr/lbin/identd identd

[johnsonpk]

thanks