Problems with cluster (Failed to load data from configuration database)

[Carme Torca]

Hi,

I have two servers in cluster with 11.23.
Now (I don't know why), one node shows me:

#cmviewcl
Failed to load data from configuration database.
Unable to perform the security token exchange with cmclconfd on node hpit12
Unable to perform the security token exchange with cmclconfd on node hpit11
Failed to refresh the cluster handle.
Unable to connect to configuration database.


But the other one, works ok.

Could anyone help me?
Thanks a lot of!

Carmen.

[Sandeep_Chaudhary]

Please read this LINK:

http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=309926

I think regeneration cmclconfig on othernode will solve this.

[Carme Torca]

Hi,

I have do it. Its runs ok, but the problem in the other node its again.

"1. run cmgetconf and get the present cluster configuration.
#cmgetconf -v -c outputfile
2. If it doesn't have the alternate node, edit the file and put that node too in the outputfile
3. Then cmcheckconf -v -C outputfile
4. cmapplyconf -v -C outputfile"



[node11]#cmapplyconf -v -C /tmp/out.txt
Checking cluster file: /tmp/out.txt
Checking nodes ... Done
Checking existing configuration ... Done
Gathering storage information
Found 34 devices on node node11
Found 58 devices on node node12
Analysis of 92 devices should take approximately 7 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%----100%
Found 9 volume groups on node node11
Found 10 volume groups on node node12
Analysis of 19 volume groups should take approximately 1 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%----100%
Volume group /dev/vgajdoc_int is configured differently on node node11 than on node node12
Volume group /dev/vgajdoc_int is configured differently on node node12 than on node node11
Volume group /dev/vgajdat_int is configured differently on node node11 than on node node12
Volume group /dev/vgajdat_int is configured differently on node node12 than on node node11
Volume group /dev/vgtemisdat_int is configured differently on node node11 than on node node12
Volume group /dev/vgtemisdat_int is configured differently on node node12 than on node node11
Volume group /dev/vgscdoc is configured differently on node node11 than on node node12
Volume group /dev/vgscdoc is configured differently on node node12 than on node node11
Volume group /dev/vgjovo is configured differently on node node11 than on node node12
Volume group /dev/vgjovo is configured differently on node node12 than on node node11
Gathering network information
Beginning network probing (this may take a while)
Completed network probing
Checking for inconsistencies
Modifying configuration on node node11
Modifying configuration on node node12

Modify the cluster configuration ([y]/n)? y
Marking/unmarking volume groups for use in the cluster
Completed the cluster creation

[Sandeep_Chaudhary]

Next step:

if inetd problem
http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1021331

if u r using bastile
http://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1155058

https://forums11.itrc.hp.com/service/forums/questionanswer.do?threadId=1123114

[Carme Torca]

Thanks, but What have I do?
The cluster now its working ok, but I can not see it in node12.

[node11]#cmviewcl

CLUSTER STATUS
cl_integracio up

NODE STATUS STATE
node11 up running

PACKAGE STATUS STATE AUTO_RUN NODE
pkg1 up running enabled node11
pkg2 up running enabled node11
pkg3 up running enabled node11
pkg4 up running enabled node11

NODE STATUS STATE
node12 up running

PACKAGE STATUS STATE AUTO_RUN NODE
pkg5 up running enabled node12
pkg6 up running enabled node12
pkg7 up running enabled node12
pkg8 up running enabled node12
pkg9 up running enabled node12
[node11]#

[Sandeep_Chaudhary]

can u come again. I havent understand what u want to say. wher is cmviewcl from node12?

[Carme Torca]

Hi,

Yes.
In one node the cluster shows ok all, but in
the other node shows me:

[node12]#cmviewcl
Failed to load data from configuration database.
Unable to perform the security token exchange with cmclconfd on node node12
Unable to perform the security token exchange with cmclconfd on node node11
Failed to refresh the cluster handle.
Unable to connect to configuration database.
[node12]#

[Avinash20]

Please check the following link and check on both the nodes

http://bizsupport.austin.hp.com/bizsupport/TechSupport/Document.jsp?lang=en&cc=us&taskId=110&prodSeriesId=3203314&prodTypeId=18964&prodSeriesId=3203314&objectID=c01368914

[Prashanth.D.S · Accepted solution]

Hi Carme,

These errors could be due to multiple reasons..

CAUSE 1:


"auth" line commented out in /etc/inetd.conf
#auth stream tcp6 wait bin /usr/lbin/identd identd


Serviceguard uses identd to validate Serviceguard commands are being performed by nodes in the cluster. If this line is disabled, Serviceguard commands will fail in various ways; one being the "security token exchange' error.

Cause 2 :

Check the /etc/nsswitch.conf file check for

ipnodes: files



This is an essential line, and adding it corrected the problem. Due to ServiceGuard's need for local hostname lookup, it is recommended that /etc/nsswitch.files be copied to /etc/nsswitch.conf as a starting configuration for hostname resolution.


Cause 3 :

Run the following command on both machines, patch level need to be the same on both nodes..

# what /usr/lbin/cmcld

Cluster Monitor Product Only $Revision: 82.2 $
Daemon
A.11.16.00 Date: 10/23/05 Patch: PHSS_33834 <<---- HERE


Cause 4 :

Check file permissions for /etc/passwd file..

#ll /etc/passwd

it should be 444...

Best Regards,
Prashanth