cluster config error

[fizan]

cmcheckconf -v -C /etc/cmcluster/cluster.conf -p pkg1.conf
Begin cluster verification...
Checking cluster file: /etc/cmcluster/cluster.conf
Note : a NODE_TIMEOUT value of 2000000 was found in line 158. This value
is recommended if the top priority is to reform the cluster as fast
as possible in case of failure. If the top priority is to minimize
reformations, consider using a higher setting. For more information see
the cluster configuration ASCII file or the Managing Serviceguard manual.
Checking nodes ... Done
Checking existing configuration ... Done
Gathering storage information
Found 9 devices on node rx26-151
Found 3 devices on node rx26-152
Analysis of 12 devices should take approximately 2 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%----100%
Found 2 volume groups on node rx26-151
Found 2 volume groups on node rx26-152
Analysis of 4 volume groups should take approximately 1 seconds
0%----10%----20%----30%----40%----50%----60%----70%----80%----90%----100%
Volume group /dev/vgpkg1 is configured differently on node rx26-151 than on node rx26-152
Volume group /dev/vgpkg1 is configured differently on node rx26-152 than on node rx26-151
Gathering network information
Beginning network probing (this may take a while)
Completed network probing
Cluster pkg1 is an existing cluster
"-p" option specified. Opening package reference file: pkg1.conf.
Parsing package file: PACKAGE_NAME pkg1 .

cmcheckconf: Nonexistent package ascii file: PACKAGE_NAME pkg1

logs;;;;

Apr 19 08:17:04 rx26-152 cmcld[6832]: The new active cluster membership is: rx26-151(id=1), rx26-152(id=2)
Apr 19 08:17:04 rx26-152 cmlvmd[6844]: Clvmd initialized successfully.
Apr 19 08:18:51 rx26-152 LVM[6878]: vgchange -a n /dev/vgpkg1
Apr 19 08:27:18 rx26-152 cmclconfd[7090]: LVML:Querying volume group /dev/vg00 for node rx26-152
Apr 19 08:27:18 rx26-152 cmclconfd[7090]: Volume group /dev/vg00 has no cluster attributes
Apr 19 08:27:18 rx26-152 cmclconfd[7090]: LVML:Querying volume group /dev/vgpkg1 for node rx26-152

-------vgchange -a y vgpkg1
vgchange: Activation mode requested for the volume group "/dev/vgpkg1" conflicts with configured mode.

Thanks

[Steven E. Protter]

Shalom,

Your volume group configuration is not consistent between the two nodes.

When building a cluster, volume group and disk information must be consistent on both nodes with regards to the shared volume groups.

You vgexport with the -p (preview) to a file, move the file to the second node to insure consistency.

This step has not been done correctly.

Also note, vg00 can not be part of the shared volume group. vg00 is the volume group each node boots off of. You can not share logical volumes or disks that are part of vg00.

SEP

[Vishu · Accepted solution]

Hi Fizan,

The error is
******
Volume group /dev/vgpkg1 is configured differently on node rx26-151 than on node rx26-152
Volume group /dev/vgpkg1 is configured differently on node rx26-152 than on node rx26-151

*******

Simply follow as:-

1. Create a VG on one node.
2. vgexport that VG in preview mode to create a map file.
3. rcp that map file to other node.
4. create VG with same name and minor number on other node.
5. vgimport that VG with the map file copied.

Now try to run cmcheckconf command.

[fizan]

cmcheckconf -p pkg1.cntl
"-p" option specified. Opening package reference file: pkg1.cntl.

cmcheckconf: Nonexistent package ascii file: . ${SGCONFFILE:=/etc/cmcluster/pkg1/pkg1.cntl}

[Raj D.]

fizan,
Have you tried creating the package ascii file and control file: if not you have to run cmmakepkg , and then to try cmcheckconf & cmapplyconf subsequently:

# cmmakepkg -v -p pkg1.conf

# cmmakepkg -v -s pkg1.cntl

#cd /etc/cmcluster/pkg1/
# ls -l #files must be there.

Then modify the pkg1.conf and pkg1.cntl file as per the need & configuration , include vg,lv,fs details on pkg1.cntl.
Copy the files in FO node ,same location.

And then run:

# cmcheckconf -v -P pkg1.conf #Remember Upper case "P" .


Cheers,

Raj.

[S.N.S]

Fizan,

Points:


1. As SEP says,the root VG for any Server- vg00 should not be shared.
The VG created on the SAN are shared in HP SG.

So as a start check the VGs on your Cluster environment...

2. cmcheckconf -v -C /etc/cmcluster/cluster.conf -p pkg1.conf <---
it should -P pkg1.conf

Step-by-Step...

Let U know your progress...and where U get stuck...
Keep the SG Admin guide near you.

Assign points to all while u are at it :-)

HTH
SNS

[Benoy Daniel]

You need to reimport the VG to secondary node and then rebuild the cluster.

[fizan]

first cretaed shared volume group executed the svg on both node n1 n2

then cmquerycl -v -C cluster.conf -n n1 -n n2

then cmcheckconf -v -C cluster.conf

then cmapplyconf -v -C cluster.conf

cmviewcl -v--- cluster up running ok. done.

pwd
/etc/cmcluster
mkdir pkg1
cmmakepkg -P pkg1.conf
cmmakepkg -P pkg1.cntl

chmod 755 pkg1.conf

cmcheckconf -v -C pkg1.conf
cmapplyconf -v -C pkg1.conf

Cluster pkg1 is an existing cluster
"-p" option specified. Opening package reference file: pkg1.conf.
Parsing package file: PACKAGE_NAME pkg1 .

cmcheckconf: Nonexistent package ascii file: PACKAGE_NAME pkg1


thanks

[fizan]

now vgchange -e y /dev/vgpkg1
sucessfully activated.

[S-M-S]

Hi Fizan,
Try the command like as follows and see what is happening


cd /etc/cmcluster/pkg1
cmcheckconf -v -P pkg1.conf
cmapplyconf -v -P pkg1.conf

[fizan]

cmcheckconf -v -C /etc/cmcluster/cluster.conf -p pkg1.conf
Begin cluster verification...

Cluster pkg1 is an existing cluster
"-p" option specified. Opening package reference file: pkg1.conf.
Parsing package file: PACKAGE_NAME pkg1 .

cmcheckconf: Nonexistent package ascii file: PACKAGE_NAME pkg1

[S-M-S]

Hi Fizan,
If you are using the full line "cmcheckconf -v -C /etc/cmcluster/cluster.conf -p pkg1.conf
" as a command I think that is the mistake


Please paste following command outputs


# ll /etc/cmcluster

# ll /etc/cmcluster/pkg1

[fizan]

CLUSTER STATUS
cluster1 up

NODE STATUS STATE
rx26-158 up running

Cluster_Lock_LVM:
VOLUME_GROUP PHYSICAL_VOLUME STATUS
/dev/vgpkg1 /dev/dsk/c7t0d4 up

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/1/2/0 lan0
PRIMARY up 0/2/1/0/5/0 lan3
STANDBY up 0/1/2/1 lan1
STANDBY up 0/2/1/0/6/0 lan4

PACKAGE STATUS STATE AUTO_RUN NODE
pkg1 up running enabled rx26-158

Policy_Parameters:
POLICY_NAME CONFIGURED_VALUE
Failover configured_node
Failback manual

Node_Switching_Parameters:
NODE_TYPE STATUS SWITCHING NAME
Primary up enabled rx26-158 (current)
Alternate up enabled rx26-157

NODE STATUS STATE
rx26-157 up running

Cluster_Lock_LVM:
VOLUME_GROUP PHYSICAL_VOLUME STATUS
/dev/vgpkg1 /dev/dsk/c5t0d4 up

Network_Parameters:
INTERFACE STATUS PATH NAME
PRIMARY up 0/1/2/0 lan0
PRIMARY up 0/2/1/0/5/0 lan3
STANDBY up 0/1/2/1 lan1
STANDBY up 0/2/1/0/6/0 lan4

Team thanks i executed wrong file.
now cluster with the pkg is up.

[fizan]

cluster up running