unable to mount the LV

[Suresh Babu G]

Hi Experts,

We are unable to mount the file system and getting the below error

# vgchange -a y vg01
vgchange: Activation mode requested for the volume group "/dev/vg01" conflicts with configured mode.

this particular mount point s for MC/SG.

awaiting for your reply.

Regards
Suresh

[R.K. #]

# vgdisplay vg01

[Suresh Babu G]

Hi,

here is the vgdisplay

# vgdisplay -v vg01 | more
vgdisplay: Volume group not activated.
vgdisplay: Cannot display volume group "vg01".

Regards
Suresh

[Fabio Ettore · Accepted solution]

Hi,

you get this error because this VG is cluster aware (it was used vgchange -c y on it). I'd guess you can activate it by

vgchange -a e vg01

After that you should note available, exclusive on the line about VG Status.

However this volume groups usually are activated by the package itself (I mean when executing cmrunpkg).

Are you sure the volume group is not already active on another cluster node? In that case you cannot activate vg01 even using vgchange -a e vg01 (you could do that by vgchange -a r vg01 but that means you have read-only access on the vg).

HTH.

Best regards,
Fabio

[madhuchakkaravarthy]

HI

check your package control script

/etc/cmcluster/package/package.cntl

[Suresh Babu G]

Hi Fabio,

Thanks for your help now the VG is activated in Exclusive Mode.

# vgchange -a e vg01
Activated volume group in Exclusive Mode.
Volume group "vg01" has been successfully changed.

the vgdisplay is as below

# vgdisplay -v vg01 | more
------------------------------------------------------------------

--- Volume groups ---
VG Name /dev/vg01
VG Write Access read/write
VG Status available, exclusive
Max LV 255
Cur LV 1
Open LV 1
Max PV 16
Cur PV 1
Act PV 1
Max PE per PV 2399
VGDA 2
PE Size (Mbytes) 64
Total PE 2399
Alloc PE 2
Free PE 2397
Total PVG 0
Total Spare PVs 0
Total Spare PVs in use 0
VG Version 1.0
VG Max Size 2399g
VG Max Extents 38384

--- Logical volumes ---
LV Name /dev/vg01/lvol1
LV Status available/syncd
LV Size (Mbytes) 64
Current LE 1
Allocated PE 2
Used PV 1


--- Physical volumes ---
PV Name /dev/dsk/c1t0d0
PV Name /dev/dsk/c2t0d0 Alternate Link
PV Name /dev/dsk/c3t0d0 Alternate Link
PV Name /dev/dsk/c4t0d0 Alternate Link
PV Status available
Total PE 2399
Free PE 2397
Autoswitch On
Proactive Polling On


when we start the package it is giving the below error

# cmrunpkg -v pkg_1
Running package pkg_1 on node testserv1
Node testserv1 is currently unable to run package pkg_1.
Unable to run package pkg_1 on node testserv1
Check the syslog and pkg log files for more detailed information
cmrunpkg: Unable to start some package or package instances

and in syslog it is not giving any error

Regards
Suresh

[Kapil Jha]

go to /etc/cmcluter/
in pkg_name.log file you would see error.

BR,
Kapil+

[Horia Chirculescu]

Hello,

You should execute:

vgchange -c y /dev/vg01

in order to return the volume group to the cluster.

Horia.

[Suresh Babu G]

Hi,

# cat pkg_1.cntl.log
/etc/cmcluster/pkg_1/pkg_1.cntl[434]: Syntax error at line 441 : `(' is not expected.
#

the above out we are getting

Regards
Suresh

[Michael Steele_2]

Is that vg already in use by the other node?

[saran2010]

Hi Suresh,

Kindly check the following things

1. which is the primary node for the package u r trying to start.

2. check whether u are able to activate the VG in that primary node.

3 if u are unable to activate means kindly check whether the VG is activated on any of the secondary node.

4 If it is activated on any of the node means deactivate it and try to start the package.(pls make sure that VG should not be activated in any of the node before starting the package)

Thanks

Saravanan

[Suresh Babu G]

Hi Michael,

Is that vg already in use by the other node?

no it is not used in another node

regards
suresh

[Suresh Babu G]

Hi Saravanan,

Hi Suresh,

Kindly check the following things

1. which is the primary node for the package u r trying to start.

the hostname of the primary node is testserv1

2. check whether u are able to activate the VG in that primary node.

with the exclusive mode only i am able to start


3 if u are unable to activate means kindly check whether the VG is activated on any of the secondary node.

no it is not activitated on the other node.

4 If it is activated on any of the node means deactivate it and try to start the package.(pls make sure that VG should not be activated in any of the node before starting the package)

i am getting the below error after using vgchange -a e vg01

# cmrunpkg pkg_1
Running package pkg_1 on node testserv1
Node testserv1 is currently unable to run package pkg_1.
Unable to run package pkg_1 on node testserv1
Check the syslog and pkg log files for more detailed information
cmrunpkg: Unable to start some package or package instances

Thanks

Saravanan

[saran2010]

Hi suresh,

In point no.2 u have mentioned that u activated the Vg in primary node..

Did u deactivate it before starting package?

Kindly deactivate it and start the package,

Thanks

[Suresh Babu G]

Hi,

can some one check the syntax of the cntl file. i have attached in this thread.

the cntl.log file output is as below

/etc/cmcluster/pkg_1/pkg_1.cntl[434]: Syntax error at line 441 : `(' is not expected.


regards
Suresh

[Michael Steele_2]

cmquerycl -v -C /etc/cmcluster/verify.ascii -n node1 -n node2
cmcheckconf -v -C /etc/cmcluster/verify.ascii

Note the name 'verify.ascii' is similar to the actual 'cluster.ascii'.

Do not overwrite the 'cluster.ascii' file. Create a 2nd file for trouble shoot and NOTE THE ERRORS>

##########################
cmscancl -n node1 -n node2 -o output_file


##########################
Read MCSG log on later version:

cmreadlog /var/opt/cmom/cmomd.log

[Suresh Babu G]

Hi,

Initially i got the below error
# cmcheckconf -v -C /etc/cmcluster/cmclconfig.ascii
Begin cluster verification...
Checking cluster file: /etc/cmcluster/cmclconfig.ascii
No value specified for FIRST_CLUSTER_LOCK_VG at line 87
cmcheckconf: Error found in cluster file: /etc/cmcluster/cmclconfig.ascii.

after that i have given the FIRST_CLUSTER_LOCK_VG as /dev/vg01

but again i got the below error


# cmcheckconf -v -C /etc/cmcluster/cmclconfig.ascii
Begin cluster verification...
Checking cluster file: /etc/cmcluster/cmclconfig.ascii
Note : a NODE_TIMEOUT value of 2000000 was found in line 139. 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.
Undefined variable FIRST_CLUSTER_LOCK_PV for node testserv1.
Provide first cluster lock physical volume entry.
cmcheckconf: Error found in cluster file: /etc/cmcluster/cmclconfig.ascii.

regards
suresh

[Fabio Ettore]

Hi Suresh,

so I understand you have 2 problems:

- first one, about syntax error in .cntl file:

I think it's due to this function (defined manually under function customer_defined_halt_cmds):

[...]
kill -9 (ps -ef |grep -v grep |grep "xclock" |cut -c10-14)
[...]

After the kill -9 command (which is the line 441) you cannot use just "(", I'd guess you missed a "$", I mean to have something like:

kill -9 $(ps -ef |grep -v grep |grep "xclock" |cut -c10-14)

Anyway I suggest trying this function out of .cntl file before inserting it there.

- second issue, definition of FIRST_CLUSTER_LOCK_PV: you must define it into the cluster ascii file under each node definition (the following definitions are just a sample, you have to replace them with right values for your cluster):

NODE_NAME pippo
NETWORK_INTERFACE lan0
STATIONARY_IP 10.0.0.1
FIRST_CLUSTER_LOCK_PV /dev/dsk/c0t3d0

HTH.

Best regards,
Fabio

[Michael Steele_2]

Run the cmscancl command. This command's primary job is to verify your lan connections. The file is long, after you vi it, search for CONNECTIONS or LAN, etc.

Note that MCSG runs on MAC addresses. So familiarize yourself with the 'linkloop' command and run a 'arp -a' report.

Collect all the IP's and MACs of both nodes, they'll be listed in the cmscancl report.

[Michael Steele_2]

# Policy 1:
USER_NAME john
USER_HOST bit
USER_ROLE PACKAGE_ADMIN
# Policy 2:
USER_NAME john
USER_HOST bit
USER_ROLE MONITOR

Could you review your config for Policies and roles of admins?

The only 'conflicts' in MC/SG that I can currently find that are flagged as errors are located here. In the above, 'john' appears' twice. In MC/SG, 'john' can only appear once.

Also, I'm a little confused about how this error can be coming up know for the first time. Is this a new build?

[Suresh Babu G]

Hi Fabio,

getting the below error after making the changes

Entries made

NODE_NAME testserv1
NETWORK_INTERFACE lan0
HEARTBEAT_IP 10.98.10.92
NETWORK_INTERFACE lan1
HEARTBEAT_IP 172.17.1.19
FIRST_CLUSTER_LOCK_VG /dev/vg01
FIRST_CLUSTER_LOCK_PV /dev/dsk/c1t0d0
# CLUSTER_LOCK_LUN

# Warning: There are no standby network interfaces for lan0.
# Warning: There are no standby network interfaces for lan1.

NODE_NAME testserv2
NETWORK_INTERFACE lan0
HEARTBEAT_IP 10.98.10.98
NETWORK_INTERFACE lan1
HEARTBEAT_IP 172.17.1.20
# FIRST_CLUSTER_LOCK_VG /dev/vg01
FIRST_CLUSTER_LOCK_PV /dev/dsk/c1t0d0


Error

# cmcheckconf -C /etc/cmcluster/cmclconfig.ascii
First cluster lock volume group /dev/vg01 needs to be designated as a cluster aware volume group.
cmcheckconf: Unable to verify cluster file: /etc/cmcluster/cmclconfig.ascii.


regards
suresh

[Suresh Babu G]

Hi Fabio,

getting the below error after making the changes

Entries made

NODE_NAME testserv1
NETWORK_INTERFACE lan0
HEARTBEAT_IP 10.98.10.92
NETWORK_INTERFACE lan1
HEARTBEAT_IP 172.17.1.19
FIRST_CLUSTER_LOCK_VG /dev/vg01
FIRST_CLUSTER_LOCK_PV /dev/dsk/c1t0d0
# CLUSTER_LOCK_LUN

# Warning: There are no standby network interfaces for lan0.
# Warning: There are no standby network interfaces for lan1.

NODE_NAME testserv2
NETWORK_INTERFACE lan0
HEARTBEAT_IP 10.98.10.98
NETWORK_INTERFACE lan1
HEARTBEAT_IP 172.17.1.20
# FIRST_CLUSTER_LOCK_VG /dev/vg01
FIRST_CLUSTER_LOCK_PV /dev/dsk/c1t0d0


Error

# cmcheckconf -C /etc/cmcluster/cmclconfig.ascii
First cluster lock volume group /dev/vg01 needs to be designated as a cluster aware volume group.
cmcheckconf: Unable to verify cluster file: /etc/cmcluster/cmclconfig.ascii.


regards
suresh

[madhuchakkaravarthy]

do u have any file with

.cntl and .cntl.bak under /etc/cmcluster



i guess some one might changed the script and may include user defined alias.

when u run cmrunpkg of a particular node.
package should be up,,

confirm with two files present or not..

regards

MC

[Suresh Babu G]

Hi MC,

except me no one has access to this server and only one file is present (pkg_1.cntl) under /etc/cmcluster/pkg_1 folder and if i run the package it is giving the below error

# cmrunpkg pkg_1
Running package pkg_1 on node testserv1
Node testserv1 is currently unable to run package pkg_1.
Unable to run package pkg_1 on node testserv1
Check the syslog and pkg log files for more detailed information
cmrunpkg: Unable to start some package or package instances

regards
suresh

[Horia Chirculescu]

>Check the syslog and pkg log files for more detailed information
cmrunpkg: Unable to start some package or package instances

Did you checked those log files?

Horia.