11-26-2013 12:45 AM
Environment
OS = Solaris10
SFHA = 6.0
Total cluster nodes = Two nodes cluster
"Commands optput of both nodes"
bash-3.00# /opt/VRTS/bin/gabconfig -c -x
GAB driver already configured. Honored other options.
#lltconfig
LLT is running
#gabconfig
GAB Port Memberships
====================================================
Node2
bash-3.00# cat /etc/gabtab
/sbin/gabconfig -c -n2
bash-3.00# cat /etc/llttab
set-node PK13
set-cluster 29564
link bge2 /dev/bge:2 - ether - -
link-lowpri aggr1 /dev/aggr:1 - ether - -
Node1
bash-3.00# cat /etc/gabtab
/sbin/gabconfig -c -n2
bash-3.00# cat /etc/llttab
set-node PK12
set-cluster 29564
link bge2 /dev/bge:2 - ether - -
link-lowpri aggr1 /dev/aggr:1 - ether - -
bash-3.00# /opt/VRTS/bin/hastatus
attempting to connect....
VCS ERROR V-16-1-10600 Cannot connect to VCS engine
attempting to connect....not available; will retry
attempting to connect....retrying
Solved! Go to Solution.
11-26-2013 05:12 AM
This means that either LLT or GAB is not running on any of the nodes or that there is no network connection through the LLT links.
Please check and start:
To check:
#gabconfig -a
#lltconfig
#lltstat -nvv
to start:
http://www.symantec.com/business/support/index?page=content&id=TECH49819
To test LLT links:
http://www.symantec.com/business/support/index?page=content&id=TECH159065
To start HAD demon GAB needs to be seeded.
regards,
Dan
11-26-2013 05:12 AM
This means that either LLT or GAB is not running on any of the nodes or that there is no network connection through the LLT links.
Please check and start:
To check:
#gabconfig -a
#lltconfig
#lltstat -nvv
to start:
http://www.symantec.com/business/support/index?page=content&id=TECH49819
To test LLT links:
http://www.symantec.com/business/support/index?page=content&id=TECH159065
To start HAD demon GAB needs to be seeded.
regards,
Dan
11-26-2013 09:41 PM
Hi Zahid,
I think the problem caused by you start GAB by "gabconfig -c -x". Restart GAB by :
gabconfig -U
sh /etc/gabtab
Maybe this will help.
11-27-2013 02:49 AM
Hi Zahid,
Is the version same on both the nodes ? If yes,
I would suggest to stop the entire stack & try once again from bottom up approach
On both the nodes
# gabconfig -U (unconfigure gab)
# modinfo |grep -i gab (find module ID for gab)
# modunload -C <gab_module ID>
# lltconfig -U (unconfigure LLT)
# modinfo |grep -i llt (find module ID for LLT)
# modunload -C <LLT_module_id>
Start again
# /etc/init.d/S70llt start OR # lltconfig -C
# lltconfig (it should say "LLT is running)
# /etc/init.d/S92gab start OR #gabconfig -C
now check status of gab whether you have both the ports in gab
If you still face issues, check the LLT communication as suggested by Dan ..
G
11-27-2013 05:39 AM
FYI--
An alternative to Gaurav's instructions on how to restart the entire VCS stack, I find it more convenient and possibly more comprehensive to use the CPI to do the same:
# /opt/VRTS/install/installvcs601 -stop # stops all drivers and processes of a product # /opt/VRTS/install/installvcs601 -start # starts all drivers and processes of a product
But as others have already pointed out, it is important to confirm your LLT configuration before taking other steps. 'lltstat -nvv active' should show you whether LLT is working; no point doing anything else until you have resolved that... Then move on to GAB.
Of course, 'hastatus' is still going to show you the error as you have shown unless it is already running, so you need to start 'had' before issuing an 'hastatus' command...
-kjb