Zone hang issue

I have a solaris10 zone running an oracle db in it. The zone hangs after some time the db is started and i 'm unable to log into the zone after that so i have to halt and restart the zone along with db .. but after few hrs again it hangs. What would be the reason for it. I did svcs -xv and got following result which i do not think could be the issue of zone hang.

svc:/network/rpc/keyserv:default (RPC encryption key storage)
 State: maintenance since Wed Apr 28 00:06:42 2010
Reason: Start method exited with $SMF_EXIT_ERR_CONFIG.
   See: http://sun.com/msg/SMF-8000-KS
   See: man -M /usr/share/man -s 1M keyserv
   See: /var/svc/log/network-rpc-keyserv:default.log
Impact: 1 dependent service is not running:
        svc:/network/rpc/nisplus:default

what is inside the log (/var/svc/log/network-rpc-keyserv:default.log) file?


[ Aug  5 00:40:00 Executing start method ("/usr/sbin/keyserv") ]
[ Aug  5 00:40:00 Method "start" exited with status 0 ]
[ Sep 27 23:05:54 Executing start method ("/usr/sbin/keyserv") ]
[ Sep 27 23:05:54 Method "start" exited with status 0 ]
[ Nov 12 21:40:38 Executing start method ("/usr/sbin/keyserv") ]
[ Nov 12 21:40:38 Method "start" exited with status 0 ]
[ Dec 31 05:00:47 Executing start method ("/usr/sbin/keyserv") ]
[ Dec 31 05:00:47 Method "start" exited with status 0 ]


thats all? where are the newer messages? there should be something from Wed Apr 28 00:06:42 2010...

what says dmesg ?

How do you log in to the zone ?
How does it fails ?
What is the zone status ?
Are there processes still running in the zone ?
Is there some capping set in the zone configuration ?

Asnwers to Jillagre's questions

How do you log in to the zone ?  zlogin, zlogin -C & ssh from putty
How does it fails ?    After few hrs i'm unable to connect to zone using any of the above methods

What is the zone status ?  The status remains running

Are there processes still running in the zone ?  The smon, pmon processes of the 
db continues to run 

Is there some capping set in the zone configuration:  There is no capping sent  

DukeNuke2


[ Apr 27 21:26:10 Executing start method ("/usr/sbin/keyserv") ]
[ Apr 27 21:26:10 Method "start" exited with status 96 ]
[ Apr 28 00:04:52 Enabled. ]
[ Apr 28 00:05:05 Executing start method ("/usr/sbin/keyserv") ]
[ Apr 28 00:05:05 Method "start" exited with status 96 ]
[ Apr 28 00:06:42 Leaving maintenance because clear requested. ]
[ Apr 28 00:06:42 Enabled. ]
[ Apr 28 00:06:42 Executing start method ("/usr/sbin/keyserv") ]
[ Apr 28 00:06:42 Method "start" exited with status 96 ]

Check out the memory status, hope there is no dedicated memory assigned for zones. Start the DB in zone and test the memory utilization in Global. If not memory issue, it will something else then :frowning:

afaik the "keyserv" is for secure nfs. do you use nfs inside the zone? maybe you have data on a nfs share which get in a "stall" state?

No i 'm not using any kind of nfs share/mount in this zone.

if there are no nfs or nis+ issues this might not be the cause of the hanging zone...

zlogin with no -C option should always work on a running zone. This is odd.

You might leave a

zlogin -C yourzone

open until the zone hangs to get zones system messages. Also have a look at <zonepath>/root/var/adm/messages* for clues.

What says

prstat -z yourzone

when it hangs ?

Actually in my absence someone had configured memroy capping for this zone and capped its memory to 2G only .. i changed it to 4G and its running fine since then. Thanx to all you guys for assisting me in resolving the issue. :b:

Too bad you replied earlier

without checking.