X86 sc3.1-0805 sol10-0606 - Doesn't boot in cluster mode

Question

Hi,
i'm at my first experience with Sun Cluster on x86.
I've already tried at home with two p4 whiteboxes and now repeating the experiment here at work with a similar conf.( i happily run 6 v490 in 3 clustered pair with 3510Fc and a test 2 u10 clustered pair with Multipack).
no matter what i try i ever end up with the same results:
nodes boot up outside of the cluster. Interconnects doesn't start and (i think maybe cause of that) global devices don't get initialized.
I already tried many reinstall, already tried to add etc/cluster/nodeid to the filelist.ramdisk and update boot archive e reconfigure, like described in an infodoc to workaround a well known problem, but nothing changed
This is the situation as i start either one of the nodes:
mordor-nodo2 # svcs -x
svc:/system/cluster/mountgfsys:default (Suncluster mountgfsys service)
State: maintenance since Tue Aug 08 15:37:51 2006
Reason: Restarter svc:/system/svc/restarter:default gave no explanation.
See: http://sun.com/msg/SMF-8000-9C
See: /var/svc/log/system-cluster-mountgfsys:default.log
Impact: 13 dependent services are not running. (Use -v for list.)
svc:/system/cluster/gdevsync:default (Suncluster gdevsync service)
State: maintenance since Tue Aug 08 15:37:51 2006
Reason: Restarter svc:/system/svc/restarter:default gave no explanation.
See: http://sun.com/msg/SMF-8000-9C
See: /var/svc/log/system-cluster-gdevsync:default.log
Impact: 13 dependent services are not running. (Use -v for list.)
svc:/network/multipath:cluster (Network Monitor Daemon)
State: maintenance since Tue Aug 08 15:37:39 2006
Reason: Maintenance requested by an administrator.
See: http://sun.com/msg/SMF-8000-63
See: in.mpathd(1M)
See: /etc/svc/volatile/network-multipath:cluster.log
Impact: This service is not running.
Only the public interface is up in sc_ipmp0 group
Cluster interconnects are 3com elxl interface in all two nodes and are connected with cross-cables elxl0->>elxl0 elxl1-->elxl1 (verified that it works)
I've removed switches and put cross-cables while troubleshooting to have a simpler setup.
/etc/vfstab - every fs is mirrored - metadb are in s7 , globalfs in s3
/dev/md/dsk/d0 - - swap - no -
/dev/md/dsk/d10 /dev/md/rdsk/d10 / ufs 1 no -
/devices - /devices devfs - no -
ctfs - /system/contract ctfs - no -
objfs - /system/object objfs - no -
swap - /tmp tmpfs - yes -
#/dev/md/dsk/d20 /dev/md/rdsk/d20 /globaldevices ufs 2 yes -
/dev/md/dsk/d20 /dev/md/rdsk/d20 /global/.devices/node@1 ufs 2 no global
Planning to add a multipack for the multihost disks. But i'd like to solve this problem before
Nothing useful appears on the logs.
I only got the initial
Not booting in cluster mode
and nothing more
Maybe missing something about versions of the software/hardware am using that for any reason can't work togheter?
or some fix is needed?
Any hint would be appreciated
I stay at disposal for any kind of info
thanks

Answer 1

Hi,
you should elaborate a bit on your hardware.
Is it x64 or x32, what is your shared storage.
The "not booting in cluster mode" appears in example if you want to install SC 3.1 and SC3.2 on x32 hardware. If this your goal, you should start with Solaris Express and Solaris Cluster express.
Kind regards
Detlef