You (gaurav saxena) wrote:
> Hello
> 
> 2010/6/30 Matthias Pfützner <matth...@pfuetzner.de>
> 
> > Gaurav,
> >
> > Jun 29 20:26:08 opensolaris gdm-binary[6634]: [ID 702911 daemon.warning]
> > WARNING: Failed to start X server several times in a short time period;
> > disabling display :0
> >
> > Sadly, there's no indication, as to WHY that fails...
> >
> >
> Ok. yes its really sad :(.

I assume, it has something to do with the below error messages regarding the
rpool. But, sadly, I'm not sure...

> > And: It seems, you have a single-node cluster, right?
> >
> >
> Yes I am having a single node cluster.

Which is OK, but for learning purposes might be a bit to lightwight... A
two-node cluster is better...

> > Jun 23 12:51:47 opensolaris genunix: [ID 965873 kern.notice] NOTICE: CMM:
> > Node opensolaris (nodeid = 1) with votecount = 1 added.
> > Jun 23 12:51:47 opensolaris genunix: [ID 843983 kern.notice] NOTICE: CMM:
> > Node opensolaris: attempting to join cluster.
> > Jun 23 12:51:47 opensolaris genunix: [ID 525628 kern.notice] NOTICE: CMM:
> > Cluster has reached quorum.
> > Jun 23 12:51:47 opensolaris genunix: [ID 377347 kern.notice] NOTICE: CMM:
> > Node opensolaris (nodeid = 1) is up; new incarnation number = 1277277706.
> > Jun 23 12:51:47 opensolaris genunix: [ID 108990 kern.notice] NOTICE: CMM:
> > Cluster members: opensolaris.
> > Jun 23 12:51:47 opensolaris genunix: [ID 279084 kern.notice] NOTICE: CMM:
> > node reconfiguration #1 completed.
> > Jun 23 12:51:50 opensolaris genunix: [ID 499756 kern.notice] NOTICE: CMM:
> > Node opensolaris: joined cluster.
> >
> > And you have same errors in your config of the zones for the cluster, as
> > far
> > as the storage setup is concerned:
> >
> >
> I see these messages at boot time. Are these errors ?

No, that's OK!

> > Jun 23 13:09:52 opensolaris Cluster.RGM.global.rgmd: [ID 224900
> > daemon.notice] launching method <hastorageplus_validate> for resource
> > <clusterresource>, resource group <mygroup>, node <opensolaris>, timeout
> > <1800> seconds
> > Jun 23 13:09:52 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_validate]:
> > [ID 148650 daemon.notice] Started searching for devices in '/dev/dsk' to
> > find the importable pools.
> > Jun 23 13:09:53 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_validate]:
> > [ID 547433 daemon.notice] Completed searching the devices in '/dev/dsk' to
> > find the importable pools.
> > Jun 23 13:09:53 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_validate]:
> > [ID 792255 daemon.warning] Failed to update the cachefile contents in
> > /var/cluster/run/HAStoragePlus/zfs/rpool.cachefile to CCR table
> > rpool.cachefile for pool rpool : file
> > /var/cluster/run/HAStoragePlus/zfs/rpool.cachefile open failed: No such file
> > or directory.
> > Jun 23 13:09:54 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_validate]:
> > [ID 471757 daemon.error] cannot unmount '/' : Device busy
> > Jun 23 13:09:54 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_validate]:
> > [ID 316615 daemon.error] Failed to export :rpool
> >
> > And a bit later:
> >
> > Jun 23 13:11:15 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_init]: [ID
> > 792255 daemon.warning] Failed to update the cachefile contents in
> > /var/cluster/run/HAStoragePlus/zfs/rpool.cachefile to CCR table
> > rpool.cachefile for pool rpool : file
> > /var/cluster/run/HAStoragePlus/zfs/rpool.cachefile open failed: No such file
> > or directory.
> > Jun 23 13:11:15 opensolaris
> > SC[,SUNW.HAStoragePlus:8,mygroup,clusterresource,hastorageplus_init]: [ID
> > 471757 daemon.error] cannot unmount '/' : Device busy
> > Jun 23 13:11:15 opensolaris Cluster.RGM.global.rgmd: [ID 515159
> > daemon.notice] method <hastorageplus_init> completed successfully for
> > resource <clusterresource>, resource group <mygroup>, node <opensolaris>,
> > time used: 0% of timeout <1800 seconds>
> > Jun 23 13:11:15 opensolaris Cluster.RGM.global.rgmd: [ID 443746
> > daemon.notice] resource clusterresource state on node opensolaris change to
> > R_OFFLINE
> > Jun 23 13:11:15 opensolaris Cluster.RGM.global.rgmd: [ID 529407
> > daemon.notice] resource group mygroup state on node opensolaris change to
> > RG_OFFLINE
> > Jun 23 13:11:15 opensolaris Cluster.RGM.global.rgmd: [ID 529407
> > daemon.notice] resource group mygroup state on node opensolaris change to
> > RG_PENDING_ONLINE
> > Jun 23 13:11:15 opensolaris Cluster.RGM.global.rgmd: [ID 784560
> > daemon.notice] resource clusterresource status on node opensolaris change to
> > R_FM_UNKNOWN
> >
> > So, some of these problems (X server start et.al.) might be related to
> > these
> > errros.
> >
> > First: The rpool is a local filesystem (better: a local pool), as it
> > contains
> > the root-filesystem This can not be made as an fail-over filesystem. But it
> > seems, you did configure that way.
> >
> >
> No actually i just created two zfs filesystems on rpool as I have told
> earlier.
> Is it wrong ?

I assume so, yes... As the thing, that can be moved are the POOLS, not the zfs
Filesystems. Therefore, you would have needed a new pool, which can be moved
from system to system... you can't MOVE the rpool, as that's needed for the
basic underlying OS... ;-)

> > So, if I were you, I would delete all the CONFIGURATIONS for things in the
> > cluster, so that you get a fresh OpenSolaris 2009.06 with the OHAC packages
> > installed.
> >
> > I think I can reinstall opensolaris only and install OHAC on it again but I
> have reinstalled this for the second time only. :(

I guess, you would NOT need to re-install, but to just DECONFIGURE all the
resources, storage et.al. that you did in the cluster config.

> > Then, I would reboot the server (are you sure, you want only a single node
> > cluster? Or am I missing something here?), and see, if the X-server comes
> > up
> > again. Once that running again, I would recommend re-reading the basic
> > cluster
> > setup manuals for how to setup a failover zone and the storage needed to
> > host
> > the filesystems for those zones...
> >
> >    Matthias--
> >    Matthias Pfützner    | mailto:pfu...@germany | Am Anfang wäre das Stück
> >  @work: +49 6103 752-394 | @home: +49 6151 75717 | schnell zu Ende.
> >  SunCS, Amp?restraße 6  | Lichtenbergstraße 73  | (21.10.93)
> >    63225 Langen, FRG    | 64289 Darmstadt, FRG  | Hellmuth Karasek
> >
> Actually in the manual it was suggested to create a mirror zpool of the root
> filesystem and then install zfs on it or create a pool on file but I thought
> I could use the original rpool only.

For testing, you don't need the mirror.
Yes, as stated above, the "thing" that will be "failed-over" is the
POOL. Therefore you need a pool. And you simply can't use the rpool...

> Is this the reason for all problems ?

It might be!

> But as far as I remember display is not working since I ran scinstall and I
> did this all zone work from the console only so I don't think this could be
> the reason.

That's the strange thing... But, in order to be sure, when that really
started, I would propose to really de-configure the actual cluster setup, and
staring all over again (not with INSTALL, only with configuring!)

> -- 
> Thanks and Regards ,
> Gaurav Saxena

  Matthias
-- 
    Matthias Pfützner    | mailto:pfu...@germany | Am Anfang wäre das Stück
 @work: +49 6103 752-394 | @home: +49 6151 75717 | schnell zu Ende.
  SunCS, Ampèrestraße 6  | Lichtenbergstraße 73  | (21.10.93)
    63225 Langen, FRG    | 64289 Darmstadt, FRG  | Hellmuth Karasek
_______________________________________________
zones-discuss mailing list
zones-discuss@opensolaris.org

Reply via email to