Piotr Jasiukajtis wrote: > Hi, > > I created 2node Colorado cluster on OpenSolaris 2009.06 b111 and I had > a few issues: > > - scinstall has problems with creating vnics on interconnect interfaces > > - scinstall says that I don't have enough NICs for interconnects (I > had to plumb intr NIC before 'scinstall') >
Piotr, I'm glad you're trying this out! There are several open bugs still in the VNICs area, among them: http://bugs.opensolaris.org/view_bug.do?bug_id=6816906 http://bugs.opensolaris.org/view_bug.do?bug_id=6834248 http://bugs.opensolaris.org/view_bug.do?bug_id=6833691 http://bugs.opensolaris.org/view_bug.do?bug_id=6816908 http://bugs.opensolaris.org/view_bug.do?bug_id=6834209 http://bugs.opensolaris.org/view_bug.do?bug_id=6834579 http://bugs.opensolaris.org/view_bug.do?bug_id=6834262 If none of these describe the problems you're seeing, please file a new bug. And note that VNICs do not seem to work properly in VirtualBox. > - my interconnect was broken (one intr setup, no quorum device/server > yet), so the second node was waiting for the first node forever. > It's normal that I was not able to log in to the system but I tried to > shutdown the node by 'power button'.. > System said 'WARNING: Power off requested from power button or SC, > powering down the system!', > but after a few minutes of doing nothing (well, no logs on console) it > said that there was a problem (I don't know what kind of problem) and > the shutdown was aborted. > No, I don't have logs/cores since I rollbacked the snapshot.. > > - I think that 'cluster shutdown' should really power down the nodes > instead of just 'init 0'. What about a new option -p like powerdown > 'cluster shutdown -p'? File an RFE. > > - I got this after 'clnode remove' on second node: > NOTICE: softmac: received DL_ERROR_ACK to DL_BIND_ACK; DLPI errno > 0x7ffff08a, UNIX errno 1 > dump on /dev/zvol/dsk/rpool/dump size 2048 MB > Apr 25 10:47:01 svc.startd[8]: svc:/system/cluster/scmountdev:default: > Method "/usr/cluster/lib/svc/method/scmountdev start" failed with exit > status 2. > Apr 25 10:47:01 svc.startd[8]: svc:/system/cluster/scmountdev:default: > Method "/usr/cluster/lib/svc/method/scmountdev start" failed with exit > status 2. > Apr 25 10:47:01 svc.startd[8]: svc:/system/cluster/scmountdev:default: > Method "/usr/cluster/lib/svc/method/scmountdev start" failed with exit > status 2. > Apr 25 10:47:01 svc.startd[8]: system/cluster/scmountdev:default > failed: transitioned to maintenance (see 'svcs -xv' for details) > pseudo-device: dtrace0 > dtrace0 is /pseudo/dtrace at 0 > /usr/cluster/bin/scdidadm: Could not load DID instance list. > /usr/cluster/bin/scdidadm: Cannot open /etc/cluster/ccr/global/did_instances. > Error: /etc/cluster/ccr/global/infrastructure does not exist > UNRECOVERABLE ERROR: /etc/cluster/ccr/infrastructure file is corrupted > Please reboot in noncluster mode(boot -x) and Repair > syncing file systems... done > Press any key to reboot. > I haven't seen this one before. Can you file a bug? Thanks, Nick
