Ok guys & gals here is the scenario: 1.) Host RHEL 4 U3 2.6.9-34.0.2.EL 2.) OCFS2 latest version 3.) Successfully formatted & mounted OCFS2 filesystems on 2 nodes /dev/sdb1 /u02/oradata/usdev/voting /dev/sdc1 /u02/oradata/usdev/data01 /dev/sdd1 /u02/oradata/usdev/data02 /dev/sde1 /u02/oradata/usdev/data03 4.) Downloaded & installed ASMLib 2.0 on both nodes 5.) Ran /etc/init.d/oracleasm configure 6.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 7.) Shutdown node 2 8.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 9.) Unmounted OCFS2 filesystems 10.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Success 11.) Ran /etc/init.d/oracleasm createdisk vol2 /dev/sdd1 & /dev/sde1 Success 12.) Remount OCFS2 filesystems: mount -a -t ocfs2 Error: ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" 13.) /etc/init.d/oracleasm listdisks VOL1 VOL2 VOL3 SO...Are my OCFS2 partitions corrupted? Or can I not mount them once they're configured for ASM? Is this the correct behavior, or am I missing something? Thanks!! Chris Taylor Sr. Oracle DBA Unique Solutions www.unisolinc.com Office: 336-667-2447 xt. 2242 Cell: 336-262-5545 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://oss.oracle.com/pipermail/ocfs-users/attachments/20060804/dadb72c1/attachment.html
Hi, I can't understand why you want to use OCFS2 and ASM together in the same partition. For a Oracle rack instalation you MUST use OCFS2 or ASM for each partition. you can't use the two together. You you add your ocfs2 partition to asm you will corrupt that partition, possibly loosing all your data. and you will need to format again the partition. Best Regards Leonardo Rodrigues de Mello -----Original Message----- From: ocfs-users-bounces@oss.oracle.com on behalf of Chris Taylor Sent: sex 4/8/2006 10:35 To: ocfs-users@oss.oracle.com Cc: Subject: [Ocfs-users] OCFS2 and ASM Question Ok guys & gals here is the scenario: 1.) Host RHEL 4 U3 2.6.9-34.0.2.EL 2.) OCFS2 latest version 3.) Successfully formatted & mounted OCFS2 filesystems on 2 nodes /dev/sdb1 /u02/oradata/usdev/voting /dev/sdc1 /u02/oradata/usdev/data01 /dev/sdd1 /u02/oradata/usdev/data02 /dev/sde1 /u02/oradata/usdev/data03 4.) Downloaded & installed ASMLib 2.0 on both nodes 5.) Ran /etc/init.d/oracleasm configure 6.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 7.) Shutdown node 2 8.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 9.) Unmounted OCFS2 filesystems 10.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Success 11.) Ran /etc/init.d/oracleasm createdisk vol2 /dev/sdd1 & /dev/sde1 Success 12.) Remount OCFS2 filesystems: mount -a -t ocfs2 Error: ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" 13.) /etc/init.d/oracleasm listdisks VOL1 VOL2 VOL3 SO...Are my OCFS2 partitions corrupted? Or can I not mount them once they're configured for ASM? Is this the correct behavior, or am I missing something? Thanks!! Chris Taylor Sr. Oracle DBA Unique Solutions www.unisolinc.com Office: 336-667-2447 xt. 2242 Cell: 336-262-5545 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://oss.oracle.com/pipermail/ocfs-users/attachments/20060804/6db29b7d/attachment.html
Thats kind of what I was looking for. The Install Guides I'm looking at didn't make it clear that they were separate. Reading more closely, I notice that the specific guide I'm going through only formats one partition as OCFS2 and sets up the others as ASM. As you can tell, I'm still learning the new 10g RAC architecture options and appreciate the input. Thanks, Chris Taylor Sr. Oracle DBA Unique Solutions www.unisolinc.com <http://www.unisolinc.com/> Office: 336-667-2447 xt. 2242 Cell: 336-262-5545 _____ From: Leonardo Rodrigues de Mello [mailto:Leonardo.Mello@planejamento.gov.br] Sent: Friday, August 04, 2006 12:59 PM To: Chris Taylor; ocfs-users@oss.oracle.com Subject: RE: [Ocfs-users] OCFS2 and ASM Question Hi, I can't understand why you want to use OCFS2 and ASM together in the same partition. For a Oracle rack instalation you MUST use OCFS2 or ASM for each partition. you can't use the two together. You you add your ocfs2 partition to asm you will corrupt that partition, possibly loosing all your data. and you will need to format again the partition. Best Regards Leonardo Rodrigues de Mello -----Original Message----- From: ocfs-users-bounces@oss.oracle.com on behalf of Chris Taylor Sent: sex 4/8/2006 10:35 To: ocfs-users@oss.oracle.com Cc: Subject: [Ocfs-users] OCFS2 and ASM Question Ok guys & gals here is the scenario: 1.) Host RHEL 4 U3 2.6.9-34.0.2.EL 2.) OCFS2 latest version 3.) Successfully formatted & mounted OCFS2 filesystems on 2 nodes /dev/sdb1 /u02/oradata/usdev/voting /dev/sdc1 /u02/oradata/usdev/data01 /dev/sdd1 /u02/oradata/usdev/data02 /dev/sde1 /u02/oradata/usdev/data03 4.) Downloaded & installed ASMLib 2.0 on both nodes 5.) Ran /etc/init.d/oracleasm configure 6.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 7.) Shutdown node 2 8.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Error: asmtool: unable to clear device /dev/sdc1 9.) Unmounted OCFS2 filesystems 10.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 Success 11.) Ran /etc/init.d/oracleasm createdisk vol2 /dev/sdd1 & /dev/sde1 Success 12.) Remount OCFS2 filesystems: mount -a -t ocfs2 Error: ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" ocfs2_hb_ctl: Bad magic number in superblock while reading uuid mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: "Operation not permitted" 13.) /etc/init.d/oracleasm listdisks VOL1 VOL2 VOL3 SO...Are my OCFS2 partitions corrupted? Or can I not mount them once they're configured for ASM? Is this the correct behavior, or am I missing something? Thanks!! Chris Taylor Sr. Oracle DBA Unique Solutions www.unisolinc.com Office: 336-667-2447 xt. 2242 Cell: 336-262-5545 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://oss.oracle.com/pipermail/ocfs-users/attachments/20060804/ca8634c7/attachment.html
ASM is, in-effect, a userspace filesystem. So what you did is similar to if you had run mkfs.ext3 followed by mkfs.xfs on a device and then expected the device to still be in ext3 format. Chris Taylor wrote:> > Ok guys & gals here is the scenario: > > 1.) Host RHEL 4 U3 2.6.9-34.0.2.EL > 2.) OCFS2 latest version > 3.) Successfully formatted & mounted OCFS2 filesystems on 2 nodes > /dev/sdb1 /u02/oradata/usdev/voting > /dev/sdc1 /u02/oradata/usdev/data01 > /dev/sdd1 /u02/oradata/usdev/data02 > /dev/sde1 /u02/oradata/usdev/data03 > > 4.) Downloaded & installed ASMLib 2.0 on both nodes > 5.) Ran /etc/init.d/oracleasm configure > 6.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 > Error: asmtool: unable to clear device /dev/sdc1 > > 7.) Shutdown node 2 > 8.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 > Error: asmtool: unable to clear device /dev/sdc1 > > 9.) Unmounted OCFS2 filesystems > 10.) Ran /etc/init.d/oracleasm createdisk vol1 /dev/sdc1 > Success > 11.) Ran /etc/init.d/oracleasm createdisk vol2 /dev/sdd1 & /dev/sde1 > Success > 12.) Remount OCFS2 filesystems: mount -a -t ocfs2 > Error: > ocfs2_hb_ctl: Bad magic number in superblock while reading uuid > mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: > "Operation not permitted" > ocfs2_hb_ctl: Bad magic number in superblock while reading uuid > mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: > "Operation not permitted" > ocfs2_hb_ctl: Bad magic number in superblock while reading uuid > mount.ocfs2: Error when attempting to run /sbin/ocfs2_hb_ctl: > "Operation not permitted" > > 13.) /etc/init.d/oracleasm listdisks > VOL1 > VOL2 > VOL3 > > SO...Are my OCFS2 partitions corrupted? Or can I not mount them once > they're configured for ASM? > Is this the correct behavior, or am I missing something? > > Thanks!! > > > Chris Taylor > Sr. Oracle DBA > Unique Solutions > www.unisolinc.com > Office: 336-667-2447 xt. 2242 > Cell: 336-262-5545 > > ------------------------------------------------------------------------ > > _______________________________________________ > Ocfs-users mailing list > Ocfs-users@oss.oracle.com > http://oss.oracle.com/mailman/listinfo/ocfs-users >