Hi, Irecently setup a Oracle 10g RAC test environment with two HP/Compaq DL380s and a shared disk array. The array is presented as a single 136GB disk. The systems are running RedHat AS 3.0 Update 4. OCFS 1.0.13-PROD1 build: 5e87201061935f464cf6541c56eb2827 Oracle 10g 10.1.0.3.0 I was able to get everything working, had a database created and running an instance on each system. The problems started when I attempted to import an exported database. The import failed, then the OCFS filesystem seemed to have gone 'downhill' from there. The systems were rebooted, then the system refused to recognize the OCFS filesystem that was created. I had ORACLE_HOME on a local drive (/x00) then the shared drive had the cluster support files (Cluster Register and Voting File) along with the database files were on the (/x01) shared volume. These are some of the errors found in the syslog, there are no other operating system or hardware messages, only the OCFS errors. The module does load and 'lsmod | grep ocfs' produces: ocfs 299272 0 ---- SYSLOG Messages ---- Oracle Cluster FileSystem 1.0.13-PROD1 Tue Aug 10 11:31:03 PDT 2004 (build 5e87201061935f464cf6541c56eb2827) (1158) ERROR: Invalid volume signature, Common/ocfsgenmisc.c, 820 (1158) ERROR: Device (104,17) failed verification, Linux/ocfsmount.c, 233 (1158) ERROR: Error mounting volume, Linux/ocfsmain.c, 314 ------------------------ I had a similiar test using GFS from redhat, but that failed as soon as any appreciable load was put on the system. I was able to get farther with OCFS, but it seems that as soon as I put a decent sized load on it, the filesystem seemed to have failed. Anyone else experience this problem? Any ideas on how to recover? Or maybe I missed something when I set it up. Any help, comments, suggestions would be most appreciated. Thank you everyone in advance. Chris Steinke- chris.steinke@caltech.edu Senior UNIX systems administrator California Institute of Technology - Administrative Technology Center
what are you using for shared storage ? where was your export file ? On Sun, Jan 16, 2005 at 09:59:21AM -0800, Chris Steinke wrote:> Hi, > > Irecently setup a Oracle 10g RAC test environment with two HP/Compaq > DL380s and a shared disk array. The array is presented as a single 136GB > disk. > > The systems are running RedHat AS 3.0 Update 4. > OCFS 1.0.13-PROD1 build: 5e87201061935f464cf6541c56eb2827 > Oracle 10g 10.1.0.3.0 > > I was able to get everything working, had a database created and > running an instance on each system. > > The problems started when I attempted to import an exported database. > The import failed, then the OCFS filesystem seemed to have gone > 'downhill' from there. The systems were rebooted, then the system > refused to recognize the OCFS filesystem that was created. > > I had ORACLE_HOME on a local drive (/x00) then the shared drive had the > cluster support files (Cluster Register and Voting File) along with the > database files were on the (/x01) shared volume. > > These are some of the errors found in the syslog, there are no other > operating system or hardware messages, only the OCFS errors. The module > does load and 'lsmod | grep ocfs' produces: > > ocfs 299272 0 > > ---- SYSLOG Messages ---- > Oracle Cluster FileSystem 1.0.13-PROD1 Tue Aug 10 11:31:03 PDT 2004 > (build 5e87201061935f464cf6541c56eb2827) > (1158) ERROR: Invalid volume signature, Common/ocfsgenmisc.c, 820 > (1158) ERROR: Device (104,17) failed verification, Linux/ocfsmount.c, > 233 > (1158) ERROR: Error mounting volume, Linux/ocfsmain.c, 314 > ------------------------ > > I had a similiar test using GFS from redhat, but that failed as soon as > any appreciable load was put on the system. > > I was able to get farther with OCFS, but it seems that as soon as I put > a decent sized load on it, the filesystem seemed to have failed. > > Anyone else experience this problem? Any ideas on how to recover? Or > maybe I missed something when I set it up. > > Any help, comments, suggestions would be most appreciated. Thank you > everyone in advance. > > > > Chris Steinke- chris.steinke@caltech.edu > Senior UNIX systems administrator > California Institute of Technology - Administrative Technology Center > > > > > _______________________________________________ > Ocfs-users mailing list > Ocfs-users@oss.oracle.com > http://oss.oracle.com/mailman/listinfo/ocfs-users
The error listed concerns failiure to mount. Appears something overwrote atleast the first block on disk. email me the output of the following: dd if=/dev/shareddisk of=outfile bs=1024 count=3000 Remember to gzip it. :-) Lets see if the contents shed any light as to the extent of damage and possibly more. Sunil Chris Steinke wrote:>Hi, > >Irecently setup a Oracle 10g RAC test environment with two HP/Compaq >DL380s and a shared disk array. The array is presented as a single 136GB >disk. > >The systems are running RedHat AS 3.0 Update 4. >OCFS 1.0.13-PROD1 build: 5e87201061935f464cf6541c56eb2827 >Oracle 10g 10.1.0.3.0 > >I was able to get everything working, had a database created and >running an instance on each system. > >The problems started when I attempted to import an exported database. >The import failed, then the OCFS filesystem seemed to have gone >'downhill' from there. The systems were rebooted, then the system >refused to recognize the OCFS filesystem that was created. > >I had ORACLE_HOME on a local drive (/x00) then the shared drive had the >cluster support files (Cluster Register and Voting File) along with the >database files were on the (/x01) shared volume. > >These are some of the errors found in the syslog, there are no other >operating system or hardware messages, only the OCFS errors. The module >does load and 'lsmod | grep ocfs' produces: > >ocfs 299272 0 > >---- SYSLOG Messages ---- >Oracle Cluster FileSystem 1.0.13-PROD1 Tue Aug 10 11:31:03 PDT 2004 >(build 5e87201061935f464cf6541c56eb2827) >(1158) ERROR: Invalid volume signature, Common/ocfsgenmisc.c, 820 >(1158) ERROR: Device (104,17) failed verification, Linux/ocfsmount.c, >233 >(1158) ERROR: Error mounting volume, Linux/ocfsmain.c, 314 >------------------------ > >I had a similiar test using GFS from redhat, but that failed as soon as >any appreciable load was put on the system. > >I was able to get farther with OCFS, but it seems that as soon as I put >a decent sized load on it, the filesystem seemed to have failed. > >Anyone else experience this problem? Any ideas on how to recover? Or >maybe I missed something when I set it up. > >Any help, comments, suggestions would be most appreciated. Thank you >everyone in advance. > > > >Chris Steinke- chris.steinke@caltech.edu >Senior UNIX systems administrator >California Institute of Technology - Administrative Technology Center > > > > >_______________________________________________ >Ocfs-users mailing list >Ocfs-users@oss.oracle.com >http://oss.oracle.com/mailman/listinfo/ocfs-users > >
Maybe Matching Threads
- Odd errors while mounting an OCFS filesystem
- OCFS write/read across 2 nodes
- ocfs doesn't free space?
- [LLVMdev] Running "make check" on Windows yields lots of unwarrented unexpected failures (HTML free this time:-)
- How to sync /etc/smbpasswd with /etc/passwd ?