similar to: Odd errors while mounting an OCFS filesystem

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

Hi Experts, We have a 4nodes RAC running and recently one is down due to hardware (fibre optics card) failure. Since running on 3-nodes RAC, the surviving server just keep crashing. We cannot figure out why is this happening but checking /var/log/messages we have these error (notice the msg before crashing at 8:32): Dec 12 08:30:45 x335-142 kernel: (2) ERROR: file entry name did not match inode,

Hi, If I have ocfs on 2 nodes should I be able to do the following: t=0 Node A echo "1" > /ocfsmount/a Node B cat /ocfsmount/a >>>>Question 1: At this point should I see "1"? t=1 Node A echo "2" > /ocfsmount/a Node B cat /ocfsmount/a >>>>Question 2: At this point should I see

We are seeing these messages in our system log. any idea what they might be? Unfortunately I'm not sure what was happening at the time. Oct 26 12:57:39 houided003 kernel: (13996) ERROR: status = -999, Common/ocfsgendlm.c, 1304 Oct 26 12:57:39 houided003 kernel: (13996) ERROR: status = -999, Common/ocfsgendlm.c, 1501 Oct 26 12:57:39 houided003 kernel: (13996) ERROR: status = -999,

The src won't compile with gcc 2.96 or 3.23: /lib/modules/2.4.21-4.ELsmp/build/include/linux/bitops.h:47: warning: no previous prototype for `gen eric_fls' /lib/modules/2.4.21-4.ELsmp/build/include/linux/bitops.h:76: warning: no previous prototype for `get _bitmask_order' Linux/ocfsmain.c:113: warning: initialization from incompatible pointer type Linux/ocfsmain.c:1492: conflicting

Oracle appears to have Wim chained in the basement, forced to answer mailing list questions at all hours. I do appreciate it. Our cluster has been stable since we installed RAC, but a few minor issues have me concerned. First, our storage array seems to maintain continuous low-level activity even when the database is shut down. The CPUs spend a modest amount of time in iowait state while this

we are running OCFS on 2.4.21-15.0.4.ELsmp ocfs-2.4.21-EL-smp-1.0.12-1 ocfs-support-1.0.10-1 ocfs-tools-1.0.10-1 I have been deleting datafiles, more than 3 times successfully for rman duplication, from a mount point /data1 (total 191G). Last week when I tried to delete datafiles from the same directory, it did delete datafiles but did not release (reclaim) all space. It still showed that 20G of

Hi all, i have the following configuration: RHAS 2.1 2.4.9-e.40enterprise #1 SMP (2 node cluster). ocfs-support-1.1.2-1 ocfs-tools-1.1.2-1 ocfs-2.4.9-e-enterprise-1.0.12-1 i followed the users guide to install and configure ocfs. During definition and formatting of ocfs-partitions the ocfstool crashed. Now restarting ocfstool fails with the following error: ocfstool Abnormal termination!

What steps do I need to take to change a host's name? I modified the /etc/ocfs.conf file but ocfstool still reports the old name under the "Configured Nodes" tab. I'm also wondering where ocfstool stores the partition information that it shows under the device list. I dumped a couple ocfs partitions but they still show up in ocfstool. Thanks, Don

I am trying to mount the ocfs partitions using the following command Mount -t ocfs -o uid=oracle,gid=dba /dev/sda /ocfs01 as user oracle and group dba. However it mounts the volume as root. But if I use the ocfstool for the first and mount it as oracle:dba, the subsequent mounts using the above command line mounts the volume as oracle:dba. Is there something that I am missing or I will have

In installing 10g\RAC we hit an issue that has been indentified as a bug with CRS where the cluster locking files cannot be implemented under OCFS. Supposedly they work fine when implemented as Raw Devices. We needed to reclaim some space from existing LUN's to create the raw devices as we had expected to able to put all Oracle related files under OCFS (Yes, we belived the hype). This

I've created a few OCFS filesystems (RHEL 3, OCFS 1.0.9-12) while attempting to assign ownership by oracle:dba. I've done this both with ocfstool and mkfs.ocfs. They both seem to accept the parameters, but the resulting filesystem is always owned by root:root. This is not a fatal problem, since we can just throw all of our Oracle files into a properly-owned subdirectory in each

Hi, Before I describe the problem, I'll describe my environment. This is a development environment built over 1.5 yrs ago: * We have a 2 node Oracle RAC Cluster. * built on two Dell PowerEdge 1850 Servers (Intel(R) Xeon(TM) CPU 3.20GHz) * OS is RHEL3 U3 * Kernel 2.4.21-20.ELsmp * these 2 servers are attached to an EMC AX100 SAN via fibre channel * HBA (QLogic QLA6312 PCI to Fibre

Hi all! Doing my first steps with OCFS (1.0.9), I ran across a nifty little "feature".... We've been trying if DMP with QLogic 2300 HBAs works without having DMP activated (ok, blame on me...). ocfstool allows you to mount different partitions on the same mountpoint, but after we tried that, everything went to state "D" and we chose to reboot the whole cluster (4

We finally seem to have worked most of the kinks out of our OCFS/RAC setup, and are running a mass import to clone our production database for testing. While this has been going on, I fired up ocfstool to check out the states of the various shared volumes. In several cases, I saw (in the File Listing tab) multiple entries for some directories and files at the filesystem root. The duplicate

Had a problem starting Oracle after expanding an EMC Metalun. We get the following errors: >WARNING: OemInit2: Opened file(/oradata/dbf/quorum.dbf 8), tid = main:1024 file = oem.c, line = 491 {Sun Apr 17 10:33:41 2005 } >ERROR: ReadOthersDskInfo(): ReadFile(/oradata/dbf/quorum.dbf) failed(5) - (0) bytes read, tid = main:1024 file = oem.c, line = 1396 {Sun Apr 17 10:33:41 2005 }

hi, i have a problem with ocfs. i did some tests with dbms_stats on a 2-node RAC and got the following error: Thu Feb 17 15:36:31 2005 KCF: write/open error block=0x7800 online=1 file=13 /db/GKEZ/tablespaces/voeb/idx01_kez.dbf error=27072 txt: 'Linux Error: 5: Input/output error Additional information: 30719' Automatic datafile offline due to write error on file 13:

I'm trying to compile ocfs 1.0.9 and when I execute ./configure I receive the following error msg: *** the gtk-config script installed by GTK could not be found *** If GTK wall installed in PREFIX, make sure PREFIX/bin is in *** your path, or set the GTK_CONFIG environmet variable to the *** full path to gtk-config configure: warning GTK+ not found, ocfstool will not be be built when I run

Looking for any ideas where I need to look to fix this: I'm installing RHEL3 AS (update 4) on Dell PowerEdge 6850's. I've installed the hugemem kernels on these boxes and need to install and run ocfs. Kernel: ------- 2.4.21-27.0.4.ELhugemem Loaded the ocfs rpm's --------------------- # rpm -qa | grep ocfs ocfs-2.4.21-EL-smp-1.0.14-1 ocfs-support-1.1.5-1 ocfs-2.4.21-EL-1.0.14-1

Hi, I'm setting up ocfs on opensuse 10.3 with kernel 2.6.22.5-31 . Appreciate if you could guide on the solution to this error. I have a secondary hard disk (80G) created raw paritions primary /dev/sdb1 - 12G extended /dev/sdb2 - 12G comprising of /dev/sdb5 - 4G /dev/sdb6 - 3G /dev/sdb7 - 5G The error is as follows: Could not get a raw device slot for disk access. Please free up some