OCFS2 supported on RHEL3.0?
Does anyone know if OCFS2 is supported on RHEL3.0? The documentation in OTN shows that is supported on RHEL4.0 but doesn't seem to mention RHEL3.0 anywhere.
Thanks!
- Tim
Hi, Tim.
Unfortunatelly, no. You may find this and other useful information at "OCFS2 - Frequently Asked Questions" (http://oss.oracle.com/projects/ocfs2/dist/documentation/ocfs2_faq.txt). This is the answer for your question:
"Q02 Can OCFS volumes and OCFS2 volumes be mounted on the same machine
simultaneously?
A02 No. OCFS only works on 2.4 linux kernels (Red Hat's AS2.1/EL3 and SuSE's
SLES8). OCFS2, on the other hand, only works on the 2.6 kernels
(Red Hat's EL4 and SuSE's SLES9)."
Regards,
Danilo
Similar Messages
-
JRockit 1.4.2_03 Linux64 on AS3.0 -Xmx -Xms error!
Hi,
I download the JRockit 1.4.2_03 SDK for Linux(64 bit) from the "http://commerce.bea.com/showallversions.jsp?family=WLJR"
When I run Java with -Xms -Xmx options on AS EL3.0, whatever heap size, the error
messages are thrown out:
Negative or invalid size for maximum heap size - reset to default!
Negative or invalid size for initial heap size - reset to default!
When run on AS2.1, it works.
I don't know why. JRockit 1.4.2_03 doesn't support AS EL3.0?
Thx,
XuekunHmm, that doc page must be a mistake. We are working on getting
NPTL-support on RHEL3.0 into the next service pack of JRockit 1.4.2. I'm
very sorry that the documentation suggests otherwise.
Thanks,
/Staffan
Xuekun Hu wrote:
Yes. By setting LD_ASSUME_KERNEL=2.4, it works. But in http://e-docs.bea.com/wljrockit/docs142/intro/newftrs.html,
it says JRockit 1.4.2_03 support NPTL at Red Hat Enterprise Linux 3.
And I just want to test the JVM performance using NPTL.
The command line is very simple. java -Xms1g -Xmx1g
Thx,
Xuekun
Staffan Larsen <[email protected]> wrote:
This is a confirmed problem on RHEL3.0 and does not exist on any of the
supported platforms. Note that RHEL3.0 is not on the list of supported
platforms: http://e-docs.bea.com/wljrockit/docs142/certif.html
Running without NPTL (by setting LD_ASSUME_KERNEL=2.4) should work
better, though.
Regards,
/Staffan
Xuekun Hu wrote:
Hi,
I download the JRockit 1.4.2_03 SDK for Linux(64 bit) from the "http://commerce.bea.com/showallversions.jsp?family=WLJR"
When I run Java with -Xms -Xmx options on AS EL3.0, whatever heap size,the error
messages are thrown out:
Negative or invalid size for maximum heap size - reset to default!
Negative or invalid size for initial heap size - reset to default!
When run on AS2.1, it works.
I don't know why. JRockit 1.4.2_03 doesn't support AS EL3.0?
Thx,
Xuekun -
MQ/WebLogic Topology and Architecture
I am new to MQ and am trying to get my head around a couple of basic "big
picture" things; I'll explain what I think is going on and hopefully someone
will jump in when I'm off in the weeds. I need to do is integrate a J2EE
application with a legacy MQ system and as you might expect, I cannot alter
that system, I must accommodate it (which J2EE is designed to do).
MQ is designed to be distributed; the idea that you'll install an MQ Server
on every Application Server machines seems like a special case. In my case,
it's academic.
First MQ itself:
There appears to be some internal and proprietary things MQ needs to do for
efficiency and possibly historical reasons. To achieve that, you must install
an MQ Client on each machine so that access appears to be local for Applications
that talk to MQ. I have written some code that talks directly to the server
using some of the Client libraries and I was able to send but could not get
messages off the queue; I found some documentation that said that this would
be the case in some circumstances. I suspect that the reason for this is
that I may have not used the "correct" client libraries since I found some
example code that doesn't yet run (we can ignore that for now).
The Client install does give you Java classes that provide connectivity and
it implies JMS support but in what form, since the classes appear to use
proprietary code?
Does the Client have the ability to run as a JMS Server that appears to be
a local JMS nexus that I can talk to with standard JNDI lookups?
WebLogic J2EE Integration:
The ultimate goal here is to integrate an existing MQ installation so I can
send and recieve from a J2EE Application I am working on.
I have been reading that there's at least three different ways to make this
connection.
1) Roll your own using the proprietary lib clients in a startup class
2) Using WebLogic's Bridge technology
3) Use WebLogic's Foreign Destination configuration.
Detailed questions below by item number:
1) -> I understand that directly using IBM's client libraries is supposed
to function the same as running the Client locally, that doesn't sound like
it would give the scalability of pooling and that solution won't route to
drive MDBs.
2) -> A lot of posts state that for WL/MQ integration this is not necessary.
I don't know a lot about the specifics of what I will need to do to talk
to this other system yet, I am just trying to get my head around the general
cases.
3) -> This is the most appealing to me since it sounds like it is a matter
of using the WL console to set up the JMS destination once I have the correct
libraries in the server's classpath. It would seem that the JMS queues look
like they are local, I can drive MDBs from this with all the manageability
built in. However, does the server use the same provider factory class as
I would in my own code or are there special considerations?
Comments and pointing to specific reading encouraged (don't just say, to
read the 677 page Java MQ docs).
ThanksWow! All I can say is "Thanks", it's gonna take we a while to get through
all this but I do appreciate the help. I tried to get the Client stuff installed
on my Linux dev box but it's not clear which rpms are needed and some complained
because I am still running RH7.3. I am in the process of moving to a supported
platform RHEL3.x or above... stay tuned.
Hello Tom,
> Hi!
>
> If you haven't already done it, the best place to start is to read
> through the entire "Integrating Remote JMS Providers" FAQ:
>
> http://e-docs.bea.com/wls/docs81/faq/interop.html
>
> I'm also appending my own integration notes which mention MQ (and
> which you may have already run across). I update these notes from
> time to time. Enjoy!
>
> Tom, BEA
>
> JMS Integration of Foreign Vendors with BEA WebLogic Server
> -----------------------------------------------------------
>
> The following notes are derived mostly from
> "http://dev2dev.bea.com/technologies/jms/index.jsp".
>
> For additional questions, a good forum for WebLogic questions in
> general is "newsgroups.bea.com". These can be mined for information by
> using Google's newsgroup search function.
>
> JMS Integration Overview
> ------------------------
> - For integration with "non-Java" and/or "non-JMS" platforms, see
> "Non-Java Integration Options" below.
>
> - For a foreign JMS vendor to participate in a WL transaction it must
> support XA. Specifically, it must support the javax.jms.XA*
> interfaces.
>
> - In WL versions 6.0 and up it is possible to make synchronous calls
> to foreign JMS vendors participate in a WL transaction as long as the
> foreign vendor supports XA.
>
> - WL 6.0 and 6.1 MDBs can be driven by foreign vendors
> non-transactionally. They can be driven transactionally by a select
> few foreign vendors (MQ is not part of the select few)
>
> - WL 7.0 and later, MDBs can be driven by foreign vendors
> transactionally and non-transationally.
>
> - WL 6.1 and later WL provides a messaging bridge feature. Messaging
> bridges forward messages between any two JMS destinations, including
> foreign destinations, and can transfer messages transactionally or
> non-transactionally.
>
> - WL 8.1 JMS provides additional features that simplify transactional
> and JNDI integration of foreign vendors. See
> http://edocs.bea.com/wls/docs81/jms/intro.html#jms_features and
> http://e-docs.bea.com/wls/docs81/faq/interop.html
>
> Integration with 8.1 Details
> ----------------------------
> To start, first read the "Integrating Remote JMS Providers FAQ"
> (released in Dec 2004) at:
> http://e-docs.bea.com/wls/docs81/faq/interop.html
>
> A good overview of 8.1 JMS interop capability is the presentation
> "Integrating Foreign JMS Providers with BEA WebLogic Server" here:
>
> http://www.bea.com/content/files/eworld/presentations/Wed_03_05_03/App
> lication_Servers/1097-Foreign_JMS_Providers_WLS.pdf
>
> This document refers to helpful new 8.1 features, which simplify
> integration. These include:
> http://edocs.bea.com/wls/docs81/ConsoleHelp/jms_config.html#accessing_
> foreign_providers
> http://edocs.bea.com/wls/docs81/jms/j2ee_components.html#1033768
> And are also summarized here (under interoperability):
> http://edocs.bea.com/wls/docs81/jms/intro.html#jms_features
> Also read the MDB documentation, which extensively covers integrating
> foreign vendors:
>
> http://edocs.bea.com/wls/docs81/ejb/message_beans.html
>
> The 8.1 features are likely sufficient for most 8.1 integration needs,
> but you may want to refer to the "Using Foreign JMS Providers With
> WLS" white-paper mentioned below, which is 7.0 specific but contains
> specific examples of configuring non-WebLogic JMS vendors. See also
> notes on "MQ" below.
>
> Integration with 6.1 and 7.0 Details
> ------------------------------------
> Read the "Using Foreign JMS Providers With WLS" white-paper:
>
> http://dev2dev.bea.com/products/wlserver/whitepapers/jmsproviders.jsp
>
> Note that this white-paper does not take into account 8.1 features.
>
> For 7.0 read the extensive 8.1 MDB documentation, which largely also
> applies to 7.0:
>
> http://edocs.bea.com/wls/docs81/ejb/message_beans.html
>
> Non-Java Integration Options
> ----------------------------
> - WL JMS has a JNI based C client which is available for Windows and
> some UNIX platforms. This C client supports 7.0 and up, and will be
> officially packaged with WLS in 9.0 (virtually unchanged). The C API
> is currently only supported through the jms newsgroup. See "JMS C
> API", here:
>
> http://dev2dev.bea.com/technologies/jms/index.jsp
>
> - WL supports direct Windows COM access through its "JCOM" feature.
> This doesn't include the JMS API, but one can invoke EJBs which in
> turn invoke JMS. See
>
> http://e-docs.bea.com/wls/docs61/jcom.html
>
> http://e-docs.bea.com/wls/docs70/jcom/
>
> http://e-docs.bea.com/wls/docs81/jcom/
>
> - Similar to JCOM, but more advanced and supported on more platforms,
> WL supports access via the standard IIOP protocol. You can use the
> BEA Tuxedo C client for this purpose (no license fee). This doesn't
> include the JMS API, but one can invoke EJBs which in turn invoke JMS.
> See
> http://e-docs.bea.com/wls/docs81/rmi_iiop/
> http://e-docs.bea.com/wls/docs70/rmi_iiop/
> http://e-docs.bea.com/wls/docs61/rmi_iiop/
> Unlike most other approaches, the IIOP client approach also allows the
> client to begin and commit user (JTA) transactions (not configured).
> - If you already have a BEA Tuxedo license, one option is communicate
> through BEA Tuxedo (which has various APIs on Windows) and configure a
> WebLogic Server to respond to these requests via the WTC bridge.
> Search for "WTC" in the BEA docs. Unlike most other approaches, the
> Tuxedo API approach also allows the client to begin and commit user
> (JTA) transactions.
>
> - Another approach is to interop via web-service standards. Or even to
> simply to invoke a servlet on the WL server using a basic HTTP call
> from the client. These operation in turn can invoke the JMS API.
> There is a white-paper on "Interoperability Study of BEA WebLogic
> Workshop 8.1 and Microsoft .NET 1.1 Web Services", that demonstrates
> web-services here:
> http://ftpna2.bea.com/pub/downloads/WebLogic-DotNet-Interop.pdf
>
> - Yet another approach is to use a third party product that is
> designed to wrap any JMS vendor. There are even open source
> versions. In no particular order, here are some examples: Open3
> WinJMS, CodeMesh, Active JMS, SpiritSoft
>
> - Finally, there are .NET/C/C++ integration libraries that are not
> specific to JMS, some examples are JNBridge, Jace, and CodeMesh.
>
> Notes on MQ Remote Capable XA Clients
> -------------------------------------
> Until recently, IBM MQ JMS clients could not work transactionally
> unless they were running on the same host as their MQ server. This is
> a limitation unique to MQ that was relaxed with the introduction of
> IBM's new "WebSphere MQ Extended Transactional Client". See:
>
> http://publibfp.boulder.ibm.com/epubs/pdf/csqzar00.pdf
>
> The product is new, and for some reason, configuration of this client
> seems to be tricky, even when WebLogic is not involved at all. Oddly,
> the main sticking point seems to be simply making sure that class
> paths refer to the required IBM jars:
>
> - Required on WLS where MQ objects are bound into JNDI:
> com.ibm.mq.jar, com.ibm.mqjms.jar
>
> - Required only if MQ objects are bound into JNDI on a different
> server: com.ibm.mq.jar
>
> If there are problems when using this client, first get it to work
> using a pure IBM client without any BEA classes involved. Once that
> is working, search the WL JMS newsgroup for answers and/or contact BEA
> customer support.
>
> Notes on Oracle AQ Integration
> ------------------------------
> If problems are encountered integrating Oracle's built-in queuing
> (Oracle AQ) JMS client, there is publicly available wrapper code that
> can aid integrating AQ directly into MDBs, JMS, or the messaging
> bridge. The solution is titled "Startup class to bind
> AQ/Referenceable objects to WLS JNDI", is not supported by BEA, and is
> posted to:
>
> http://dev2dev.bea.com/codelibrary/code/startupclass.jsp (older
> version) http://xa-compliant-oracleaq.projects.dev2dev.bea.com (newer
> version)
>
> Caveats:
>
> It may be that the solution doesn't directly support concurrent
> consumers. Perhaps Oracle requires that concurrent consumers each
> have a unique JMS connection? As a work-around, parallel message
> processing can be achieved indirectly by forwarding AQ messages into a
> WL JMS destination - which do support concurrent processing.
>
> Up-to-date versions of Oracle may be required. For more information,
> google search the weblogic.developer.interest.jms newsgroup for
> "Oracle" and "AQ".
>
> MDB Thread Pool Notes
> ---------------------
> WL7.0SP? and WL8.1 and later support the "dispatch-policy" field to
> specify which thread pool an MDB uses to run its instances. In most
> cases this field should be configured to help address potential
> performance issues and/or dead-locks:
>
> http://edocs.bea.com/wls/docs81/ejb/DDreference-ejb-jar.html#dispatch-
> policy
>
> (Note that "dispatch-policy" is ignored for non-transactional foreign
> vendors; in this case, the MDB "onMessage" callback runs in the
> foreign vendor's thread.)
>
> MDB Concurrency Notes
> ---------------------
> Queue MDBs driven by foreign providers can run multiple instances
> concurrently. Topic MDBs driven by foreign providers are limited to
> one instance (not sure, but transactional foreign driven topic MDBs
> may not have this limitation). The size of the thread pool that the
> MDB runs in and the "max-beans-in-free-pool" descriptor limit how many
> instances run concurrently.
>
> Design Guide-Lines and Performance Tuning Notes
> -----------------------------------------------
> The "WebLogic JMS Performance Guide" white-paper contains detailed
> design, performance, and tuning information for Clustering, Messaging
> Bridge, JMS, and MDBs.
>
> http://dev2dev.bea.com/products/wlserver/whitepapers/WL_JMS_Perform_GD
> .jsp
> -
R12 Upgrade Path Advice (RHEL 3)
Dear All,
I would like you to provide me some input on R12 upgrade.
Our current setup is
OS: Red Hat Linux AS 3.0 32-bits
EBS: 11.5.10.2
DB: 10.2.0.4
New setup
OS: Red Hat Linux 6.x 64-bits
EBS: 12.1.3
DB: 11.2.0.3
I have searched various notes, but can not find anything specific to it RHEL 3.0.
R12 is not supported on RHEL3.0, so I can't upgrade it on current system and then migrate it to RHEL6 system using note 471566.1.
Can you please advice me what options we have here ?.
Thanks in advance
RegardsOur current setup is
OS: Red Hat Linux AS 3.0 32-bits
EBS: 11.5.10.2
DB: 10.2.0.4
New setup
OS: Red Hat Linux 6.x 64-bits
EBS: 12.1.3
DB: 11.2.0.3
I have searched various notes, but can not find anything specific to it RHEL 3.0.
R12 is not supported on RHEL3.0, so I can't upgrade it on current system and then migrate it to RHEL6 system using note 471566.1.
Can you please advice me what options we have here ?.You do not need RHEL 3.0 and you just need to lay down R12 on RHEL 6.x 64-bit and do the upgrade from that node. The 11i node is not used during the upgrade, and there is no point of installing R12 on an OS and migrate it to a different OS.
Oracle E-Business Suite Installation and Upgrade Notes Release 12 (12.1.1) for Linux x86-64 [ID 761566.1]
Oracle E-Business Suite Upgrades and Platform Migration [ID 1377213.1]
Interoperability Notes Oracle EBS 11i with Oracle Database 11gR2 (11.2.0) [ID 881505.1]
Oracle E-Business Suite Release 12.1.3 Now Available
https://blogs.oracle.com/stevenChan/entry/ebs_1213_available
Thanks,
Hussein -
OCFS Linux Distribution confusion and planned production release
Hello, as far as I know are Oracle10g and Oracle9iR2.
supportet for SLES9. There is no support for RHEL4.
But up to now OCFS2 is developed for RHEL4.
Now I've 2 questions:
1. Will OCFS2 supported for SLES9?
2. When it's planned to release the production version of OCFS2?You are asking realy hard questions.
And I don't think there is a person who is able to give you specific answer. (maybe Mark Fasheh or 2.6 kernel maintener)
Anyway there was OCFS2 added to linux kernel (2.6.12-rc5-mm1) but it is not still stable version but it seems good and I hope it will become stable and included in stable linux kernel soon.
When the OCFS2 wil be included in stable version of kernel then all new distributions will be support it. And also Companies (such as Redhat and Suse) could backport the feature for old kernels and for old distributions.
Till (I sugest) use OCFS(1) on production systems. -
We are planning to build 2 RHEL 5.5 (64bit) Servers with EMC SAN LUNs shared between them and we are going to use OCFS instead of ASM to share the disks for our application Oracle RAC 11g.
As per our design we need to have LUNs upto the size of 24TB From EMC SAN. In RHEL 5.5 with ext3 and ext4 the maximum supported filessytem size is 16TB. I want to know, what is the maximum LUN/filesystem size supported by OCFS ?
If I had to do a RAC with ASM disks, I know that we can present smaller LUNs of size 3000GBx8=24TB and the RAC will stipe across them and form a large usable space with in RAC. But in OCFS how do I do that? Can I present multiple LUNS of small size and form a bigger LUNs out of it inside OCFS? The space we are planing is 12 TB for Database files, 300 GB Redo Logs, 8 TB Flash Recovery, 24 TB of Backup Space to backup the Database files. How do I break down the LUN NUmbers for each of these? I can make a LUN as large as 64TB out of EMC Storage so there is no problem there.
Thanks for your help in advance.I want to know, what is the maximum LUN/filesystem size supported by OCFS ?http://www.oracle.com/us/technologies/linux/025995.htm
http://oss.oracle.com/projects/ocfs2/dist/documentation/v1.6/ocfs2-1_6-usersguide.pdf
OCFS2 supports cluster sizes of 4KB, 8KB, 16KB, 32KB, 64KB, 128KB, 256KB, 512KB and 1MB.
A file system with 1MB cluster size can go up to *4PB*, while a file system with a 4KB cluster size can address up to 16TB. -
OCFS2 - Any idea when it will be supported?
Hi,
I am at the beginning stages of a 11.5.10 development effort. We are building our dev hardware environment with the intentions of mimicking prod. The problem I have is clustering/RAC and what to do about the shared file system. Right now, OCFS2 is not supported for production. This would somewhat force our hand to look at Red Hat GFS for the shared file system (expensive - $2100 per node) or something similar, since we are planning a shared $ORACLE_HOME and $APPL_TOP.
Does anyone have an idea when OCFS2 will be supported? Are there other alternatives to Red Hat GFS?
Thanks kindly,
BJ
PS: Raw devices not an option.That certainly does - a bunch!
Go live on this project is not until November 1 2005, so I can start development with beta OCFS2, when OCFS2 goes GA I'll just backup everything and rebuild the file system with the supported version.
Thanks for checking it out, very much appreciated!
BJ -
What is the Oracle supported method for backing up OCFS2?
Hi
My customer has been trying to backup Oracle Cluster File System (ver 2) with netbackup 6.5, but not via RMAN.
They want to know if this method of backing OCFS up is suppported?
But I think the supported method has to use a netbackup Oracle agent and RMAN
Can anybody confirm this please?
Regards
jethro_pThe recommended method is thru RMAN only. Directly backing up OCFS thru netbackup is not a reliable backup.
We faced an critical oracle error & the database went down. We are not able to bring the database with the direct backup thru veritas. Then we restored the database only thru rman backup. -
I have a x86_64 red hat Rel 4 update 3 Installation where I am trying to get OCFS2 working. It is mostly working but I am unable to get it to mount by Label.
I keep getting the error "No Such partition" and was wondering if any one has run into this before or if there is a mistake on what I am doing.
Here is the exact commands and errors I would get:
[root@xxxxxx 10.2.0.1]# uname -a
Linux xxxxxx .xxxxxx .xxxxxx 2.6.9-34.0.1.ELsmp #1 SMP Wed May 17 16:59:36 EDT 2006 x86_64 x86_64 x86_64 GNU/Linux
[root@xxxxxx 10.2.0.1]# df -k /ocfs*
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb 15728640 401024 15327616 3% /ocfs1
/dev/sdc 15728640 400896 15327744 3% /ocfs2
/dev/sdj 20971520 532224 20439296 3% /ocfs3
[root@xxxxxx 10.2.0.1]# mounted.ocfs2 -d
Device FS UUID Label
/dev/sdb ocfs2 76504271-6be6-4c3d-adfc-6aed537d3135 ClusterFiles
/dev/sdc ocfs2 46f87ea8-0994-4341-afcb-ad4cabbeecb8 OraDumps
/dev/sdj ocfs2 e49af5ca-5bdb-4436-a120-640e9febe4b6 DiskShare
[root@xxxxxx 10.2.0.1]# umount /ocfs3
[root@xxxxxx 10.2.0.1]# mount -L "DiskShare" /ocfs3
mount: no such partition found
[root@xxxxxx 10.2.0.1]# mount /ocfs3
[root@xxxxxx 10.2.0.1]# df -k /ocfs*
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/sdb 15728640 401024 15327616 3% /ocfs1
/dev/sdc 15728640 400896 15327744 3% /ocfs2
/dev/sdj 20971520 532224 20439296 3% /ocfs3I encountered this issue also i.e. not being able to mount an OCFS2 filesystem by label.
I was able to resolve this issue by updating the util-linux package on my Red Hat system. I had a fresh install of RHEL 4 release 4 and the version of util-linux that was present on my system was:
# rpm -qa | grep util-linux
util-linux-2.12a-16.EL4.6
So I grabbed the latest version of util-linux from Red Hat and updated the util-linux package. Now I am able to mount a OCFS2 filesystem by label.
So the version of util-linux on my system is now:
# rpm -qa | grep util-linux
util-linux-2.12a-16.EL4.20
Apparently, Oracle provided a patch for mounting OCFS2 by label which is included in this release of util-linux.
I hope this works for you. -
11.2.0.3 grid installation fails while selecting OCFS2 for ocr files
We are installing 11GR2 (11.2.0.3) cluster on a 64 bit system. we have OCFS2 filesystem for shared devices. version 1.6.3.
While selecting ocr file locations , we get the following error
[INS-41321] Invalid oracle cluster register [OCR] Location
Cause- The installer detects that the storage type of location is not supported for Oracle Cluster registery
Action - Provide a supported storage location for the Oracle Cluster Registry
Additional information
/crp2db01/OCR/ocr_1 is not shared
However , this mountpoint is shared across both the nodes.
Note: 11201 grid installation was successful and it accepted the above locations for OCR. however ,we need 11.2.0.3 cluster for 11.2.0.3 databaseAs for your current problem, just because Oracle "allows" OCFS2 in a GRID environment, I would never suggest nor implement that. It adds a layer of complexity that is totally unnecessary when a GRID/ASM implementation performs circles around OCFS2. ASM is much easier to manage, maintain, expand and shrink than OCFS2. Especially at version 11.2.0.3. When working at a large telco a few years ago, we had a 300TB+ ASM environment. OCFS2 could not even begin to be that big. ASM will provide you a MUCH more stable environment than OCFS2. And with ASM there is a lot of "magic" that happens with OCR/Voting that makes your life MUCH easier. If you "require" shared application files, then use ASM/ACFS. It is a much better "volume manager" than OCFS2.
Since you must present devices to the system for OCFS2, you should not have any problems doing the same for ASM. (and don't use ASMLib as it is going away and is not necessary - just make sure you use a partition that skips the first 1M (usually cylinder 1) and you should be good to go!)
I also would not use a "shared ORACLE_HOME" on either ACFS or OCFS2. The biggest reason is that you lose the ability to do a "rolling" upgrade and when you have a VLC, that becomes much more important that saving a few GB worth of storage.
I would also pay attention to this:
http://docs.oracle.com/cd/E11882_01/install.112/e22489/storage.htm#CDEDAHGB
3.1.4.2 General Storage Considerations for Oracle RAC
Use the following guidelines when choosing the storage options to use for each file type:
You can choose any combination of the supported storage options for each file type provided that you satisfy all requirements listed for the chosen storage options.
If you plan to install an Oracle RAC home on a shared OCFS2 location, then you must upgrade OCFS2 to at least version 1.4.1, which supports shared writable mmaps.
Oracle recommends that you choose Oracle ASM as the storage option for database and recovery files.
For Standard Edition Oracle RAC installations, Oracle ASM is the only supported storage option for database or recovery files. -
Problem about use OCFS2 install clusterware
hi,all :
i set two disk use OCFS2, one for OCR, another for voting disk,i format the two disk on rac1 ,and mout on rac1 and rac2
[root@rac2 ocfs2]# mkfs.ocfs2 -b 4K -C 32K -N 2 -L ora_votingdisk /dev/sdb1
[root@rac2 ocfs2]# mkdir /opt/votingdisk
[root@rac2 ocfs2]# chown oracle.oinstall /opt/votingdisk/
[root@rac2 ocfs2]# mount -t ocfs2 /dev/sdb1 /opt/votingdisk/
[root@rac2 votingdisk]# mkfs.ocfs2 -b 4K -C 32K -N 2 -L ora_OCR /dev/sdc1
[root@rac2 votingdisk]# mkdir /opt/OCRfile
[root@rac2 votingdisk]# chown oracle.oinstall /opt/OCRfile
[root@rac2 votingdisk]# mount -t ocfs2 /dev/sdc1 /opt/OCRfile/and i use response file to installl crs
n_storageTypeOCR=2
s_ocrpartitionlocation="/opt/OCRfile"
s_ocrMirrorLocation=""
n_storageTypeVDSK=2
s_votingdisklocation="/opt/votingdisk"and install successed,but when i run root.sh ,both fail on each node,and here is error info,i do not know why?
[root@rac2 opt]# /opt/ora/product/10.2.0/crs_1/root.sh
WARNING: directory '/opt/ora/product/10.2.0' is not owned by root
WARNING: directory '/opt/ora/product' is not owned by root
WARNING: directory '/opt/ora' is not owned by root
Checking to see if Oracle CRS stack is already configured
/etc/oracle does not exist. Creating it now.
Setting the permissions on OCR backup directory
Setting up NS directories
PROT-1: Failed to initialize ocrconfig
Failed to upgrade Oracle Cluster Registry configurationand i checked the log output
[root@rac2 client]# cat ocrconfig_8035.log
Oracle Database 10g CRS Release 10.2.0.1.0 Production Copyright 1996, 2005 Oracle. All rights reserved.
2011-07-29 23:26:53.612: [ OCRCONF][5408976]ocrconfig starts...
2011-07-29 23:26:53.613: [ OCRCONF][5408976]Upgrading OCR data
2011-07-29 23:26:53.614: [ OCROSD][5408976]utopen:6'': OCR location /opt/OCRfile configured is not valid storage type. Return code [37].
2011-07-29 23:26:53.614: [ OCRRAW][5408976]proprinit: Could not open raw device
2011-07-29 23:26:53.614: [ default][5408976]a_init:7!: Backend init unsuccessful : [37]
2011-07-29 23:26:53.615: [ OCRCONF][5408976]Exporting OCR data to [OCRUPGRADEFILE]
2011-07-29 23:26:53.615: [ OCRAPI][5408976]a_init:7!: Backend init unsuccessful : [33]
2011-07-29 23:26:53.615: [ OCRCONF][5408976]There was no previous version of OCR. error:[PROC-33: Oracle Cluster Registry is not configured]
2011-07-29 23:26:53.615: [ OCROSD][5408976]utopen:6'': OCR location /opt/OCRfile configured is not valid storage type. Return code [37].
2011-07-29 23:26:53.615: [ OCRRAW][5408976]proprinit: Could not open raw device
2011-07-29 23:26:53.615: [ default][5408976]a_init:7!: Backend init unsuccessful : [37]
2011-07-29 23:26:53.616: [ OCROSD][5408976]utopen:6'': OCR location /opt/OCRfile configured is not valid storage type. Return code [37].
2011-07-29 23:26:53.616: [ OCRRAW][5408976]proprinit: Could not open raw device
2011-07-29 23:26:53.616: [ OCRAPI][5408976]a_init:6b!: Backend init unsuccessful : [37]
2011-07-29 23:26:53.616: [ OCRCONF][5408976]Failed to initialized OCR context. error:[PROC-37: Oracle Cluster Registry does not support the storage type configured]
2011-07-29 23:26:53.617: [ OCRCONF][5408976]Exiting [status=failed]...Edited by: 859340 on 2011-7-29 上午12:59instead this dot ( . ) on oracle.oinstall
[root@rac2 votingdisk]# chown oracle.oinstall /opt/OCRfileuse colon ( : )
[root@rac2 votingdisk]# chown oracle:oinstall /opt/OCRfile -
Ok this question may be a little complicated so I'll start by explaining what I'm trying to achive.
I have 3 OVM 2.2.1 servers in 1 Server pool running OCFS2 on an EMC SAN.
I also ave an Oracle Enterprise Linux server which connects to the same ocfs2 cluster for performing backups of the VMs.
Currently I have to pause all VMS and backup the shared storage in order to backup all of my VMs.
I would like to use reflink so I can do this without pausing them.
Can I achieve this by
1 upgrading all OVM servers and my OEL Server to OCFS2 1.6 or is the Kernel upgrade not supported in OVM 2.2.1
or
2 Upgrading my OEL server to OCFS2 1.6 and doing the reflinks there, but leave the OVM machines on 1.4 as the doco states the 1.4 and 1.6 are compatible with one another.
Thanks in advanceHi,
unfortunately OCFS2 1.6 is not yet supported on OVM 2.2.1.
I am waiting all day that Oracle will finally release OVM 3.0, which will incorporate the new OCFS2 1.6 with all it functions.
This would also resolv your problem then...
Regarding your second option, I don't know if this will work. But it seems the only solution you have right now...
Sebastian -
Mount ocfs2 error in oracle 10g installation on vmware
Dear All this is very urgent.....kindly help me...
i m running this command for mount ocfs folder but not run gives error below.......
[root@rac1 ~]#mount -t ocfs2 -o datavolume,nointr /dev/sda1 /ocfs
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"
--------I have run this command for checking----------
$ mounted.ocfs2 -d
$ mounted.ocfs2 -f
$ cat /etc/ocfs2/cluster.conf
$ /etc/rc.d/init.d/ocfs2 status [ as root ]
---------the output is below--------------------------------
--------From node 1----------------------------------------------
[root@rac1 ~]# mounted.ocfs2 -d
Device FS UUID Label
[root@rac1 ~]# mounted.ocfs2 -f
Device FS Nodes
[root@rac1 ~]# cat /etc/ocfs2/cluster.conf
node:
ip_port = 7777
ip_address = 172.16.31.195
number = 0
name = rac1
cluster = ocfs2
node:
ip_port = 7777
ip_address = 172.16.31.197
number = 1
name = rac2
cluster = ocfs2
cluster:
node_count = 2
name = ocfs2
[root@rac1 ~]#
[root@rac1 ~]# /etc/rc.d/init.d/ocfs2 status
[root@rac1 ~]#
----------------------from node 2----------------------------------
output from node2
[root@rac2 ~]# su -
[root@rac2 ~]# mounted.ocfs2 -d
-bash: mounted.ocfs2: command not found
[root@rac2 ~]# mounted.ocfs2 -f
-bash: mounted.ocfs2: command not found
[root@rac2 ~]# cat /etc/ocfs2/cluster.conf
node:
ip_port = 7777
ip_address = 172.16.31.195
number = 0
name = rac1
cluster = ocfs2
node:
ip_port = 7777
ip_address = 172.16.31.197
number = 1
name = rac2
cluster = ocfs2
cluster:
node_count = 2
name = ocfs2
[root@rac2 ~]# /etc/rc.d/init.d/ocfs2 status
[root@rac2 ~]#
the both node is installed on VMWARE AND LINUX IS OEL4....THE CONFIGRATION OF vmx file below
------from first node vmx file configration----------------------
.encoding = "windows-1252"
config.version = "8"
virtualHW.version = "7"
scsi0.present = "TRUE"
scsi0.virtualDev = "lsilogic"
memsize = "1056"
ide0:0.present = "TRUE"
ide0:0.fileName = "D:\vmware\ractesting\rac1.vmdk"
ide1:0.present = "TRUE"
ide1:0.autodetect = "TRUE"
ide1:0.deviceType = "cdrom-image"
floppy0.startConnected = "FALSE"
floppy0.fileName = ""
floppy0.autodetect = "TRUE"
ethernet0.present = "TRUE"
ethernet0.wakeOnPcktRcv = "FALSE"
ethernet0.addressType = "generated"
usb.present = "TRUE"
ehci.present = "TRUE"
sound.present = "TRUE"
sound.fileName = "-1"
sound.autodetect = "TRUE"
serial0.present = "TRUE"
serial0.fileType = "thinprint"
pciBridge0.present = "TRUE"
pciBridge4.present = "TRUE"
pciBridge4.virtualDev = "pcieRootPort"
pciBridge4.functions = "8"
pciBridge5.present = "TRUE"
pciBridge5.virtualDev = "pcieRootPort"
pciBridge5.functions = "8"
pciBridge6.present = "TRUE"
pciBridge6.virtualDev = "pcieRootPort"
pciBridge6.functions = "8"
pciBridge7.present = "TRUE"
pciBridge7.virtualDev = "pcieRootPort"
pciBridge7.functions = "8"
vmci0.present = "TRUE"
roamingVM.exitBehavior = "go"
displayName = "rac1"
guestOS = "rhel4"
nvram = "rac1.nvram"
virtualHW.productCompatibility = "hosted"
printers.enabled = "TRUE"
extendedConfigFile = "rac1.vmxf"
disk.locking="FALSE"
disklib.dataCacheMaxSize="0"
scsi1.sharedBus="virtual"
scsi0:0.present = "TRUE"
scsi0:0.fileName = "D:\sharedstorage\ocfs2disk.vmdk"
scsi0:0.mode = "independent-persistent"
scsi0:0.deviceType="disk"
scsi0:1.present = "TRUE"
scsi0:1.fileName = "D:\sharedstorage\asmdisk1.vmdk"
scsi0:1.mode = "independent-persistent"
scsi0:1.deviceType="disk"
scsi0:2.present = "TRUE"
scsi0:2.fileName = "D:\sharedstorage\asmdisk2.vmdk"
scsi0:2.mode = "independent-persistent"
scsi0:2.deviceType="disk"
scsi0:3.present = "TRUE"
scsi0:3.fileName = "D:\sharedstorage\asmdisk3.vmdk"
scsi0:3.mode = "independent-persistent"
scsi0:3.deviceType="disk"
ethernet1.present = "TRUE"
ethernet1.connectionType = "hostonly"
ethernet1.wakeOnPcktRcv = "FALSE"
ethernet1.addressType = "generated"
ide1:0.fileName = "E:\OEL4\Enterprise-R4-U4-i386-disc4.iso"
ethernet0.generatedAddress = "00:0c:29:de:6c:7a"
ethernet1.generatedAddress = "00:0c:29:de:6c:84"
uuid.location = "56 4d a1 a0 2f 33 08 24-12 e5 7a 39 a7 de 6c 7a"
uuid.bios = "56 4d a1 a0 2f 33 08 24-12 e5 7a 39 a7 de 6c 7a"
cleanShutdown = "FALSE"
replay.supported = "FALSE"
replay.filename = ""
ide0:0.redo = ""
scsi0:0.redo = ""
scsi0:1.redo = ""
scsi0:2.redo = ""
scsi0:3.redo = ""
pciBridge0.pciSlotNumber = "17"
pciBridge4.pciSlotNumber = "21"
pciBridge5.pciSlotNumber = "22"
pciBridge6.pciSlotNumber = "23"
pciBridge7.pciSlotNumber = "24"
scsi0.pciSlotNumber = "16"
usb.pciSlotNumber = "32"
ethernet0.pciSlotNumber = "33"
ethernet1.pciSlotNumber = "34"
sound.pciSlotNumber = "35"
ehci.pciSlotNumber = "36"
vmci0.pciSlotNumber = "37"
vmotion.checkpointFBSize = "16777216"
ethernet0.generatedAddressOffset = "0"
ethernet1.generatedAddressOffset = "10"
vmci0.id = "-1478595462"
checkpoint.vmState = ""
tools.syncTime = "TRUE"
tools.remindInstall = "FALSE"
ide1:0.startConnected = "FALSE"
------and also paste vmx file configration from node2---------------------
.encoding = "windows-1252"
config.version = "8"
virtualHW.version = "7"
scsi0.present = "TRUE"
scsi0.virtualDev = "lsilogic"
memsize = "1056"
ide0:0.present = "TRUE"
ide0:0.fileName = "rac2.vmdk"
ide1:0.present = "TRUE"
ide1:0.autodetect = "TRUE"
ide1:0.deviceType = "cdrom-image"
floppy0.startConnected = "FALSE"
floppy0.fileName = ""
floppy0.autodetect = "TRUE"
ethernet0.present = "TRUE"
ethernet0.wakeOnPcktRcv = "FALSE"
ethernet0.addressType = "generated"
usb.present = "TRUE"
ehci.present = "TRUE"
sound.present = "TRUE"
sound.fileName = "-1"
sound.autodetect = "TRUE"
serial0.present = "TRUE"
serial0.fileType = "thinprint"
pciBridge0.present = "TRUE"
pciBridge4.present = "TRUE"
pciBridge4.virtualDev = "pcieRootPort"
pciBridge4.functions = "8"
pciBridge5.present = "TRUE"
pciBridge5.virtualDev = "pcieRootPort"
pciBridge5.functions = "8"
pciBridge6.present = "TRUE"
pciBridge6.virtualDev = "pcieRootPort"
pciBridge6.functions = "8"
pciBridge7.present = "TRUE"
pciBridge7.virtualDev = "pcieRootPort"
pciBridge7.functions = "8"
vmci0.present = "TRUE"
roamingVM.exitBehavior = "go"
displayName = "rac2"
guestOS = "rhel4"
nvram = "rac2.nvram"
virtualHW.productCompatibility = "hosted"
printers.enabled = "TRUE"
extendedConfigFile = "rac2.vmxf"
ethernet1.present = "TRUE"
ethernet1.connectionType = "hostonly"
ethernet1.wakeOnPcktRcv = "FALSE"
ethernet1.addressType = "generated"
disk.locking="FALSE"
disklib.dataCacheMaxSize="0"
scsi1.sharedBus="virtual"
scsi0:0.present = "TRUE"
scsi0:0.fileName = "D:\sharedstorage\ocfs2disk.vmdk"
scsi0:0.mode = "independent-persistent"
scsi0:0.deviceType="disk"
scsi0:1.present = "TRUE"
scsi0:1.fileName = "D:\sharedstorage\asmdisk1.vmdk"
scsi0:1.mode = "independent-persistent"
scsi0:1.deviceType="disk"
scsi0:2.present = "TRUE"
scsi0:2.fileName = "D:\sharedstorage\asmdisk2.vmdk"
scsi0:2.mode = "independent-persistent"
scsi0:2.deviceType="disk"
scsi0:3.present = "TRUE"
scsi0:3.fileName = "D:\sharedstorage\asmdisk3.vmdk"
scsi0:3.mode = "independent-persistent"
scsi0:3.deviceType="disk"
ide1:0.fileName = "E:\OEL4\Enterprise-R4-U4-i386-disc4.iso"
ide1:0.startConnected = "FALSE"
ethernet0.generatedAddress = "00:0c:29:dc:e1:c9"
ethernet1.generatedAddress = "00:0c:29:dc:e1:d3"
tools.syncTime = "TRUE"
uuid.location = "56 4d c6 2a 5b 77 ee e5-79 46 84 cd 2b dc e1 c9"
uuid.bios = "56 4d c6 2a 5b 77 ee e5-79 46 84 cd 2b dc e1 c9"
cleanShutdown = "FALSE"
replay.supported = "FALSE"
replay.filename = ""
ide0:0.redo = ""
scsi0:0.redo = ""
scsi0:1.redo = ""
scsi0:2.redo = ""
scsi0:3.redo = ""
pciBridge0.pciSlotNumber = "17"
pciBridge4.pciSlotNumber = "21"
pciBridge5.pciSlotNumber = "22"
pciBridge6.pciSlotNumber = "23"
pciBridge7.pciSlotNumber = "24"
scsi0.pciSlotNumber = "16"
usb.pciSlotNumber = "32"
ethernet0.pciSlotNumber = "33"
ethernet1.pciSlotNumber = "34"
sound.pciSlotNumber = "35"
ehci.pciSlotNumber = "36"
vmci0.pciSlotNumber = "37"
vmotion.checkpointFBSize = "16777216"
usb:1.present = "TRUE"
ethernet0.generatedAddressOffset = "0"
ethernet1.generatedAddressOffset = "10"
vmci0.id = "735896009"
usb:1.deviceType = "hub"What's the output of "/etc/init.d/o2cb status"
Have you formatted the partition i.e. mkfs -t ocfs2 /dev/sdb1 -
Sun Studio 10 Linux, RHEL3 U4, install fails on openmotif21.
Hello everyone,
As a long-time user of Sun WorkShop... I mean Forte.. I mean Studio; it was with great anticipation that I downloader Sun Studio 10 for Linux.
On one machine it failed to install and on another it installed properly.
Here are the details:
- Machine 1: RHEL3 U2 (x86/i686), small install: Installation of Studio 10 went fine.
- Machine 2: RHEL3 U4 (x86/i686), complete install : Installation of Studio 10 failed. The following can be gathered from the logs under /var/opt/sun:
[....snip....]
Installing Sun Studio Software
Log file: /var/opt/sun/install/logs/Sun_Studio_Software_install.A04060111
Installed: /var/prod/com.sun.studio_10/uninstall_Sun_Studio_Software.class
Uninstaller is at: /var/prod/com.sun.studio_10/uninstall_Sun_Studio_Software.class
Standalone Tools: Building Software
Installing RPM: sun-dmake-10.0-1
[....snip....]
System packages supporting Sun Studio Software
Installing RPM: openmotif-2.1.31-2_IST
error: Failed dependencies:
openmotif < 2.2.2 conflicts with openmotif21-2.1.30-9.RHEL3.4
Error: RPM openmotif-2.1.31-2_IST install failed}.
Install complete. Package: openmotif-2.1.31-2_IST
rpm: Skipping rpm remove of openmotif-2.1.31-2_IST, not installed.
rpm: Skipping rpm remove of sun-svc-locale-zh-10.0-1, not installed.
[....snip....]
And everything is removed. Is this issue documented somewhere? Is there a workaround?
Thanks for reading,
VincentDeselecting 'system supporting packages' from the Sun Studio 10 installation list made the trick and Studio finished installing.
However, I have a few concerns:
1) Since Studio will be using the system's openmotif, will it be supported? (I guess not).
2) Since openmotif-2.1.31-2_IST.i386.rpm from the Studio 10 IDE kits lands under /usr/OpenMotif-2.1.31 and should not conflict with any kind of other/system openmotif, why didn't SUN just create a sun-openmotif-support package? This would prevent a naming conflict when trying to install on RHEL3 if RHEL's openmotif21 is already installed?
3) It appears to work fine, even without SUN's provided openmotif.
Cheers,
Vincent -
10.2.0.4 RAC / OCFS2 upgrade to 11g R2 RAC / ASM
We are in the planning phase of an upgrade from 10.2.0.4 RAC on OCFS2 Cluster to an 11g R2 RAC and ASM. Is there documentation that will guide me in the best possible and least problematic direction? I have recently moved into the DBA world from a Senior Developer position so my questions might be fairly beginner and I apologize for that now. What I am thinking I would do is create a new RAID10 partition on our SAN and have my ASM disks there. Once the 11g database is installed and configured my plan was to take the current database out of service, back it up, and then recover to the 11g ASM ... is this possible when coming from OCFS2? Additionally I have found the certification matrix in support.oracle.com but that seems to be only a certification of the database against a specific OS ... is there a certification matrix for the hardware I will be installing this on especially my QLogic Infiniband HCA and the firmware version there on.
You should try this question on the Database Upgrade forum here. http://goo.gl/hOYZ
Maybe you are looking for
-
Flash simple image gallery problem
Hello I have what I think is an easy problem to solve but I can´t find a solution. I´m creating an image gallery with thumbnails on the right inside a symbol so that they can all fade in at the same time, and the bigger images on the left that also f
-
Error D4O-1007 Discoverer Plus
Hi, When I create a time dimension in the AWM. The I try to connect to Discoverer PLus and get this message "D4O-1007 An error occurred in Oracle OLAP during the execution of the query" Why??? Thanks a lot
-
Hey everyone. I'm running 10.4 server and I'm trying to get iMac(G4) and Mac Mini(G4) clients running 10.3 to auto login into a student profile. Previously we were running 10.3 server and these clients were auto logging fine, but when we upgraded to
-
How I fixed: You have been signed out problem in 2 minutes!
Very frustratingly I was having the same problem with Creative Cloud saying you have been signed out all the time, so I was reading through the forum and got some ideas, so this is what I did and finally I can get back to work and hopefully help some
-
Aperture 3 Photos "Group by date"?
Hi, I recently upgraded from iPhoto, and I miss the feature where photos are grouped by date in the grid view. To see what I mean, see how the image previews in Import have a header bar that splits pictures into dates. I'm wondering if there's a sett