Quantcast
Channel: VMware Communities : Discussion List - Virtualizing Oracle
Viewing all 450 articles
Browse latest View live

Increasing vCPU's on Oracle DB server

$
0
0

My company has an application that currently runs on a physical Solaris 10/SPARC server with ISAM databases.   We want to move to a RHEL 5/Oracle VM.   The application vendor says that the virtual machine requires the same amount of RAM and same number of CPU's (4) as a physical server, which seems like overkill.

 

My first question is can the number of vCPU's be increased on the VM after installation if I find I need more CPU resources without having any ill effects on the VM and Oracle in particular ( I was thinking of starting with 2 vCPU's and working up if needed)?

 

Second question - the # cores/socket setting - for 2 vCPU's is there any benefit in setting 2 cores/socket rather than 1 (host has 2 quad-core CPU).   If I understand what I've read correctly, that really only helps when the # of available cores exceeds the hard limit of CPU's in an OS.

 

Thank you for your help.


Oracle 9i on Win2003EE - How do I perform a system level backup?

$
0
0

Okay -

 

I'm going to put this to rest but I'm going to reach out here first to see if there is any stone unturned.

 

I have reluctantly come to the conclusion that it isn't safe to snapshot an Oracle database.

I was trying to use Veeam to backup the database using pre-freeze/post-thaw scripts but Veeam and VMware have each told me that without VSS or shutting down the dbms that I cannot do this.  Threads in this forum indicate possible db corruption.

 

We have virtualized some pre-10.2.0.3 Oracle databases (e.g. no VSS) and hosted them in Win2003EE guest VMs.

It was the right thing to do.  And I admit life is peachy but for the backup/DR strategy.

The systems are not easily shutdown for maintenance so hot backups are the norm.

 

We are:

performing RMAN backups and rounding that out with nightly exports

backing up all guest VMs (including Oracle servers) with Veeam

Excluding the oracle data stores from the Veeam backup

 

I was digging the incremental block changes and push-button restore of Veeam.  But Veeam requires a snapshot of the dbms which Oracle doesn't like.

 

- so here's my question, more of a head-scratcher.

 

How do I perform a system level recovery of an Ora server?

 

While I may have exports and RMAN backups I don't have the actual dbms server backed up, at least not entirely.

Without the oracle data stores there is no target for RMAN to restore.

In fact, without the datastores I don't think VMware will boot the vm guest.

 

Does it look like this?

1 Restore VM guest to host (missing ora datastores)

2 Create ora datastores

3 Boot VM

4 allocate ora datastores

5 Restore db using RMAN

6 Recover ora db

 

Any insight would be greatly appreciated.

I have alot of experience with Oracle but only 6 months of running hard in VMware.  I hate Windows.

 

Thanks

Rick

Oracle HA and vmotion with windows cluster

$
0
0

Hello guys. Just a moments ago i saw a webcast about "Virtualizing Microsoft and Oracle on VmWare VSphere, benefits and best practices". Some where in the presentation they talked about windows cluster being compatible with vmotion and HA.

 

Is this true or did i get it wrong? and if it is, is it compatible with an Oracle implementation? because i cant seem to find any information about this, does anybody has any detailed info or something?

 

Thanks and have a nice day.

Oracle Extended RAC on VM

$
0
0

Hello all,

 

Can anyone point me to any documentation (or sites) regarding how to setup Oracle extended RAC on VMware?

 

We are in the beginning stages of looking at extended RAC and there is some confusion about the storage requirements associated with it.

 

Thanks for your help...

DBCA Build Stops with ORA-04030 out of Mem

$
0
0

I am running vSphere 4.1.

 

I built a VM and installed Windows 2008 Enterprise Edition.

The VM has 6 GB of RAM (as seen in the edit settings of the VM).

 

I then installed Oracle 11g Standard One.

 

I am using DBCA to build a DB from a DBCA template that was saved under Oracle 10g.

 

The DB build proceed until the Oracle JVM is being installed

Then I get the ora-04030 error (out of memory).

 

The attached vCenter performance screen shows the VM has only 4 GB of RAM?

BUT, the Windows Task Mgr is showing 6 GB matching the VM settings.

 

Since this is Windows Server 2008 Enterprise, it looks like the Windows VM is seeing 6 GB, but the vCenter Perf screen is only showing the VM as having 4 GB.

 

 

I assume the ora-04030 is because the install is only seeing 4 GB of RAM?

 

Any suggestion on what is going on?

 

thanks

Think BIG!

$
0
0

I've been asked to research Oracle Discoverer running in VMWare.  Can it be done well, to what depth, maximum users, size of databases, security, problems, etc.  Think Big!  Would anyone be willing to provide links to documents I might explore?

Experience with VMotion and PeopleSoft

$
0
0

We are looking at a VM architecture for PeopleSoft application server (Tuxedo) tier.

 

There would be multiple VMs, spread across 2 hosts. If one host goes down, the VMs are migrated in-memory to the other host, using VMotion. It is expected that this functionality will provide real-time high-avaliability, with no disruption of service.

 

My understanding is that the VMotion move is not instantaneous, but rather takes a few seconds. We are wondering if anyone has any experience with Tuxedo-based applications (e.g. PeopleSoft) using VMotion, specifically wondering if there are any known issues with the tuxedo domains becoming unstable during a VMotion event, stale DB connections, etc.

Cloning Oracle ASM disks

$
0
0

Hi,

 

We'll virtualize a physical Linux server having an  Oracle Database with ASM.  The physical server will remain untouched and  the new virtual will be under testing.  What do we must do respect to  the LUNs / volumes on which ASM has database data?  I guess we can clone them via SAN, but not sure what's next on the virtual server.

 

Thank you very much.


Oracle ASM, SAN and VMware

$
0
0

Hello,

 

I have a question about using Oracle ASM. If I only have one physical SAN disk (2 disks mirrored) isolated for an Oracle VM is there any point in implementing ASM in the VMs filesystem by creating multiple virtual disks to support ASM if any IO is still really only accessing one physical disk? This is the first time I will have to do an Oracle VM install and I want to make sure the DBAs have the right info for setting up the DB.

 

I see that it is tip #14 in the Essentiual Deployment tips doc to use Oracle ASM but it doesn't say anything about assumptions...eg: Is there more than one physical disk available? I am thinking the DBA doesn't want to bother with ASM as there is only one physical disk under this VM but we are in a virtual world now and I don't think the same that would be true for a physical server is necessarily the case for a virtual one.

 

Thanks

Oracle RAC on Windows 2008, no support??

$
0
0

It looks like Oracle does not support setting up the RAC within a virtualized environment running on windows.  It must be Linux/Unix based.

 

http://www.oracle.com/technetwork/database/virtualizationmatrix-172995.html

 

Anyone successfully, in a production environment, running Oracle RAC 11.2.03 on Windows 2008?  Are you using shared RDM's or the "multi-writer" mode as described in KB 1034165?

 

The VMware RAC Deploy Guide (http://www.vmware.com/files/pdf/partners/oracle/vmware-oracle-rac-deploy-guide.pdf) describes the configuration using the "multi-writer" mode KB 1034165 however the OS listed is only Oracle Linux.  No mention of Windows 2008.

Error uploading configuration data to the repository

$
0
0

During installation of Oracle in Virtual Machine, I am getting "Error uploading configuration data to the repository..." warning message. Please find the attached for more details.

 

If I ignore this error and continue the setup then till the end of the setup there will be no issue. After that listener is running with 1521 port number but web console is not opening using "emctl start dbconsole" command, giving me one more error.

 

1. Can we ignore this error?

2. How to reconfigure using emca script?

 

My Environment is,

VM is created by VMware Studio 2.5

Oracle version = 11.2 64-bit

OS version = CentOS 5.5 64-bit

 

Plese let me know if you require more information.

Oracle 2nd node fails to start in cluster due to locking

$
0
0

Greetings,

 

I’m attempting to setup an oracle cluster (ASM) with shared disks on elinux 5.8. I modified the vmx file with the parameters below to supposedly allow disk sharing between the nodes:

 

 

disk.locking = "FALSE"

diskLib.dataCacheMaxSize = "0"

diskLib.dataCacheMaxReadAheadSize = "0"

diskLib.dataCacheMinReadAheadSize = "0"

diskLib.dataCachePageSize = "4096"

diskLib.maxUnsyncedWrites = "0"

 

scsi1.present = "TRUE"

scsi1.virtualDev = "lsilogic"

scsi1.sharedBus = "VIRTUAL"

 

scsi1:1.present = "TRUE"

scsi1:1.mode = "independent-persistent"

scsi1:1.fileName = "

/vmfs/volumes/4f709747-0ec9e0a0-219b-001e0bccdf5c/test1/test1_1.vmdk"

scsi1:1.deviceType = "scsi-hardDisk"

scsi1:1.redo = ""

 

 

I can start either node separately, however if I start them both, the second fails because the first has the file locked.

 

From the log:

 

Apr 13 13:00:45.313: vmx| DISK: OPEN '/vmfs/volumes/4f709747-0ec9e0a0-219b-001e0bccdf5c/test2_1/test2.vmdk' Geo (1958/255/63) BIOS Geo (0/0/0)

Apr 13 13:00:45.397: vmx| DISK: OPEN scsi0:1 '/vmfs/volumes/4f709747-0ec9e0a0-219b-001e0bccdf5c/test1/test1_1.vmdk' independent-persistent R[]

Apr 13 13:00:45.437: vmx| AIOGNRC: Failed to open '/vmfs/volumes/4f709747-0ec9e0a0-219b-001e0bccdf5c/test1/test1_1-flat.vmdk' : Failed to lock the file (67) (0x2017).

 

Does anyone out there know how to setup disks in VM that can be shared by multiple nodes?

 

Thanks!

Oracle VM Server as virtual machine

$
0
0

We want to setup a virtual machine with Oracle VM Server as the guest.   Does anyone know what Linux O/S is best used for this process?

Oracle VMWare network question

$
0
0

We currently have a live Oracle server, that has an LACP connection over 4 1 Gb NICs to a netapp FAS2040.   Unfortunately it's failing so we're going to create a VM.  We have it online right now, but I have a question.  The VMWare server has 4 1 Gb NICs trunked out, will a single VMXNET 3 NIC be able to push as much traffic as the 4 trunked NICs could?  Or do I need to create a similar setup?  Has anyone else done something like this?

 

Thanks in advance.

Fight the FUD - Oracle Support and Licensing on VMware vSphere

$
0
0

I have recently published an article that I hope will help customers make a well informed decision when evaluating how best to virtualize Oracle workloads on vSphere. The article will help customers determine fact from fiction, and help to remove the FUD that generally comes from some professionals around Oracle virtualization. If you like this article I'm sure you'll like my Oracle sessions at VMworld (APP-BCA1432 and APP-BCA1624).

 

http://longwhiteclouds.com/2012/07/21/fight-the-fud-oracle-licensing-and-support-on-vmware-vsphere/


oracle database consolidation effort - physical to VMware

$
0
0

Hate to post sort of a generic question but just trying to figure out  what I might look for to try and narrow down the root cause of this  issue.
Old environment is RH4 64-bit Linux servers running primarily oracle 10g EE.
Due to licensing costs etc. moving to a virtualized environment (VMware vSphere) with VM's running RH5 64-bit.

 

Couple of db moves seemed to go ok but the etl jobs that run to Dev Data  Warehouse take much longer to complete like 60 minutes to now 100 minutes.   So it seems to be related to queries over database links so far but  more testing to be done with other apps as we migrate.

I know lots of variables involved here so maybe hard to make any suggestions but to summarize the move is from older Intel servers to much newer and faster servers with huge memory footprints and newer faster SAN array.
- tried increasing the ram and vcpu's made available to the VM and in turn bumped up sga_max_size and shared_pool_size
- increased redo log sizes but so far no noticable improvement

 

By monitoring the sql job on both the source and destination databases  the connection details shows a high # for the Time Waited column in the  "SQL*Net message" entry.
- tried adding TCP.NODELAY = YES to the sqlnet.ora file
- no change
- is this an indication of a network latency or setting issue or a red herring?

Via an Oracle article is says to enable HugePages on Linux  [ID 361468.1]
- done but not exactly sure how to substantiate the change other than I can see that "grep Huge /proc/meminfo" now shows some values.

Maybe it is just poor sql?

Seeking Beta Participants for Oracle Enterprise Manager 12c plugin for Vmware

$
0
0

Greetings Vmware 'Virtualizing Oracle' Community Members:

 

Blue Medora is seeking interested participants for the 2nd Beta release of a forthcoming Oracle Enterprise Manager 12c based Plug-in for monitoring and managing VMware environments.

 

More details on the features / capabilities of EM12c Plug-in for VMware can be found here (http://bit.ly/RbM4iv).

 

If you have any interest in participating in the beta, you can signup online (http://bit.ly/NJXoSC) or you can directly reach out to our Beta coordinator at bryce.kaiser@bluemedora.com.

 

Regards, --Nathan

email: nathan.owen@bluemedora.com

blog: bluemedora.com/blog

Transparent failover

$
0
0

Hello All,

 

I'd like to have some feedback on what we are trying to achieve here...

 

On the same site, we have 2 distant rooms for disaster recovery on the same site. ESX host(s) in each room with its own SAN. We would like to have the best possible solution for an Oracle Database with transparent failover and thus (perfect / live) replication between those two rooms.

 

Our current idea would be to use RAC and ASM - with an ASM Disk Group that has disks from the SAN in each room. The idea is to avoid using any kind of SAN replication and let Oracle handle everything. I would then configure the Oracle Virtual Servers in room #1 and #2 with disks from the SAN in room #1 and from the SAN in room #2. All of them in the same ASM disk group.

 

The way I understand this, Oracle ASM will handle the "replication" and if Room #1 burns down, RAC will allow business to continue transparently.

 

Am I day dreaming?

What about having disks from two SAN in different rooms in an ASM disk group? Is this viable?

 

Any other ideas / possibilities to solve our issue?

 

Thank you very much!

 

X.

Virtualizing Oracle ECM solution

$
0
0

Hi All,

 

I am working on virtualizing Oracle ECM solution with the below functions.

 

Oracle DB
Web Center Content
Web cCenter Portal
Oracle Web Tier
MGMT & Backup Solution for Oracle

 

 

Need guidance to virtualize the above solution. Like, need pros and cons, licensing and support from Oracle for the above functions, best hw for Oracle solutions, any best practices will be really helpful.

 

Thanks for your time and guidance.

 

Regards,

JKPK

Oracle ECM\UCM on VMware vSphere 5.0

$
0
0

Hi All,

 

I am working with providing a solution for Oracle ECM\UCM on VMware vSphere 5.0.

 

Need documents\suggestions\solutions guide in implementing the above.

 

Regards,

JKPK

Viewing all 450 articles
Browse latest View live




Latest Images