NFS supports Oracle RAC with TAF?

Similar Questions

• Oracle RAC with MRS 5.1

  Hi, good day to you all...

  Did anyone here ever has implemented or tested Oracle RAC protected with Vmware SRM?

  VMware SRM 5.1 does support Oracle RAC?

  AFAIK, the VMs CARS they shared virtual disk to another RAC VMs. in my VMs RAC Setup - B had 7 virtual disk mapped to RAC - A VMs and everything works very well in a production environment, and they intend to protect this RAC VMs with VMware SRM 5.1. We use replication Array with EMC RecoverPoint as SRA.

  When I try to test failover my protection group which is containing these two VMs, VMS in RAC - B not able to start with the following error

  

  The virtual disk mapping prevented virtual machines to start upward, and there is no snapshot of the CAR - A or B - CARS.

  Can I modify the virtual machine starts action of the stimulus package, im setting of the RAC - B to not power on the recovering site. map manually the virtual disk after the failover of the trial and the RAC - B still cannot start the system.

  Please give me some lights, how to map the virtual disk after failover to the recovering site.

  Thanks before

  Rgds

  Dika

  Good find!

  CTK files are created by ESX when CBT (changed block Tracking) is enabled. CBT is used by the backup software and is not supported with disks using shared bus.

  So it can certainly cause problems. Do you have a backup solution for virtual machines?

  CBT can be turned off (requires a virtual machine downtime):

  http://KB.VMware.com/kb/1031873

  Michael.

• Installation of Oracle RAC with Oracle Linux 5.8(64bit) and 11 GR 2 on EXSI 5.5

  Well, I'm trying to install oralce RAC on oracle Linux 5.8(64bit) with 11 GR 2.

  After installation oracle Linux package required by it I m is problem for oracleasmlib - 2.0.4 - 1.el6.x86_64.rpm.

  I triyed with rpm - Uvh oracleasmlib - 2.0.4 - 1.el6.x86_64.rpm and the result is rpmlib (FileDigests) and (PayloadisXz) rpmlib need for oracleasmlib - 2.0.4 - 1.el6.x86_64.rpm

  I triyed with yum localinstall oracleasmlib - 2.0.4 - 1.el6.x86_64.rpm and get an error of public key for 2.0.4 - oracleasmlib - 1.el6.x86_64.rpm is not installed...

  I tried on 3 machines and same result.

  With 32-bit I m not getting any problem

  Found the solution

  the inadequacy of the kernel version and oracle asm

• two node Oracle RAC with two Oracle databases completely different on each node?

  We currently have two different applications that use Oracle databases.  Current configuration is two different oracle homes dbhome_1, dbhome2 with separate mounting LINUX points containing the Oracle homes and *.dbf files on a single host server.

  Previously, the idea was to set up a second server identical host configured as a 'data guard' for both of these databases.  Now, management wants to implement a cluster of CARS of two nodes, with each node running two databases.  With the Standard edition.

  I only did CARS of high availability and scalability for a single database.  Is it possible to run two completely different databases Oracle on a two-node RAC cluster?  I am inclined to think that they it has divided into

  two clusters RAC to two nodes on four small servers host.  Any suggestions or comments would be appreciated?

  Is it possible to run two completely different databases Oracle on a two-node RAC cluster?

  The short answer is Yes... it is possible.  It will work without any problem, as long as you have the resources available to an additional database.

  I am inclined to think that they it has divided into two clusters RAC to two nodes on four small servers host.

  There is another option.  In the past I have two databases on a RAC cluster that has been implemented for reasons of cost of license / (I don't think that this applies to itself, but you will need to check).

  However having seen on separate hosts that you have only potentially interruptions of service on a database if you decide to do something like upgrading from one of the databases and thus upgrade the GI as well.  This would mean interruptions in service for all your databases using the GI.

• Installing Oracle RAC: Unix, Windows ASM nodes

  I have a question about the Oracle RAC configuration. I never did no prior installation RAC or ASM. Maybe a stupid question to some of you.
  
  Is it possible to install Oracle RAC with following options?
  
  2 node RAC with Sun Solaris
  Shared storage using ASM in Windows server
  
  Any additional information you can provide will be greatly appreciated.
  
  Thanks in advance

  In very simple terms, when using ASM, database server process read the CRUDE device and process DBWR write devices ROUGH containing tablespace information. With ASM, these processes simply ask ASM 'where the disk read/write. Therefore, ASM provides simply a "table of contents". (It is much more than that, but this seems to be a good way to start the discussion.)

  ASM MUST run on the same computer as the database instance that is to use it.

  In your scenario, the Windows computer may be Server iSCSI providing raw devices to Solaris machines. These devices would then be managed by ASM and reading/praav1s directly by database processes.

• Is it possible create Oracle RAC on NFS datastore?

  Hello

  Is it possible create Oracle RAC on NFS datastore?   With the VMFS data store, we use VMDK files as the Oracle RAC shared virtual disks with the Comptrollership and multi-writer SCSI Paravirtual, what about the NFS datastore? is the controller SCSI Paravirtual and writer multi function supported on the NFS datastore?

  Unless I'm missing something, this is not supported on NFS.

• RAC with Oracle VM and Windows Server

  Hi all,
  
  We want to implement the system with Oracle database 11g or 10g 64 bit (CARS) on the OS of Microsoft Server 2008/2003. We do not have enough licenses for all the hearts, and we want to find a way to implement the RAC database with existing licenses.
  
  Is this a good solution to use virtualization, as Oracle VM?
  How much is safe to use virtualization?
  Can we do Oracle Real Application Clusters (RAC) environments Oracle VM with Windows Server 2008 64-bit (or 2003 Server) operating system and use hard partition?
  Is it better to invest in licenses or try with Oracle VM?
  
  
  Oracle VM can be configured so that it is recognized as a difficult score.
  Hard partitions allow customers of one license these CPU used by the partition instead of license all the processors on the physical server.
  Source: oracle.com
  
  
  Thank you!

  Virtualization is not certified for Oracle RAC on Windows.

  + Software on Oracle VM [464754.1 ID] + notes States certified:

  Oracle Real Application Clusters (RAC)
  
      Oracle 10.2.0.4 and up (10gR2) and 11.1.0.7 and up (11gR1) and 11gR2 RAC for Linux x86 and Linux x86_64 certified on Oracle VM
          Guest OS: Oracle Linux 5.1 (and above) RHEL 5.1 (and above) for Linux x86 / Linux x86_64
          Paravirtualized (PV) mode only (Guest OS and drivers)
          Only supported on Oracle VM 2.1.2 and above
          Live-migration of an Oracle RAC VM is supported with Oracle VM 2.2.1 and above.
          Previous versions are not supported. Please refer to this link for best practices.
  
          Over-committing CPUs is not recommended, but supported with the following restrictions:
              The total amount of VCPUs allocated to guest domains (running Oracle RAC guests), should not exceed
              two times (2x) the amount of real CPUs / cores in the Oracle VM server.
              The amount of VCPUs allocated to a single guest domain should not exceed the amount of real CPUs /
              cores in the Oracle VM server.
              Maintain Oracle VMs default VCPU allocation for dom-0: Oracle VM will allocate 1 VCPU for each real CPU or core to dom-0.
              CPU pinning is only recommended for hard partitioning. If no hard partitioning is required, CPU pinning should not be used.
  
          Static support only (dynamic support is being planned):
              Dynamic resizing of guest virtual machine is not supported (VCPU, memory and I/O)
              Virtual Machine Pause/Restore of an active Real Application Cluster virtual machine is not supported. 
  

• FRA on NFS Oracle RAC one node

  Hi all
  
  We have installed Oracle RAC one node on Oracle Linux. Everything seems to work fine except one small thing: we try to change the database in archivelog mode, but when we try to move the database, we get ORA-19816 "WARNING: files may exist in... that are not known to the database." and "Linux-x86_64 error: 37: no available lock.
  
  The FRA is mounted as NFS share with the following options: "rw, bg, hard, nointr, rsize is 32768, wsize = 32768, proto = tcp, noac, worms = 3, suid.
  
  I searched a lot on the Internet but couldn't find any hint. Can someone point me to the right installation guide?
  
  Thanks in advance

  Hello

  user10191672 wrote:
  Hi all

  We have installed Oracle RAC one node on Oracle Linux. Everything seems to work fine except one small thing: we try to change the database in archivelog mode, but when we try to move the database, we get ORA-19816 "WARNING: files may exist in... that are not known to the database." and "Linux-x86_64 error: 37: no available lock.

  The FRA is mounted as NFS share with the following options: "rw, bg, hard, nointr, rsize is 32768, wsize = 32768, proto = tcp, noac, worms = 3, suid.

  I searched a lot on the Internet but couldn't find any hint. Can someone point me to the right installation guide?

  Check if service NFSLOCK works... and if not start it.

  # service nfslock status
  

  * Mounting for Oracle files options when used with NAS [359515.1 ID] devices *.
  Mounting for Oracle data files options

  rw,bg,hard,nointr,rsize=32768, wsize=32768,tcp,actimeo=0, vers=3,timeo=600
  

  For games of backups RMAN, copies images and Data Pump dump files, the "NOAC" mount option should not be specified - it's because RMAN and Data Pump do not check this option and specifying this can adversely affect performance.

  The following NFS options must be specified for 11.2.0.2 disc RMAN backup directory:

  opts="-fstype=nfs,rsize=65536,wsize=65536,hard,actime=0,intr,nodev,nosuid"
  

  Hope this helps,
  Levi Pereira

  Published by: Levi Pereira on August 18, 2011 13:20

• Taking JDBC connection configuration supported the RAC in Oracle SQL Developer?

  How can one Oracle SQL Developer be configured to successfully connect to a [Oracle RAC | http://en.wikipedia.org/wiki/Oracle_RAC] environment, that is, not only the installation program always to connect to one of the individual nodes?
  
  [http://programmersjournal.blogspot.com/2008/08/jdbc-connection-string-for-oracle-rac.html] is an example of a classic and taking version supported the equivalent CARS of a JDBC connection string. I've used a CAR of the same format connection string from within Java applications custom without problem. I thought that the same thing should be configurable in Oracle SQL Developer, by setting the 'Connection Type' drop-down menu on "Advanced", then enter the connection string JDBC RAC-aware as "Custom URL JDBC". However, after the test connection, the following error message is displayed at the bottom of the dialog box:
  
  Status: Failure-Test failed: IO exception: NL Exception has been generated
  
  There is no additional information shown on the display 'Page Logging'. With the help of Oracle SQL Developer 1.5.5.
  
  Thank you!

  You could always give the current 2.1 ai2 test... continuous development may have fixed the issues in this area...

  Hope that helps,
  K.

• How can I add data file to an existing table on Oracle RAC ASM with no OMF? Thank you!

  How can I add data file to an existing table on Oracle RAC ASM with no OMF? Thank you!

  Hello

  So I guess you have some files in ASM, see your first existing file structure

  Select file_name

  from dba_data_files;

  and to add to an existing table, the example below (even if you're better sticking with OMFs!)-is that what you are looking for?

  SQL > create tablespace TEST1

  2 datafile '+ DATA' size 1 M;

  Created tablespace.

  SQL > select file_name in dba_data_files;

  FILE_NAME

  --------------------------------------------------------------------------------

  +Data/orcl2/datafile/users.259.859820983

  +Data/orcl2/datafile/undotbs1.258.859820983

  +Data/orcl2/datafile/SYSAUX.257.859820983

  +Data/orcl2/datafile/system.256.859820981

  +Data/orcl2/datafile/example.269.859821049

  +Data/orcl2/datafile/Test1.271.859843053

  6 selected lines.

  SQL > alter tablespace TEST1

  2 Add datafile ' + DATA / mynewfile01.dbf ' size 2 m;

  Tablespace altered.

  SQL > select file_name in dba_data_files;

  FILE_NAME

  --------------------------------------------------------------------------------

  +Data/orcl2/datafile/users.259.859820983

  +Data/orcl2/datafile/undotbs1.258.859820983

  +Data/orcl2/datafile/SYSAUX.257.859820983

  +Data/orcl2/datafile/system.256.859820981

  +Data/orcl2/datafile/example.269.859821049

  +Data/orcl2/datafile/Test1.271.859843053

  + DATA / mynewfile01.dbf

  Thank you

• Oracle RAC 12.1.0.2 (GI battery) deployment on OEL 7 (3.8.13 - 35.3.4.el7uek) fails with ORA-27102

  Hi guys,.

  I am currently trying to deploy a complete installation (GI battery) of Oracle RAC 12.1.0.2 cool on OEL 7 (3.8.13 - 35.3.4.el7uek) with 2 nodes for validation purposes. However the installation itself went well, but 'Creation repository for Oracle Grid Infrastructure Management container' step fails with "ORA-01034: ORACLE not available / ORA-27102: out of memory / Linux-x86_64 error: 12: cannot allocate memory". The steps of runInstaller validation completed successfully, but I never take this step of configuration without jumping.

  Here is my configuration which should normally avoid such errors ORA.

  Host

  -bash - $4.2 uname - a

  Linux OELRAC1 3.8.13 - 35.3.4.el7uek.x86_64 #2 SMP Tue Jul 29 23:24:14 CDT 2014 x86_64 x86_64 x86_64 GNU/Linux

  "Creation of repository database for Oracle Grid Infrastructure Management container" error in /oracle/base/cfgtoollogs/dbca/_mgmtdb/trace.log

  From restoration to August 9, 14

  allocated channel: ORA_DISK_1

  channel ORA_DISK_1: SID = 12 type of device = DISK

  channel ORA_DISK_1: from the restore backup set data file

  channel ORA_DISK_1: specifying datafile (s) to restore from backup set

  channel ORA_DISK_1: restore datafile 00003 to + GRID

  channel ORA_DISK_1: backup /oracle/grid/12102/assistants/dbca/templates/MGMTSeed_Database.dfb piece reading

  channel ORA_DISK_1: ORA-19870: error when restoring the backup /oracle/grid/12102/assistants/dbca/templates/MGMTSeed_Database.dfb piece

  ORA-19504: could not create the file "+ GRID.

  ORA-17502: ksfdcre:4 cannot create the file + GRID

  ORA-15001: diskgroup 'GRID' does not exist or is not mounted

  ORA-01034: ORACLE not available

  ORA-27102: out of memory

  Linux-x86_64 error: 12: cannot allocate memory

  Additional information: 2663

  Additional information: 1565392897

  Additional information: 161480704

  switch to the previous backup

  Number of folder of the data file = 3 name = + GRID

  RMAN-00571: ===========================================================

  RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.

  RMAN-00571: ===========================================================

  RMAN-03002: failure of the restore command at 09/08/2014 19:41:47

  ORA-01119: error in creating file of database "+ GRID.

  ORA-17502: ksfdcre:4 cannot create the file + GRID

  ORA-15001: diskgroup 'GRID' does not exist or is not mounted

  ORA-01034: ORACLE not available

  ORA-27102: out of memory

  Linux-x86_64 error: 12: cannot allocate memory

  Additional information: 2663

  Additional information: 1565392897

  Additional information: 1614807040

  RMAN-06956: create the data file failed; try again after removing + OS GRID

  SHM / CPI (check the id of shm with previous information)

  -bash - $4.2 ipcs - a

  -Shared memory segments-

  key shmid owner perms bytes nattch status

  1565360128 0 x 00000000 grid 640 4096 0

  1565392897 0 x 00000000 grid 640 4096 0

  0xfba47600 1565425666 640 24576 29 grid

  Memory of the ASM instance setting

  SQL > see the memory settings

  VALUE OF TYPE NAME

  ------------------------------------ ----------- ------------------------------

  whole big memory_max_target 1076M

  whole large memory_target 1076M

  ASM disk groups

  SQL > select NAME, STATE, TOTAL_MB, USABLE_FILE_MB from v$ asm_diskgroup;

  NAME STATE TOTAL_MB USABLE_FILE_MB

  ------------------------------ ----------- ---------- --------------

  RACK MOUNTED 6144 4868

  / dev/shm to memory_target (more than enough free space)

  -bash - $4.2 df-h

  Size of filesystem used Avail use % mounted on

  tmpfs, 630M 2.4 G 3.0 G 21% / dev/shm

  Kernel for SHM limits set to unlimited parameter

  -bash - $4.2 sysctl - a | grep shm

  kernel.shmall = 1152921504606846720

  kernel.shmmax = 922337203685477580

  User limits on unlimited for memory

  -bash-4, $ 2 cat /etc/security/limits.conf

  # Oracle settings

  grid soft nproc 2047

  grid hard nproc 16384

  grid soft nofile 1024

  grid hard nofile 65536

  Oracle nproc 2047 soft

  Oracle nproc 16384 hard

  Oracle soft nofile 1024

  Oracle hard nofile 65536

  * hard memlock unlimited

  * soft memlock unlimited

  -bash - 4, $ 2 known - grid

  -bash - $4.2 ulimit - a

  the file size (blocks, - c) of base 0

  (kbytes, - d) data seg size unlimited

  scheduling priority (-e) 0

  size of the file (blocks, f) - unlimited

  pending signals (-i) 23953

  Max locked memory (kbytes, - l) unlimited

  size of the memory (k, m) max - unlimited

  open files (-n) 1024

  a size (512 bytes, - p) 8 hose

  (Bytes, - q) POSIX message queues 819200

  real-time priority (-r) 0

  size (Ko, - s) 8192 battery

  time processor (seconds,-t) unlimited

  Max user process (-u) 2047

  virtual memory (KB), - v) unlimited

  the locks on files (-x) unlimited

  So what the hell is wrong here? Why the instance ASM (+ ASM1) returns the error below, even if there is no memory limit / problem. The error with the disk group is also absolutely no sense.

  ------------------------------------------------------------------------------------------------------------------------

  ORA-01119: error in creating file of database "+ GRID.

  ORA-17502: ksfdcre:4 cannot create the file + GRID

  ORA-15001: diskgroup 'GRID' does not exist or is not mounted

  ORA-01034: ORACLE not available

  ORA-27102: out of memory

  Linux-x86_64 error: 12: cannot allocate memory

  Additional information: 2663

  Additional information: 1565392897

  Additional information: 1614807040

  ------------------------------------------------------------------------------------------------------------------------

  Someone at - he encounter the same problem with 12.1.0.2 by deploying the CDB for Grid Infrastructure Management repository? Is there something special with UEK3 (3.8.13 - 35.3.4.el7uek) on 7 OEL? Any necessary special kernel parameter (although the runInstaller controls do not mention)? I'm totally tapped by GI 12.1.0.2.

  Thank you.

  Best regards

  Stefan

  Hi guys,.

  I was finally able to solve this problem.

  She was related to a memory on the provisioning of problem in the virtual environment as both nodes are VMs. Unfortunately none of these errors of memory were populated / pushed somehow in the virtual machine.

  Best regards

  Stefan

• How Oracle Clusterware Oracle RAC and 3rd-party Clustering with them

  I have some questions on how Oracle, Oracle RAC and 3^rd party clustering CLusterware fit between them.

  Q1. My understanding is that the Infrastructure Grid Oracle Clusterware is Oracle's Clustering solution that allows applications to cluster? that is, it is a general clusterware solution to compete with those already on the market. Is this correct?

  Q2. So why CARS is necessary for the Oracle cluster databases? An Oracle database is not just considered one application like any other? So why can not just use you GI / Clusterware to gather what you would do for any other application? Why so you need RAC to the Clusterware Summit?

  Q3. Is it is possible to combine an Oracle database using 3^rd party of clustering and not use Oracle Clusterware and CARS at all that is can you say cluster, an Oracle database using say Sun Clusters, Clustering AIX or Linux native same clustering?

  Q4. If the RAC is purely for Oracle Cluster databases - what is with the title "Real Application Clusters"? What's real on this subject?, what to ask?

  Q5. I also read that RAC can use 3^rd party of clustering. However if you decide that this is the case, then you need to install Oracle Clusterware when even (I think for interconnection must be created to allow the merger with Cache between Instances of node?).

  Is this the case?  If yes why never bore you with 3^rd party clusters? -Since you will have to install Oracle Clustering anyway - and I probably have to license (all I can think about is the scenario that you already have a cluster of 3^rd party in place and you decide that you need to use the same material for a database cluster)?

  All wisdom greatly appreciated,

  Jim

  Yes, with 10g, you had to use the CRS, and with 11g, to use the GI.  For 9i RAC, you would use Verits Cluster and Sun Cluster or anything else.

  11g RAC to GI.  As above, you can use clusterware extra if needed.  I gave the example above where to a 10g RAC we used the CRS to the cluster database/listener/vip and allowing groups of volume and filesystem in HACMP cluster.  File systems were not shared, but could switch to another node failure node, who has been treated by HACMP.  The volume of the groups containing the raw devices were shared volume groups.  For 11g, I never used extra clusterware on top of CRS and have always used ASM which is part of the GI.

  The application provider only certified database to run on 10g on AIX with raw devices.  What is strange about that?  It would be unusual that you probably expect to use ASM instead of raw devices, but it is what it is.

  See the following Note of Oracle for more information: using Oracle Clusterware with vendor Clusterware FAQ (Doc ID 332257.1)

  It is perhaps easier to explain if you can explain what problem you are trying to solve?

• Oracle RAC - TAF

  Dear all,
  
  I have a knot two Oracle - RAC configuration in place.
  
  OS: Windows Server 2003 EE 64-bit
  Database: Oracle Database 10 g Release 2 Patch - 3 (64-bit)
  
  Instance type: two-node RAC
  
  Node 1: testdb1
  Node 2: testdb2
  
  I use the following tns alias to connect to the database:
  
  TESTDB_RAC =
  (DESCRIPTION = (LOAD_BALANCE = WE)(FAILOVER = ON) (ADDRESS = (PROTOCOL = TCP) (HOST = testdb1-VIP)(PORT = 1521)) (ADDRESS = (PROTOCOL = TCP) (HOST = testdb2-VIP)(PORT = 1521)) (CONNECT_DATA = (SERVICE_NAME = testdb)(SERVER = dedicated) (FAILOVER_MODE = (TYPE = SELECT)(METHOD = PRECONNECT) (ATTEMPTS = 20)(DELAY = 3) (BACKUP = TESTDB2))))
  
  What I want to achieve:
  
  (1) when I start SQL * Plus (or any application using the alias above) and connect to the base, I expect to connect to the two instances (using the PRECEONNECT clause).
  
  (2) shoot node-1
  (3) my database connection must exist yet with the second instance (node-2)
  (4) set up the node-1
  (5) to lower node-2. Now, the connection should failover to node 1.
  
  Rather than fail to node-1, my database connection fell, and I have to reconnect to the database.
  
  We would like to have this transparency when applications to talk to the database. Applications should NOT feel failure of one of the nodes. Interruption of node should be transparent to the application.
  
  Appreciate your comments on this.
  
  Concerning
  
  Prem

  >

  What's doing BACKUP keyword in the FAILOVER clause?

  It specifies a different net service name to use for the connection of relief. A backup must be specified when you use PRECONNECT to establish connections. Specification of a BACKUP is highly recommended for the BASE methods; otherwise, it may first attempt which has just failed, adding an extension until the client reconnects.

  >

  TNS alias on the node-1 should have BACKUP = TESTDB2 while Tns alias on the node-2 should have BACKUP = TESTDB1. Please validate my statement.

  Yes... You are right...

  Please also check

  http://Stanford.edu/dept/ITSS/docs/Oracle/10G/RAC.101/b10768/example.htm

  Concerning
  Rajesh

• ORACLE RAC FAILED WITH PRVF-4007 - 11.1.0.7

  Hello
  
  We are in the middle of installing Oracle RAC on AIX 6.1 when run RAC check utility.
  
  cluvfy stage - post hwos - ud1 UD2-verbose
  
  It generates the user error
  
  Result: PRVF-4007: failure of verification of equivalence for the user 'oracle '.
  
  Post-audit for hardware and configuration of the operating system has failed.
  Controls has failed for the following nodes:
  Node1
  
  
  I did ssh, rsh from two nodes then I face issue. I modified /etc/hosts.equiv
  
  Now I can do rsh two nodes and node itself.
  
  But for ssh I can't

  Hello

  SSH node1 date (does not work)

  Try to add rsa_keys in authorized_keys and copy node2.

  See the TechNote MOS:

  * How to set up SSH for a RAC Installation [ID 300548.1] *.

  Thank you
  A H E E R X

• Oracle RAC and replacement of equipment

  Hello world

  next year our Oracle RAC (hardware) servers will be is more supported. We have a 2 node RAC (SE) running on Linux 5. For you, what is the strategy easier (and safer) to replace the equipment:

  • Reinstall complete with database exports / imports.
  • Adding new nodes to the cluster with new servers and delete 'old' (but it seems that the same version of operating system is required in accordance with the documentation).
  • Clone of the cluster according to the Oracle documentation on news servers.

  Thank you for all your ideas. I googled a bit, but I've not found much info.

  Carlos.

  • Adding new nodes to the cluster with new servers and delete 'old' (but it seems that the same version of operating system is required in accordance with the documentation).

  It's what I strive for. If your time with new nodes in the cluster is short (less than 24 hours), and you stay on the same distribution (i.e. OL5 to OL7), then several times, this isn't a problem. If you still have concerns and do not feel comfortable with this recommendation, then I'll recommend another avenue. I often used it when I was doing something similar for the production system was too critical to take no risk.

  1. Up new servers in a cluster. You can even start with the version of grid Infrastructure more later/more big who can save a future update maybe you.
     1. It will take a new OCR and voting disk new configuration.
  2. Attach the storage shared on the new cluster. For a short period of time, you will have storage mounted to both old and new.
  3. Install the RDBMS software on the new cluster. Wear on all the config files you might need more... tnsnames.ora, sqlnet.ora, etc..
  4. When you are ready to drop the database on the old cluster, cutting on the new cluster.
  5. Use "srvctl database add" to add the database to the new cluster.
  6. Use 'srvctl add instance' to add the instances to the new cluster.
  7. (Optional but normally a good idea), use 'srvctl add service' to add services to the new cluster.
     1. Steps 5 and 6 above usually use the same name of db, same instance names, same service name.
  8. You now have the database running on the new hardware. Stop the old servers

  The only other sticking point would be handling the SCAN and traditional VIP. When I made this drive, I usually my network guy change the SCAN VIP and traditional dignitaries to be alias in DNS that point to the new VIP.

  If you do your homework and have everything ready to go, the downtime can be less than 15 minutes for this approach.

  See you soon,.
  Brian