Real Application Clusters | Forum profile

crs_stat should be under the ORA_CRS_HOME env variable in the bin folder. ORA_CRS_HOME is usually @ the $ORACLE_HOME/ORACLE_CRS_HOME What gives you echo $ORACLE_HOME and echo $ORACLE_CRS_HOME ?

ca107207 wrote: Short answer is money. These are very small databases and I don't have the budget luxury of running one per server. Oracle, especially RAC, is expensive so we do the best that we can with what we have. That does not make sense. You do not have enough money for a couple of servers, one per db instance - yet there are sufficient funds for several RAC servers, shared storage, Interconnect switch and RAC licensing? What are...

I realized that the ORACLE_SID was set to the same value for both the nodes in the .bash_profile file. After mdification, the stoping and starting the instances is returning no errors. But can you plz let me know why I am not able to exe the crs_stat commands.

hi all. i hv 2 db servers in RAC environment. my server 1 is showing a lil strange behavior since last couple of weeks. its /u01 directory used to be at about 64% (total size 28G) and it remained constant for about a year .. now it has startd increasing slowly. i clear the core files generated in cdump thru script automatically. but then where its size is being used, how to find where un necessary files r being generated? can someone please...

can anybody help to check if this is network HB error or DISK HB error? thank you ocssd.log as below: CSSD 2009-11-03 14:21:24.651 14 >WARNING: clssnmPollingThread: node pcddbb (2) at 50 2.000000e+00artbeat fatal, eviction in 14.771 seconds CSSD 2009-11-03 14:21:24.651 14 >TRACE: clssnmPollingThread: node pcddbb (2) is impending reconfig, flag 1037, misstime 15229 CSSD 2009-11-03 14:21:24.651 14 >TRACE:...

From the information it looks like the VIP either the VIP is not configured correct or may have to relocated to its correct node. Try restarting CRS on one of the nodes to see if the VIP gets relocated: ora. node1 .vip application ONLINE ONLINE node2 ora. node2 .vip application ONLINE ONLINE node1

Hello, Same problem for me. Also on Solaris x86-64, 10.2.0.3. 2009-11-05 20:48:37 evmd: Remote accept socket IN_ADDR_ANY is open 2009-11-05 20:48:37 evmd: Warning: no base template for authorized event myco.myapp. 2009-11-06 09:12:39 evmd: Warning: no base template for authorized event myco.myapp. 2009-11-06 09:12:39 evmd: Remote accept socket IN_ADDR_ANY is open 2009-11-06 09:12:39 evmd: Warning: no base template for authorized event...

Finally, it is fixed. But I am not sure what is the problem. I just followed the metalink note 333611.1. to remove and add all instance back: srvctl remove instance -d demarc -i demarc1 -n ldcprdb06 srvctl remove instance -d demarc -i demarc2 -n ldcprdb07 srvctl remove instance -d demarc -i demarc3 -n ldcprdb08 srvctl remove database -d demarc srvctl add database -d demarc -o /oracle/product/10g srvctl add instance -d demarc -i demarc1 -n...

Hi buddy, What is the proper way to shutdown (not restart) a node ? You can stop the instance, the asm, the nodeapps and then stop the crs. the commands are: srvctl stop instance -d <DB_NAME> -i <INSTANCE_NAME> srvctl stop asm -n <NODE_NAME> srvctl stop nodeapps -n <NODE_NAME> crsctl stop crs Besides, after we issue crs_ctl stop crs, how could we failover the vip to node 2 ? You...

You are right, rac is not supported in solaris zones/containers. I think what you are suggesting (rac instance and clusterware in global zone, and some single-instances in containers) is supported, but please consider that your rac database might get evicted from the cluster for some reason or another which means a server reboot that would also take down all the singe instances. If you are on SPARC CMT you could use LDOMs for this. Bjoern