scotthowe.com - oracle, mssql, and general systems administration blog...

  • my portfolio
  • my code-projects
  • xml sitemap
  • contact
Home

Navigation

  • Recent posts

Primary links

  • My Portfolio
  • My Code-Projects
  • XML Sitemap
  • Contact

Flickr recent photos

Swimmerafterbathcheeriosdirtyclothes

User login

  • Create new account
  • Request new password

racgwrap: line 55: fg: no job control

scott — Wed, 06/04/2008 - 06:10

While working in a development environment I came across the need to setup a clustered oracle home (Db_2) on a box with an existing single instance (Db_1). After the cluster ready services (crs) completed and I ran through the db setup I went to run netca to setup the clustered listeners. Unfortunately, the creation of the listener failed with an odd CRS error message. I canceled out of the installer and attempted it again with the same outcome. At this point I decided to check out crs and see what was going on. A crs_stat -t revealed that two listeners were installed but had a status of unknown.

At this point a little background information would be useful. I had been building up a development Real Application Cluster (rac) cluster using a clone of an existing production 10.1.0.5 instance. After running through the appropriate steps found in the guide www.oracle.com/technology/pub/articles/chan_sing2rac_install.html it was time to create the listeners using netca. Unfortunately, the oui produced some errors and failed the install. A crs_stat revealed that it had installed the listeners, but their states were unknown and would not start. I attempted to perform a crs_unregister for each of the listeners in the rac but that too failed. I thought, duh - its failing because the application is running so let’s stop it. A srvctl revealed the following:

[oracle@lubbd01dev bin]$ srvctl stop nodeapps -n lubbd01dev
CRS-0210: Could not find resource ora.lubbd01dev.LISTENER_LUBBD01DEV.lsnr.
lubbd01dev:ora.lubbd01dev.LISTENER_BLACK_LUBBD01DEV.lsnr:/u01/app/oracle/product/10.1.0/Db_1/bin/racgwrap: line 55: fg: no job control
lubbd01dev:ora.lubbd01dev.LISTENER1_LUBBD01DEV.lsnr:/u01/app/oracle/product/10.1.0/Db_1/bin/racgwrap: line 55: fg: no job control
lubbd01dev:ora.lubbd01dev.LISTENER.lsnr:/u01/app/oracle/product/10.1.0/Db_1/bin/racgwrap: line 55: fg: no job control
CRS-0216: Could not stop resource ora.lubbd01dev.vip.

This was a peculiar error to receive but I noted the db home was mentioning Db_1. Because the rac was running 10.1.0.5, I used the 10.1.0.3 installer to install the crs as well as creating a new oracle home in Db_2. From here I was able to run the 10.1.0.5 patchset to upgrade the crs and the Db_2 home. I then copied over the contents of Db_1’s dbs and was able to startup the database, but apparently I didn’t eliminate all traces of calls to the Db_1 home.

To temporarily correct the issue I created a symlink in Db_1/crs to Db_2/crs… This allowed me to use crs_unregister and rerun netca with successful results. I still need to go back and find where it’s still referencing the old oracle home…

  • database administration
  • oracle
  • Login or register to post comments

Search

External Links

  • Richardfoote.wordpress.com
  • xkcd.com

Photo Galleries

  • 08 Spring Flowers
  • Baby Matthew
  • my portfolio
  • my code-projects
  • xml sitemap
  • contact