During a recent project upgrading a clients Maximo system to IBM Power 8 processors running AIX 7.1 and Oracle 12cR1 we encountered the following issues:
- The system would after a restart, that is restarting the Host, AIX, Oracle RDBMS, the Listeners and the Weblogic Java clients, run without a hitch for a while.
- Sporadically a query would hang and the Websphere Java cliients would after a while respawn the request.
- On investigation it was found that the hanging was as a result of a sync operation on the CTXSYS indexes specifically the SYNCRN.
None of these seemed to help.
End result the CTX index was dropped and this releived the system of most of the bottleneck around the CTXSYS.SYNCRN.
This is not the solution but rather a temporary work around until the CTX index issue can be resloved.
Has anyone experienced this before and what actions do you propose I take to re-instate the CTX indexes. These indexes are vitally important for Maximo queries to function properly.