Ok, well, just to save yourself time, you should grab the cluster.ccs file, and any gfs-related log messages from the system logs. That can help point out any misconfigurations you might be running into. Matt ________________________________ From: Crisler, Jon [mailto:Jon.Crisler@xxxxxxx] Sent: Wednesday, February 04, 2009 10:46 AM To: Matthew Zito; ORACLE-L Subject: RE: RedHat GFS and 10g poor performance I will try to get that info for you- I am not familiar with GFS setup but I am working with a bunch of sysadmins ________________________________ From: Matthew Zito [mailto:mzito@xxxxxxxxxxx] Sent: Wednesday, February 04, 2009 10:39 AM To: Crisler, Jon; ORACLE-L Subject: RE: RedHat GFS and 10g poor performance It's probably lock-related, if the SAN appears fine. What's your GULM configuration look like? Matt ________________________________ From: oracle-l-bounce@xxxxxxxxxxxxx [mailto:oracle-l-bounce@xxxxxxxxxxxxx] On Behalf Of Crisler, Jon Sent: Wednesday, February 04, 2009 10:32 AM To: ORACLE-L Subject: RedHat GFS and 10g poor performance Subtitle- Tuning GFS for 10g ? I am in the final testing stages of a RedHat 5.1 active / passive cluster, using GFS and 10.2.0.4. The database is fairly large. I am seeing very poor performance, apparently due to disk i/o issues and I cannot pin it down to the cause. Its connected via 4gb SAN, and AWR is reporting datafile service times that range from 10ms to over 80ms, which is very poor. However, the backend SAN is reporting much better service times. Disk I/O seems to be far worse than a non-GFS system- perhaps up to 10 times worse. My feeling is that it is something odd with GFS. Has anybody run into performance problem with 10g on GFS ?