I would appreciate if someone could share how they break the various Oracle parts (pls excuse, I'm not a DBA), ie. data, logs, archives, & etc. amongst the various volumes in a NetApp cluster.
I have an Oracle instance running on a singe F840 that very shortly is going to be clustered. The Oracle binaries are in a separate qtree and the rest the DBAs put in another qtree that I created for them. The qtrees and mount are by the SID. (HR)
On the toaster /vol/db/<sid>/archs archive logs /ctl control files /data dbf files /rbs dbs files /redo redo logs /system /temp
this gets mounted on a Sun DB server as filer:/vol/db/HR /rmounts/HR /oradata/HR --> /rmounts/DB/HR The DBAs refer to the database by a slink /oradata/HR in this case.
A new mandate is to utilize the clustered F840s and to separate the database (.dbf), the redo and archive logs.
In the cluster env there are now 4 volumes, 2 on each head. One for database, the other for logs (one has snapshots the other does not). Production DB is on one head, QA on the other. filer: A filer: B ------------------------ vol DB vol DB vol log vol log
Do you folks that spread your logs end up with multiple nfs mounts per database? Do I suggest to my DBAs that I will have something like:
A:/vol/db/HR/u01/data /rmounts/HR/u01 /rbs /system /temp
A:/vol/log/HR/u02/redo1 /rmounts/HR/u02 /arch1 /ctl1
B:/vol/log/HR/u03/redo2 /rmounts/HR/u03 /arch2 /ctl2 /oradata/HR --> /rmounts/HR
So I end up with 3 nfs mounts for each instance of the database? How do you people do this? There are about 10 other DBs waiting. What about a DSS; a reporting database? At the present time with all of the DB components on one volume all the DBAs have to snap the one volume and they can start up a DSS instance.
In this new scenario, I guess they will still have to snap the DB volume and also snap one of the LOG volumes. What do you do?
Sorry for being so winded but I was looking for specifics, unfortunately I couldn't find anything on either the NOW site or the archive list. Just the White Papers that describe the concept.
Thanks, George
------------------------------------------------------------------------------- George Kahler e-mail: george@yorku.ca Sr. Systems Administrator humans: (416) 736-2100 x.22699 Computing and Network Services machines: (416) 736-5830 Ontario, Canada, M3J-1P3