Hi Joan,
Please ignore all lock types other than ST and CF. Please capture all sessions.
It is important to see which session is holding the CF or ST lock when others
are waiting.
Sid 111 has had a TX lock for 1007 seconds, and has been waiting for the CF
enqueue for 3 seconds - this is because
Hi Steve,
I still couldn't figure out why we have cf enqueue lock. As you suggest, I
run your script repeatedly to see how long the problem persists. All the cf
lock associated locks are tx and tm. Here I choose 4 sid to run. I didn't
find any st lock, but we do have ts and MR lock.
RESOURCE
Hi Joan,
You don't have a problem unless there is a session WANTING a lock on the
resource. By "something interesting", I meant a session waiting for a lock on
either the CF or ST enqueues. Then you run it repeatedly to see how long the
problem persists. Because you are going to run it repeatedly
Steve,
Thank you so much for taking time to reply. I did run couple of times the
enque_lock script as you suggested. CF waiting is really minimum. I noticed
the maximum enqueue comes from MR which is Media Recovery in share lock.
Should I need to concern this or this is just a very normal symptom
Hi Joan,
This filtered into my Inbox folder because you mentioned my name. I've not been
following the thread, but I'll comment on the output below.
Normally when seeing CF and ST enqueue waits and a lot of time spent in
'enqueue' waits I would suspect the ST enqueue to be the problem, rather th
Joan,
What do I suggest? Well for one I have no idea is async io is enabled or
not on our HP's. When I last discussed it with my SA he was not too encouraging
& after a pile of reading, neither was I. ASYNC IO only allows DBWR to pass a
block or set thereof to the OS and move on. Yeah it
Dick,
Thanks, so what you suggest? That's why the other day I asked the list about
the asyn i/o on os. If asyn is enabled, is it ok to set dbwr_io_slaves? I am
not sure for that part. For checkpoint, You see, before I made the change,
we have 8 log switch per hr plus every 30 min and every 3200 bl
Oh, some more info. We have 2 control file, all the disk just have 1
datafile. The company spend a lot money invested on this. All the redolog on
separated disk too. all raw disk and mirrored. I want to know what I can do
to reduce control file lock?
Joan
-Original Message-
Sent: Thursda
Dick,
The system is not in archivelog mode. If the database crashed, I was told
they don't care at all. They just rebuilt a shell database and let getter
and feeder processes gradually feed the database. (The getter will check the
url in the database, if didn't find then will go to the origin serv