-
low open file
Hi
I got a 3 node cluster implemented with OS Linux and 10g RAC
Code:
uname -a
Linux XXXX 2.6.9-34.ELsmp #1 SMP Fri Feb 24 16:56:28 EST 2006 x86_64 x86_64 x86_64 GNU/Linux
[root@xxxx etc]# sysctl -a | grep fs.file-max
fs.file-max = 1599224
I know fs.file-max is not set to oracle recoomended value (fs.file-max = 65536) rather it is still the default value.
Now, when I start oracle i get the message in alert log as
WARNING:Oracle instance running on a system with low open file descriptor limit. Tune your system to increase this limit to avoid
severe performance degradation.
Code:
[oracle@xxxx ~]$ ulimit -a
core file size (blocks, -c) 0
data seg size (kbytes, -d) unlimited
file size (blocks, -f) unlimited
pending signals (-i) 1024
max locked memory (kbytes, -l) 50000000
max memory size (kbytes, -m) unlimited
open files (-n) 131072
pipe size (512 bytes, -p) 8
POSIX message queues (bytes, -q) 819200
stack size (kbytes, -s) 10240
cpu time (seconds, -t) unlimited
max user processes (-u) 131072
virtual memory (kbytes, -v) unlimited
file locks (-x) unlimited
So, can anybody tell me from where the warning originates? And what is the solution?
Last edited by Thomasps; 07-11-2006 at 06:32 AM.
-
what is the db_files set to - you can lower that if it set too high for what you have and the message shuld go away
-
as per the metalink fs.file-max > (db_files * 2) + control files + log files + over head. In my case it is 2000 data files and total comes to max: 4500 and only one database in the cluster. See my OS limits are far far above that.
-
well ive had the same, reduce db_files to the correct value for what i actually have and the error went away
-
Well, I changed the db_files from 2000 to 500.. still same message
Code:
Tue Jul 11 15:04:58 2006
cluster interconnect IPC version:Oracle UDP/IP (generic)
IPC Vendor 1 proto 2
Tue Jul 11 15:04:58 2006
WARNING:Oracle instance running on a system with low open file descriptor
limit. Tune your system to increase this limit to avoid
severe performance degradation.
I am missing something some where
-
Are you using CRS to start the cluster?
There is an oracle bug that is fixed in 10.2.0.3
workaround:
edit init.crsd
ulimit -n unlimited
Then restart crs to get the new settings.
-
Thanks.
yes, I am using CRS to start the cluster.
Are you refering to the bug: 4462367?. Many Thanks.
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
Click Here to Expand Forum to Full Width
|