In my production database connection suddenly starts taking significant time like 10-15 seconds.
even I try to connect from same machine by using following command : sqlplus system/impetus@DB
it is taking more then 10 seconds every time
whereas if I use follwing command : sqlplus system/impetus
connection is instant (taking only few milliseconds)
We also used tnsping which is also instant ...
As per my experience...
sqlplus system/impetus :- its a Bequeath Connection you are using.
sqlplus system/impetus@DB using the srervice so take's check the hostnames
and connects to server. That's the mininum time....
Does it happen before to this like that....
Pavan Kumar N
I am having the same slow response from the 11G enviroment.
Connecting from the server locally is fast.
TNSPING from a remote client respond very fast.
Connecting from a remote client (EM, Sqlplus, etc..) are SLOW
I also created a 10G database on the same server and the connection is pretty fast.
Kindly check the os side any firewall is enable.
Last month i have faced the same issue in 10g database.
I am unable to login the database from remote.
I have use the tnsping sometime suddenly connected or take more time.
So we have raise the request to system support group.
They people checked in os side to kill the one process. (i don't know the which process).
After the database connected smoothly.
so kindly check the network/ Os side.
I experienced a similar problem to this when we moved our Oracle server from our office network in to our production network. In this case it was caused by the database server trying to connect to the DNS server in our office network which is not available.
This setting was in the /etc/resolve.conf file.
Once this setting was fixed I could login instantly using sqlplus.
The problem is of course the DNS entry (as indicated in one of the responses above) configured either at OS installation or later. Whatever the case, it is of primary improtance. On the other hand, one might have to prepare Oracle totally disconnected from the network (which is what I prefer, anyway). This is what I usually make sure off before proceeding and I also get a static IP from my administrator first (although not a must):
eth0 interface (static):
IP : 192.168.1.101
SM : 255.255.255.0
192.168.1.101 myhost.acme.com myhost (replace myhost with your actual server hostname)
The above entires are a minimum for a succesful Oracle DB installation. Assuming you have completed the installation and have a created a database, you will find all connections working "normally" and the way it should.
Finally, (after I'm satisfied alls well), I get my network administrator to give me the correct entries for GW, DNS and make the necessary substitutions in the lan interface and /etc/hosts. Or have the administrator do it for you.
Hope this helps.
Hi, could you find out from your colleague exactly what he did and post it? I have the same issue, and its not firewall related. Tnsping is instantaneous, its only logins using tns listener that are slow.
Oracle 11g seems to ignore your /etc/host.conf i.e It always uses DNS first to try to resolve ip adresses from hostnames. If you remove the nameserver from your configuration and add all servers needed by the database to /etc/host it should run fine.
I just experienced the same problem with a RAC 11g database on Linux servers. Removing all the name servers from /etc/resolv.conf did not help though. There were three DNS servers listed in this file and what did the trick was to remove the DNS server which could not be reached from the database servers. What I would like to know though if there is a setting to prevent the Oracle database to do DNS lookups. cheers