Win a copy of Event Streams in Action this week in the Java in General forum!

Hareesh Ram Chanchali

Ranch Hand
+ Follow
since Jan 31, 2005
Cows and Likes
Cows
Total received
0
In last 30 days
0
Total given
0
Likes
Total received
0
Received in last 30 days
0
Total given
0
Given in last 30 days
0
Forums and Threads
Scavenger Hunt
expand Ranch Hand Scavenger Hunt
expand Greenhorn Scavenger Hunt

Recent posts by Hareesh Ram Chanchali

Hi Easwar,

You can directly put the same at the respective directories(exploded mode in weblogic) and try to hit 'redeploy' button.

Alternatively there is a short cut for jsps...you can directly put the same under staging directory of the managed server and restart the server.
If it is cluster, you need to restart all the servers. Hence this will be used for development modes effectively.

Please let me know if it worked or not.
10 years ago
Hi Jan/ Adeeb,

Thanks for the suggestions.
Just wondering about the performance point of view.

Also can con.setAutoCommit(con.getAutoCommit()) be used and catch the SQL exception to know if Connection is still valid or not?

Does the later one improves the efficiency in checking?

Regards
Hareesh
Will you be able to see that these message time stamps has some pattern?
Like at a specific time ? Any idea if your Managed servers are being shutdown abruptly? seems to be mostly a message in Fail-over scenario.
10 years ago
The scenario is like this.

A connection pool was configured with application server
Minimum number of connections > 0 were configured.
Hence they will be instantiated at server start up.
I used some of the connections and closed them except one.
Meanwhile DB went down.

The requirement is I want to check if this is a valid connection or not with out using a query (possibly with DUAL) and without using configurable options in application servers ( like TestonReserve).

What could be the simplest way without using the query to check the health of the connection as I need to make sure that it is healthier for next set of steps?

Also for the connection in the above scenario, if I cross check the same with null value, will it validate whether it is healthier or not?
also if I close the connection, will it somehow raise an exception as DB is down? or it will simply logically close the connection and release the same to connection pool irrespective whether DB is up or not?

Apprecaite your insights.

Regards
Hareesh
The scenario is like this.

A connection pool was configured with application server
Minimum number of connections > 0 were configured.
Hence they will be instantiated at server start up.
I used some of the connections and closed them except one.
Meanwhile DB went down.

The requirement is I want to check if this is a valid connection or not with out using a query (possibly with DUAL) and without using configurable options in application servers ( like TestonReserve).

What could be the simplest way without using the query to check the health of the connection as I need to make sure that it is healthier for next set of steps?

Also for the connection in the above scenario, if I cross check the same with null value, will it validate whether it is healthier or not?
also if I close the connection, will it somehow raise an exception as DB is down? or it will simply logically close the connection and release the same to connection pool irrespective whether DB is up or not?

Apprecaite your insights.

Regards
Hareesh
HI Vishy,

I am not sure about 7.0 but I worked with 8.1. There are couple of thing you need to do.

First select the server > monitoring > then monitor all the queues and then choose default queue and monitor all the threads and see what exactly happenning with the data you can find there.

next, through access logs or by any other analytic mechanism if available try finding the maximum number of users at given instant of time say 2 minutes period.

now this count should be less than or equal to number of managed servers multiplied by number of execute threads per server.( unless you are creating user threads explicitly)

If this calculations holds good then you need to see why threads are not sufficient? sometime some of the threads are not prpoerly released back and hence they will be consuming your execute queue.

Also take thread dumps and analyse which threads are taking more time. Dont increase the threads just like that even though WLS advises. Also you can think of putting dynamic expansion of thread count but it needs some over head too!
10 years ago
I strongly agree with Marcos.

We need to take thread dumps at regular intervals and should analyse them.

You take the thread dumps at an interval of 1 or 2 minutes and take the help of samurai tool to analyse the thread dumps. Then you will be able to know what exactly happenning and possible reasons

1) DB locks
2) Native thread problems if any
3) Huge amount of data being pulled up
4) extended time in getting responses from interface systems

with all these try checking the Garbage collection also.
10 years ago
I hope nodemanager service or daemon is running and possibly the servers were started.
Would you be able to replicate the same so that we will have clear understanding?
10 years ago
Hi Rahul,

I believe it should config.xml

10 years ago
Hi Samal,

Using in-memory replication, WebLogic Server copies a session state from one server instance to another. The primary server creates a primary session state on the server to which the client first connects, and a secondary replica on another WebLogic Server instance in the cluster. The replica is kept up-to-date so that it may be used if the server that hosts the servlet fails.

(from docs) read more at


http://download.oracle.com/docs/cd/E11035_01/wls100/cluster/failover.html#wp1022034


ALso the flow chart at the end gives you good idea of Failover with proxy plug-in


http://download.oracle.com/docs/cd/E11035_01/wls100/plugins/nsapi.html
10 years ago
Hi Murali,

Have you had any static page as a default?

Also can you replace the local host with ip address and try?

10 years ago