| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Mailing Lists -> Oracle-L -> RE: RAC, fail over and timeout .. Big Question
If your devs are using ADO, do you know if they've tried setting the
.CommandTimeout property of the relevant Connection or Command object? I'm
not sure what the analogue is in ADO.Net, tho I'd bet there is one...
Cheers,
-Roy
Roy Pardee
Programmer/Analyst/DBA
SWFPAC Lockheed Martin IT
Extension 8487
-----Original Message-----
Sent: Thursday, June 12, 2003 12:02 PM
To: Multiple recipients of list ORACLE-L
Thanks KG, Greg,
What I visualize is this ...
Specify a (sort of) timeout value. If query doesn't come back within this value, raise alert and fail over to the other side. If you get the same problem, raise the hell out of everyone and go to cache mode (good for 3-5 minutes). Hope the problem gets fixed by then.
The biggest question is how to timeout a query if it doesn't return in specified time? We got some good VB coders ... so any ideas with VB are also welcome.
Pre-connecting is something we are looking at, but sometimes application performs DML as well. So ... the plot thickens ...
I am still researching ...
Raj
It would be interesting to see if the pre-connect helps you out or not.. We are getting ready to work thru similar issues/testing...
Greg
-----Original Message-----
Sent: Thursday, June 12, 2003 12:30 PM
To: Multiple recipients of list ORACLE-L
Raj:
I am afraid, there are not much alternatives you can think of. Even the
SELECT fail over
will take around few seconds since it has to attach the PGA to the second
instance and run the query from where it is failovered, though you can
overcome
this by enabling pre-connect.
Pls let me know if you got any nice ideas ;)
Best Regards,
K Gopalakrishnan
-----Original Message-----
Jamadagni, Rajendra
Sent: Thursday, June 12, 2003 7:50 AM
To: Multiple recipients of list ORACLE-L
Hi all,
here is a (hopefully) tricky one ...
We have a two node RAC (9202), two applications run on either side, no
problems there. We have a business critical process that runs all the time
on node T2. This process needs to have a set response time or it affects
business. And it is written in VB. Currently we have already handled the
situations when a node is down (or machine is down) it fails over to the
other side and continues.
What we need to cater for is when DB is up, but sick (i.e. not responding).
We need to be able to specify a timeout in the queries and when we get no
response in the specified time, we need to automatically fail over to the
other side.
The queries in the application are optimally written, they are sub-second or
single digit second queries. So, a query taking a long time would be about
10-15 seconds.
While we are fishing for ideas, has anyone implemented anything like this?
Any ideas?
Thanks in advance
Raj
Fat City Network Services -- 858-538-5051 http://www.fatcity.com San Diego, California -- Mailing list and web hosting servicesto: [EMAIL PROTECTED] (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing). Received on Thu Jun 12 2003 - 15:05:09 CDT
---------------------------------------------------------------------
To REMOVE yourself from this mailing list, send an E-Mail message
 |
 |