Thank You Lyle,
We have the ranking configured to A as 1 and B as 2.
We shut down server A and it did not fail over to B. Wow according to what you 
stated this should of worked.
Is there anything else we should check?
Also should those processes be in a running state on the B server prior to the 
fail over from A or will they get started automatically?

In this case they were already started, however I want to be sure about that 
point also.
 
Thank  you for the input on the Bug also. I would think it should then also.


Thank you,

William Abdo

-----Original Message-----
From: Action Request System discussion list(ARSList) 
[mailto:[email protected]] On Behalf Of Lyle Taylor
Sent: Friday, April 16, 2010 5:20 PM
To: [email protected]
Subject: Re: How to accurately test fail-over

Shutting down AR Server A should do it if you have your operations rankings 
configured correctly.  If it doesn't fail over (note that it can take a couple 
minutes to fail over), then it sounds like the other servers don't think that 
they can take over those operations (for example, if their rank is null).

Note that I have not consistently seen failover if one of the sub processes has 
died - e.g., AR Server A is up, but the e-mail process on that server has died. 
 We filed a bug on that a while back - it was supposedly fixed, but I haven't 
tested it, so I can't vouch for it.

Lyle

-----Original Message-----
From: Action Request System discussion list(ARSList) 
[mailto:[email protected]] On Behalf Of William Abdo
Sent: Friday, April 16, 2010 3:04 PM
To: [email protected]
Subject: How to accurately test fail-over

Hello  All,
Do any of you know of a tried, tested and accurate method to test
fail-over for multiple AR Servers in the server group?
Some of the many functions that need to be tested are:
- AR Server
- Email
- Flash Boards
- Approval Server

To see if the system would fail over from A to B:
I have tried killing the process on A, and that proved unsuccessful.
I have tried Shut down the A server and that was also unsuccessful.  

I am looking for a method that is tried and tested to accurately test
fail-over

Please, any assistance is appreciated.

Thank you, 

William Abdo

_______________________________________________________________________________
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
attend wwrug10 www.wwrug.com ARSlist: "Where the Answers Are"


 NOTICE: This email message is for the sole use of the intended recipient(s) 
and may contain confidential and privileged information. Any unauthorized 
review, use, disclosure or distribution is prohibited. If you are not the 
intended recipient, please contact the sender by reply email and destroy all 
copies of the original message.

_______________________________________________________________________________
UNSUBSCRIBE or access ARSlist Archives at www.arslist.org
attend wwrug10 www.wwrug.com ARSlist: "Where the Answers Are"

Reply via email to