December 3, 2013 at 2:12 pm
I read an article called ReplicationAndDBM.docx maybe some recognize it?
Any way I followed the steps and whenever the publication fails over to it's mirror, the publication says retrying in the replication monitor, and if I refresh the publisher the publication disapears from the screen. If I fail back, they reappear and replication continues. The only thing it says to do is add the PublisherFailoverPartner parameter to all the agents involved. I Failed over before setting up the parameter, and it did the same thing so it's like it's ignoring the parameter.
This is the setup I have.
sqltest01 - Publisher/Principal
sqltest02 - Mirror
sqltest03 - Witness/Distributor/Subscriber
I have the publisher database mirrored between 01 and 02, the subscribing one is not mirrored and is on03.
So for the PublisherFailoverPartner I put in sqltest02. I know there is a lot more information to include. I pretty much just used the defaults when I setup transactional replication. Is there something else I need to do besides the failoverpartner parameter? Or is there another whitepaper I could follow?
December 3, 2013 at 2:34 pm
I have used that same whitepaper with great efficacy. I would double check that you have both publishers (principal and mirror) configured to use the same distributor with identical settings. Also make sure that the profile you setup to use the PublisherFailoverPartner is the configured profile for the agent (Log Reader).
December 4, 2013 at 8:39 am
Well shoot. I guess I will have to try following it again for the third time. The example uses four servers, I only have 3 to work with.
I have checked and both publishers are using the same distributor. I used sp_helpdistributor on both and all values returned matched. How could be setup differently? When creating a publication you only have to do that once so it would be impossible to setup it up differently. The same for the subscription, I only have to set that up once.
Do you know if I have to use the fqdn when specifying the partnerfailover value? I have just been using the hostname.
December 4, 2013 at 8:49 am
Ok well that's just annoying.
I failed over one more time before tearing down replication to start again and now it's working. I hadn't changed anything. Thanks for the tips any how.
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply