March 15, 2019 at 9:01 am
Hi
I have a question about the failover on sql 2016
I've installed a windows cluster with 2 node S1 and S2, my windows cluster is S0
I've one instance I1 on both servers with one database D1
One dag DAG1 with D1
My listener is DAGL1
Both instances are in mode Synchronous commit, with failover automatic, S1\I1 is the primary
Everything works fine.
On my desktop, I connect to DAGL1\I1 and lauch a simple query inserting every second a row in a table of D1
I do a manual failover; it works fine, S1\I1 is now secondary and S2\I1 is now primary but...
on my desktop my query is now stopped because of a disconnected state
I can connect again to DAGL1\I1 and lauch it again and it works fine, writing now on S2\I1 which is primary but is there a way to avoid the disconnect event or not ?
Many thanks 🙂
March 15, 2019 at 9:49 pm
I don't believe you can avoid a disconnect. In the best case you would get a warning instead of an error, that you have disconnected and reconnected. In real life, applications have retry logic so it should not be such an issue. The main issue is how long the outage will be during the failover
March 18, 2019 at 1:56 am
ok thanks 🙂
Viewing 3 posts - 1 through 2 (of 2 total)
You must be logged in to reply to this topic. Login to reply