August 18, 2015 at 10:16 am
find out why the data movement is being suspended in the first place, can you provide more detail as to what exact messages you receive, stop the sync issues to resolve the problem
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 18, 2015 at 8:08 pm
Hello Perry,
Thanks for your reply.
I found another database out of sync on another server, I tried to apply the log backup but there are not enough log backup since then, so I restored the full backup made last week and applied log backup, then run the command:
alter database DBNameset hadr availability group =[AGName]
Got error again:
Msg 35250, Level 16, State 7, Line 1
The connection to the primary replica is not active. The command cannot be processed.
Because this is production database, I have to fix this in weekend.
re the first place where the issue happens, I can only find some connection issue like below:
A connection timeout has occurred on a previously established connection to availability replica 'C\INSTANCE2' with id [119BD57D-BDE1-454A-9A87-31BE2C99C94E]. Either a networking or a firewall issue exists or the availability replica has transitioned to the resolving role.
A connection for availability group 'ORD20-I2G10' from availability replica 'A\INSTANCE2' with id [5F1D5F50-44D8-4529-9592-5041A9D60855] to 'C\INSTANCE2' with id [30A1B59E-EDF3-4142-ABC1-FC1D03E3EFBC] has been successfully established. This is an informational message only. No user action is required.
Those 3 nodes are directly connected and the speed is very fast( we spend million dollars on hardware), because we have multiple instances running on each node(4 instances on this HA environment), each instance has configured cpu affinity, I suspect the overlapped cpu cores are busy on one instance and could not handle the request for another instance, which causing AG connection dropped, but I haven't approved this yet.
August 19, 2015 at 2:05 pm
are your backups preferred on the primary or a secondary?
how often do the full backups run?
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
August 20, 2015 at 2:03 am
Our application does the backup, so all through primary, and there are both primary replicas on both node A and B, so it is balanced...
August 20, 2015 at 6:21 am
i1888 (8/20/2015)
there are both primary replicas on both node A and B, so it is balanced...
Yes, but they'll be members of different AlwaysOn groups. Keep your AG backups segregated from the remaining instance database backups, a simple check when the backup job runs to see if HADR is set and if so which group it's in then backup to a set share would be my preference.
If no full backup has run since the last sync between the primary and the secondary you may use a differential backup on the primary to bring the secondary in sync then resume data movement
-----------------------------------------------------------------------------------------------------------
"Ya can't make an omelette without breaking just a few eggs" 😉
Viewing 5 posts - 1 through 6 (of 6 total)
You must be logged in to reply to this topic. Login to reply