Forced failover is needed when automatic failover does not happen or not possible. Sometimes, customers choose asynchronous type for disaster recovery servers. In this case, if primary replica site fails, automatic failover does not happen and you should do manual failover. Please refer below: learn.microsoft.com/en-us/sql/database-engine/availability-groups/windows/perform-a-forced-manual-failover-of-an-availability-group-sql-server?view=sql-server-ver16
yes, you should create quorum witness as best practice. If you do not have quorum witness, there will be 50% probability that your whole cluster fails when one node goes down. Please refer below official doc: learn.microsoft.com/en-us/azure-stack/hci/concepts/quorum#examples
@@arslanov1886thank you very much for the prompt and detailed reply. our environment is on a vxrail. what will be the best witness to use? will a file share witness created on a nas storage work. have you got any good videos for configuring the windows failover cluster. if yes, please share.
As long as you have a constant and stable network between your servers and witness, you can choose any type of witness. There are other small points. Please refer below: learn.microsoft.com/en-us/windows-server/failover-clustering/manage-cluster-quorum#witness-configuration Please make sure that the witness is in another server. I do not have videoes but I would follow the below good docs: Step 1 learn.microsoft.com/en-us/azure/azure-sql/virtual-machines/windows/failover-cluster-instance-prepare-vm?view=azuresql&tabs=single-subnet Step 2 learn.microsoft.com/en-us/azure/azure-sql/virtual-machines/windows/failover-cluster-instance-azure-shared-disks-manually-configure?view=azuresql
Thanks for this. At last I understood AG. I have never seen this kind of detailed video for AG internals. This is very good channel for internals :)
Good session mate..please make the videos of basic troubleshooting in AOAG..thank you 🙂
Sure mate. Thanks for suggestions
cool explanation.. thanks!!🙂
Glad it was helpful!
Great content and balance between concepts and practical example!
Thank you. Glad it helped🙂
Very useful information. Thanks a lot 🙏
Most welcome 😊
Why forced manual fail over needed sir?
Forced failover is needed when automatic failover does not happen or not possible.
Sometimes, customers choose asynchronous type for disaster recovery servers. In this case, if primary replica site fails, automatic failover does not happen and you should do manual failover. Please refer below:
learn.microsoft.com/en-us/sql/database-engine/availability-groups/windows/perform-a-forced-manual-failover-of-an-availability-group-sql-server?view=sql-server-ver16
when we create windows failover cluster for SQL AG, do we need to have quorum disk / file witness? what is the best practice for the same.
yes, you should create quorum witness as best practice. If you do not have quorum witness, there will be 50% probability that your whole cluster fails when one node goes down. Please refer below official doc:
learn.microsoft.com/en-us/azure-stack/hci/concepts/quorum#examples
@@arslanov1886thank you very much for the prompt and detailed reply. our environment is on a vxrail. what will be the best witness to use? will a file share witness created on a nas storage work. have you got any good videos for configuring the windows failover cluster. if yes, please share.
As long as you have a constant and stable network between your servers and witness, you can choose any type of witness. There are other small points. Please refer below:
learn.microsoft.com/en-us/windows-server/failover-clustering/manage-cluster-quorum#witness-configuration
Please make sure that the witness is in another server.
I do not have videoes but I would follow the below good docs:
Step 1
learn.microsoft.com/en-us/azure/azure-sql/virtual-machines/windows/failover-cluster-instance-prepare-vm?view=azuresql&tabs=single-subnet
Step 2
learn.microsoft.com/en-us/azure/azure-sql/virtual-machines/windows/failover-cluster-instance-azure-shared-disks-manually-configure?view=azuresql