Please start any new threads on our new
site at https://forums.sqlteam.com. We've got lots of great SQL Server
experts to answer whatever question you can come up with.
| Author |
Topic |
|
cz
Starting Member
19 Posts |
 Posted - 2010-09-14 : 19:27:16
|
| Hi,I was wondering the most common reasons resulting in failover in a Windows/SQL clustering environment?I know a few of them: 1) too busy CPU/Memory ( easy to fix, not likely the common reason) 2) SAN ? (How to know this? not likely either, since it should be fixed before) 3) hardware failure, ie NIC, PCI (what possibility here? are they major reasons?) 4) bad SQL queries , consuming almost all resources? (this is the major reason, isn't it?)Glad to share experiences and thoughts!cz |
|
|
tkizer
Almighty SQL Goddess
38200 Posts |
Posted - 2010-09-14 : 19:30:49
|
| I've never had a failover as a result of CPU/memory issues, and I don't think this is even a possibility unless the heartbeat gets affected.If a SAN issue occurs, then it's likely a failover will fail unless the disks can be brought online on the other node.Hardware failure is the most common cause of an unplanned failover.Bad SQL queries should never cause a failover.Most failovers should be planned events, such as you need to perform maintenance on one of the nodes (such as a service pack installation for the OS).Tara KizerMicrosoft MVP for Windows Server System - SQL Serverhttp://weblogs.sqlteam.com/tarad/Subscribe to my blog |
 |
|
|
BISWAJIT74
Starting Member
5 Posts |
Posted - 2010-09-21 : 18:22:14
|
| Here are few which I face with my 200 cluster server 1)Newtwork glitch on the domain 2)The domain server have soem network hickup3)SAN did not respond due to any issueBiswajit Das |
|
|
|
|
|
|
common reasons for failover in clustering