Stopping Service in Cluster

• Rechana Rajan

  SSCertifiable

  Points: 7717

  More actions

  June 8, 2017 at 5:32 am

  #1945972

  Perry Whittle - Thursday, June 8, 2017 3:24 AM

  Rechana Rajan - Thursday, June 8, 2017 2:43 AM

  Perry Whittle - Wednesday, June 7, 2017 10:03 AM

  Rechana Rajan - Wednesday, June 7, 2017 8:12 AM

  Thanks perry

  Perry Whittle - Wednesday, June 7, 2017 4:05 AM

  Rechana Rajan - Wednesday, June 7, 2017 3:45 AM

  Perry Whittle - Thursday, May 11, 2017 7:13 AM

  Rechana Rajan - Tuesday, May 9, 2017 3:53 AM

  Stopped SQL service from Configuration manager in a cluster, is the failover supposed to happen?

  Will stopping SQL Server resource or the role from one node in cluster failover manager will initiate a failover?

  Out of the box, stopping the clustered sql server service from either FCM or SSCM will not initiate a failover.
  Can you provide more detail on the failover policy configuration from the cluster group and its resources please, have they been changed?
  Check your sql server error log and the cluster events for more information too

  Thanks Perry.

  I have changed the policy in FCM
  If Resource Fails,attempt restart on current node- to 1 minute and I turned off the SQL Service from SQL Configuration Manager to Test whether it will come online after 1 min but it didn't.

  stopping a clustered instance of sql server is supported either via FCM or SCM, it is not supported via services.msc

  Thanks Perry.
  As I mentioned earlier I did it from SCM

  Should be fine then

  Thanks Perry.

  If  I  do simulate failure from  FCM  it comes online immediately in same node,but if i stop SQL service from SQL Server Configuration manager it wont come online after 1 minute nor its doing a failover.
  What policy settings do you have set for the group and the network name resource[/quote]Thanks Perry.Please find the screenshots

  

  

• Perry Whittle

  SSC Guru

  Points: 234013

  More actions

  June 8, 2017 at 9:13 am

  #1946018

  Rechana Rajan - Thursday, June 8, 2017 5:32 AM

  Perry Whittle - Thursday, June 8, 2017 3:24 AM

  Rechana Rajan - Thursday, June 8, 2017 2:43 AM

  Perry Whittle - Wednesday, June 7, 2017 10:03 AM

  Rechana Rajan - Wednesday, June 7, 2017 8:12 AM

  Thanks perry

  Perry Whittle - Wednesday, June 7, 2017 4:05 AM

  Rechana Rajan - Wednesday, June 7, 2017 3:45 AM

  Perry Whittle - Thursday, May 11, 2017 7:13 AM

  Rechana Rajan - Tuesday, May 9, 2017 3:53 AM

  Stopped SQL service from Configuration manager in a cluster, is the failover supposed to happen?

  Will stopping SQL Server resource or the role from one node in cluster failover manager will initiate a failover?

  Out of the box, stopping the clustered sql server service from either FCM or SSCM will not initiate a failover.
  Can you provide more detail on the failover policy configuration from the cluster group and its resources please, have they been changed?
  Check your sql server error log and the cluster events for more information too

  Thanks Perry.

  I have changed the policy in FCM
  If Resource Fails,attempt restart on current node- to 1 minute and I turned off the SQL Service from SQL Configuration Manager to Test whether it will come online after 1 min but it didn't.

  stopping a clustered instance of sql server is supported either via FCM or SCM, it is not supported via services.msc

  Thanks Perry.
  As I mentioned earlier I did it from SCM

  Should be fine then

  Thanks Perry.

  If  I  do simulate failure from  FCM  it comes online immediately in same node,but if i stop SQL service from SQL Server Configuration manager it wont come online after 1 minute nor its doing a failover.

  What policy settings do you have set for the group and the network name resource

  Thanks Perry.Please find the screenshots

  

  [/quote]
  I see you have set an extremely low value for "Period for restarts".
  With the current settings above, uncheck "If restart is unsuccessful, fail over all resources in this Role".

  try taking the resource offline again and let me know what happens

  Have you modified any other resource policies in this cluster role?

  -----------------------------------------------------------------------------------------------------------

  "Ya can't make an omelette without breaking just a few eggs" 😉

• Rechana Rajan

  SSCertifiable

  Points: 7717

  More actions

  June 11, 2017 at 3:29 am

  #1946319

  Perry Whittle - Thursday, June 8, 2017 9:13 AM

  Rechana Rajan - Thursday, June 8, 2017 5:32 AM

  Perry Whittle - Thursday, June 8, 2017 3:24 AM

  Rechana Rajan - Thursday, June 8, 2017 2:43 AM

  Perry Whittle - Wednesday, June 7, 2017 10:03 AM

  Rechana Rajan - Wednesday, June 7, 2017 8:12 AM

  Thanks perry

  Perry Whittle - Wednesday, June 7, 2017 4:05 AM

  Rechana Rajan - Wednesday, June 7, 2017 3:45 AM

  Perry Whittle - Thursday, May 11, 2017 7:13 AM

  Rechana Rajan - Tuesday, May 9, 2017 3:53 AM

  Stopped SQL service from Configuration manager in a cluster, is the failover supposed to happen?

  Will stopping SQL Server resource or the role from one node in cluster failover manager will initiate a failover?

  Out of the box, stopping the clustered sql server service from either FCM or SSCM will not initiate a failover.
  Can you provide more detail on the failover policy configuration from the cluster group and its resources please, have they been changed?
  Check your sql server error log and the cluster events for more information too

  Thanks Perry.

  I have changed the policy in FCM
  If Resource Fails,attempt restart on current node- to 1 minute and I turned off the SQL Service from SQL Configuration Manager to Test whether it will come online after 1 min but it didn't.

  stopping a clustered instance of sql server is supported either via FCM or SCM, it is not supported via services.msc

  Thanks Perry.
  As I mentioned earlier I did it from SCM

  Should be fine then

  Thanks Perry.

  If  I  do simulate failure from  FCM  it comes online immediately in same node,but if i stop SQL service from SQL Server Configuration manager it wont come online after 1 minute nor its doing a failover.

  What policy settings do you have set for the group and the network name resource

  Thanks Perry.Please find the screenshots

  

  

  I see you have set an extremely low value for "Period for restarts".
  With the current settings above, uncheck "If restart is unsuccessful, fail over all resources in this Role".

  try taking the resource offline again and let me know what happens

  Have you modified any other resource policies in this cluster role?[/quote]

  Thanks Perry,

  I did as per your words and stopped service from SCM waited for 2 mins and nothing happened.

• Perry Whittle

  SSC Guru

  Points: 234013

  More actions

  June 11, 2017 at 3:53 am

  #1946321

  The resource stayed offline and didn't fail over?

  -----------------------------------------------------------------------------------------------------------

  "Ya can't make an omelette without breaking just a few eggs" 😉

• Rechana Rajan

  SSCertifiable

  Points: 7717

  More actions

  June 11, 2017 at 7:49 am

  #1946334

  Perry Whittle - Sunday, June 11, 2017 3:53 AM

  The resource stayed offline and didn't fail over?

  [/quote]

  Thats correct.Failover didnt  happen.

• Perry Whittle

  SSC Guru

  Points: 234013

  More actions

  June 12, 2017 at 4:04 am

  #1946368

  Rechana Rajan - Sunday, June 11, 2017 3:29 AM

  Thanks Perry,

  I did as per your words and stopped service from SCM waited for 2 mins and nothing happened.

  That's exactly what i would expect to happen.

  1 minute for the period restarts is way too low, what are you expecting this to do?

  -----------------------------------------------------------------------------------------------------------

  "Ya can't make an omelette without breaking just a few eggs" 😉

• Rechana Rajan

  SSCertifiable

  Points: 7717

  More actions

  June 12, 2017 at 4:36 am

  #1946371

  Perry Whittle - Monday, June 12, 2017 4:04 AM

  Rechana Rajan - Sunday, June 11, 2017 3:29 AM

  Thanks Perry,

  I did as per your words and stopped service from SCM waited for 2 mins and nothing happened.

  That's exactly what i would expect to happen.

  1 minute for the period restarts is way too low, what are you expecting this to do?

  I was expecting failover to another node . What if I make it as 15 minutes?