We use the NTServices probe to force start a crashed windows service during normal business hours. Here's the rub I'm working through: When the Service Delivery Team wants to update this service, they have to stop it before they can do any work. If they stop it, it just restarts due to the requested force start on stop. Then I have to be paged to disable the restart and paged to re-enable it afterwards(no one wants this to happen ).
It would be amazing if maintenance mode prevented this profile from running, but it doesn't.
So 1 idea that has been floated my way is:
First you could alarm only for that service. You could run a NAS rule for that server(s) to run a script that calls the db to see if the server is in maint mode. If so, do nothing, if not, have nexec on the box and call the profile that has the restart command for that service.
This is complicated and I don't know how to get that data from the database (Postgres NIS)
Question 1: Can anyone think of another way to do this that doesn't require a monitoring admin to be paged?
Question 2: How can I get the maint info from the CA_UIM database?
Thanks in advance!
Tell the Service Delivery Team to
stop the 'Nimsoft Robot Watcher'
stop the service to be upgraded
start the service to be upgraded
start the 'Nimsoft Robot Watcher'
Seems very reasonable. I'll socialize that here.