We have multiple servers running, with dozens of jobs on each one.

Each time we have a maintenance window to upgrade systems or anything else that requires us to go offline for a number of hours, it means we have to go into each job and update the triggers to have an expiration date, then after the window is over, turn back on.

We'd like to have the ability to set a window of time to have nothing run.
For example, "Thursdays from 8pm to 10pm" or "second tuesday of month, from 2am to 4 am".
The server would still be on, just nothing would run until the window was over,then everything continues on it's regular schedule.

Bonus points for including a job switch to say "include in maintenance window" to define whether to turn off or not.

We could turn off the server while the window is going on, but we've had problems with that because that means someone needs to still manually turn it back on, usually at some ghastly hour of the morning.
Also, because of the jobs we run, we don't always turn them all off. We could, but don't need to. (Hence the need for the bonus feature.)

Hi Rahn,

What you can do is make a new Time Exception called Maintenance.
Add the maintenance time in this exception and add this time exception in every job, or just the jobs you want.

During your maintenance no jobs will run and the server is still up.
You can schedule the maintenance ahead so nobody has to rush to be on time to turn the service back on.

For feature maintenance you only need to change the time exception.

So this funcionality is already in place.

Regards
Erik

---

Thanks! That seems easy enough. We'll give it a try and see how it works.

I'm going to add to this.

I have not tried, but based on the way the OP was asking for Maint Windows, I assume all of the triggers being updated were time based, or at regular intervals, and not file creation triggers, so the suggested solution will likely work for them.

I'm willing to bet that if I put a job into this suggested 'Maintenance' mode, if I have a job set up with a file trigger... and files get created during this maintenance window, that the job is not put into queue waiting for the maint window to end. By this logic, if files are placed in a directory that normally would have something done to them immediately, ONLY when new files show up, AFTER the maintenance window you would have the possibility of files sitting there that would seemingly never get processed. BEcause what is going to trigger them? They already exist, even though they didn't when the maint window began.

Does that make sense? I have jobs now since learning Variables, File Triggers, etc... that would literally not work unless the file trigger is what activates them. How would you handle something like this and does my requirement change your mind about needing to revisit missed runtimes for event triggers, etc.. due to maint windows?

Just some random thoughts for the day.

Brian