04-29-2016 07:16 AM - edited 04-29-2016 07:17 AM
Let's say it's Monday and I have a rule to check if 'no setup express' is set, and the remediation for that is a triggered script that is sceduled for Sunday. Let's say for whatever reason that that rule fires five times against insecure.device.com. Will there be 1 schedule job to fix that device on Sunday or 5? NetMRI 6.9.4.
05-31-2016 05:24 PM
Russ -- yes, if you configure a script to be triggered by a Policy Rule Failure (or any other Issue, for that matter), a separate copy of the script will be triggered every time a new instance of that policy rule failure occurs, Generally speaking, an issue instance is considered "new" if it is not a direct continution of the last time that issue fired for the same device.
In other words -- if a device fails this policy rule on Monday, and fails it again for every subsequent day, then those subsequent failures will not cause a "new" Policy Rule Failure issue to fire, but simply increases the "last seen" timestamp of the current one. Only the first failure will trigger the remediation script.
But if the device failed on Monday, passed on Tuesday, and then failed again on Wednesday, then those would be two separate instances of the issue, and the remediation script would be triggered twice.
This can potentially lead to undesirable situations, where multiple identical scripts are scheduled to run for the same device at the same time.
A safer (but more complicated) approach would be to create a script that checks for the current status of this particular Policy Rule Failure issue, and then runs or schedules the remediation script for those devices that the issue is active for at that time.
06-01-2016 04:52 AM
What if we use activity windows and resolve right away? Let's say we are using the off hours one. So a device fails a pocliy rule violation during the day, but since that failure is *not* during the activity window, the triggered job will not fire. Will it fire once we enter the activity period, or is there a delay till the next policy run (and is that say, within an hour)?
Also, how many jobs can run at the same time? Let's say we have 7500 devices in violation of a policy. How many of those actually run at either the scheduled time or when the policy rule fails during an activity window?
It seems like triggered jobs should always check if the issue is still there. Is there an RFE for that?