Hi Marcin, is there any documentation around the column descriptions for the configurable tables please? Many of the columns are self explanatory but I’m not completely sure on these 2 columns:
action_repeat_period_minutes and action_hourly_limit in the sqlwatch_config_check_template_action table.
Also, is there a concept of grouping checks when an action is triggered? If I have 5 failed jobs detected, can I send 1 action email which lists all 5 jobs that failed, rather than 5 separate emails?
Thanks.
Good idea about the grouping but sadly, there isn’t anything like it.
action_repeat_period_minutes:
Say you have a missing backup, the check fails and you get a notification. If you however never fix the backup, this option will “remind you” every n minutes that the check is still critical or warning.
action_hourly_limit
Max number of actions per hour. Say something goes terribly wrong and you get thousands of failures, this will limit how many emails/actions you can get per hour to avoid flood.
Perfect, thank you for the explanation.
could you explain the action_every_failure please?
In my experimentation/testing, having this set to “true” does just that: triggers the corresponding action(s) each time the check runs and fails. I believe if this is set to “false”, the action(s) will trigger once if the check fails, but will not trigger again if the check runs again and fails. It will only trigger the action(s) again once the check has succeeded after failure(s), and then fails.
Marcin can correct me if I’m wrong.
That’s correct (apologies I missed the question). This flag is to avoid