High Priority Alerts not raised
nateirvin
Posts: 3
Last night, we had some sort of issue with our SQL Server - all apps failed to connect for a period of about 6 hours. When I look at SQL Monitor for this timeframe, it is clearly missing data for that time period, yet no alerts were raised. I would have expected a "Machine unreachable" or a "Monitoring error (host machine data collection)" alert to be raised, but nothing.
The SQL Monitor base service and the SQL Monitor website both run on the same machine, which is different from the SQL Server machine, on the same private network (and Monitor connects using the private IP).
Any ideas why the alerts were not raised? Is this a problem with my setup or configuration, a bug, or a 'feature'?
EDIT:
We now know that the reason our apps couldn't connect was because of bad network cables. I'm not worried about the issue, I'm wondering why SQL Monitor didn't detect it.
The SQL Monitor base service and the SQL Monitor website both run on the same machine, which is different from the SQL Server machine, on the same private network (and Monitor connects using the private IP).
Any ideas why the alerts were not raised? Is this a problem with my setup or configuration, a bug, or a 'feature'?
EDIT:
We now know that the reason our apps couldn't connect was because of bad network cables. I'm not worried about the issue, I'm wondering why SQL Monitor didn't detect it.
Comments
If there was some connectivity then the alerts may not have fired. The monitoring status page indicates that those alerts are based on ping working and the remote registry getting a response at least every 2 minutes... I've request log files and a query result from the raised support call and will update here with further information.
Kind regards,
Alex
Have you visited our Help Center?
I'm putting what we've discussed in the support call here for others to take advantage of. The issue is due to the connection from SQL Monitor to the data repository being down (which was located on the remote server being monitored). In this instance the alerts would be raised, so emails would have been sent if they were configured, but not persisted since they would be unable to be entered in the data repository (i.e. they will not show in the UI). This is also the reason the log files had no entries in them. I would suggest setting up email alerts in case this were to occur again so that you would have that warning.
Kind regards,
Alex
Have you visited our Help Center?