Global Dashboard Stuck on "Loading..."

C_Davis

We've had a very stable install for a couple months now but very recently my global dashboard page was rendered empty and refuses to load. The rest of the application is functioning normally, I can manually navigate to sub-pages via my history and the server responds fast and normal. The only page impacted seems to be the global dashboard. I've reorganized any major (>80%) index fragmentation, and no queries are blocking, locking or waiting for anything of note, there doesn't seem to be any runaway processes. It started several days ago and since the first occurrence has been re-occurring (no global dashboard since).

C_Davis

Oh, one thing of note, when it first occurred we were on a 6.x version. I've upgraded to 7.0.7.7367 in an attempt to clear the issue

Alex B

Hi C_Davis,

There was an issue in version 6.0.15 (and possible one or two point releases before) that caused this issue and it was fixed in 6.0.16 and later, but if the issue had occurred then a script would need to be run to correct the issue.

It was also seen when a customer had a single node of a cluster (when this was allowed in v4) and then added the cluster later on - removing the single node resolved the issue here.

This is the script to clean up the errant entries in the data repository PLEASE BACKUP YOUR DATA REPOSITORY before running this and make sure you restart the SQL Monitor Base Monitor service after running it (you may also need to refresh and/or restart the web service or IIS page too):

DELETE M FROM [settings].Machines M 
JOIN [settings].Clusters C 
ON C.Id = M.ParentId 
WHERE C.IsCluster = 1 AND LEN(REPLACE(M.NAME, CHAR(0), '')) = 0 

DELETE MK 
FROM [data].[Cluster_Machine_Keys] MK 
INNER JOIN [data].[Cluster_Keys] CK ON MK.ParentId = CK.Id 
INNER JOIN [settings].[Clusters] CS ON CS.Name = CK._Name COLLATE SQL_Latin1_General_CP1_CI_AS 
WHERE MK.ParentId = CK.Id AND CS.IsCluster = 1 AND LEN(REPLACE(MK._Name, CHAR(0),'')) = 0

Please try this and let me know if it corrects the issue for you!

Kind regards,
Alex

C_Davis

Excellent! I will give this a run first thing in the morning and report back with the results!

C_Davis

No luck. It cleaned out one row in the second query and I restarted both sql monitor/web service. I did however recently add an AlwaysOn availability group, this would roughly coincide with the monitoring issues.

C_Davis

Update! I deleted all traces of the AlwaysOn availability group on those tables in the query and my overview came back. I'll need to do some more testing to try to get this working - our AlwaysOn group is in a production test at the moment and will be moved to a live system within the month, which I'd prefer to have monitored in Redgate