Alert Issue (AlertCaption)

Here are the Fix Scripts I wrote and applied to correct this issue:

These scripts are provided wholly without warranty, utility, or verification beyond my own environment.

By executing these scripts you take full responsibility for understanding what they do and any outcome

Please contact Solarwinds Support before applying these and open a case before applying

Fix the related node captions

UPDATE AO
 set AO.RelatedNodeCaption = N.Caption 
FROM [dbo].[AlertObjects] AO inner join nodes N on
AO.relatednodeid=N.Nodeid
and N.unmanaged=0
 where  AO.RelatedNodeCaption<>N.Caption

Fix the alert captions on Node Alerts

Update AO
 Set AO.EntityCaption=N.Caption
FROM [dbo].[AlertObjects] AO inner join nodes N on
     AO.relatednodeid=N.Nodeid
     and AO.EntityType='Orion.Nodes'
     and N.unmanaged=0
WHERE AO.EntityCaption<>N.Caption 

Fix the Alert captions on interface alerts

Update AO
SET AO.EntityCaption= I.Caption
FROM [dbo].[AlertObjects] AO inner join nodes N
     on AO.relatednodeid=N.Nodeid
     and N.Unmanaged=0
Inner join interfaces I
     ON I.Nodeid=AO.relatedNodeID
     and AO.EntityType='Orion.NPM.Interfaces'
     and i.unmanaged=0
where  AO.EntityCaption<>I.Caption
and  'I:'+cast(I.InterfaceID as varchar) = EntityNetObjectID

Nice Catch!  On a different yet related note, I actually switched from using the Caption to using Sysname where possible because of conflicts with CMDB discovery data.  Clients tend to use caption in wildly different ways and can sometimes be nothing like the actual name of the device (sysname).  So things like CI binding (we use ServiceNOW CMDB and ITSM) don't work if you pass caption because ServiceNOW has no clue what device you're trying to bind an incident to!

Note: this bug is still there in NPM 12.1 and 12.2

in our upgrade the server names were changed, and staff started troubleshooting alerts on inactive SolarWinds servers.

Also need to fix the engine names

Update AO 
    Set AO.EntityCaption=E.ServerName 
    FROM [dbo].[AlertObjects] AO , Engines E
    WHERE AO.EntityCaption<>e.ServerName
        and  AO.EntityType='Orion.Engines' 
        and  ':'+cast(E.EngineID  as varchar) = EntityNetObjectID  ;

Bug is still there in 12.4 as well.

it had always bugged me that i changed the names and they weren't reflected in all areas, i didn't realise how bad it was until l ran your select query and found over 2000 node entries ( i took your interface update and changed it to a select and have 30 entries).

I'm torn between running your provided select statements and leaving as is just in case it causes further damage.

Engine names: Be careful with this one if you have HA running as the Engine names change dependant on active poller.

Thanks for the information and the fixes though.

What about alerts for 'RESET: Average Response time of blah blah blah...  All other alerts show dns this alert shows server name, not my poller (dns)??

Bug is still present on 2020.2.6 HF2

The query to get application component with a different name in alerting : 

select AO.EntityCaption as old_name,APMC.Name as actual_name, AO.EntityNetObjectID, AO.AlertObjectID
FROM [dbo].[AlertObjects] AO
inner join [dbo].[APM_Component] APMC on AO.EntityNetObjectId = CONCAT('AM:',APMC.ID)

where AO.EntityCaption<>APMC.Name
and APMC.isdisabled is NULL

The query to update alert component name :

UPDATE AO
SET AO.EntityCaption = APMC.Name
FROM [dbo].[AlertObjects] AO
inner join [dbo].[APM_Component] APMC on AO.EntityNetObjectId = CONCAT('AM:',APMC.ID)

where AO.EntityCaption<>APMC.Name
and APMC.isdisabled is NULL