dkeyser ✭✭✭✭✭

With errors thrown or even incredibly long delays to do 30 node threshold changes this is really the only way to do mass changes safely, quickly, and en mass.

Is Q3 still bugged?

I'm wondering if the incorrect answer message is just wrong. I'm looking again at the data sheet and technically Windows system logs AND Windows application logs are in the Changes category. They are just combined onto one line and have an AND in between them. If that's the case then the incorrect answer message should be…

Just to show what I was talking about above. Dunno.

I'd like to see some sort of answer file ability here also. It wouldn't have to necessaryily be silent but having the basics pushed out using answer files from either a network install or a package push would be amazing.

Just realized I'm an idiot that doesn't read the answers completely. My fault!

Thanks! I just ordered a notepad from the Thwack store to go with it.

I don't believe that this is what was being asked by the OP. I believe they want to have an audit event created when a user manually clears an alert from the all alerts view. The picture you have above shows when an alert note is changed or the alert is acknowledged but nothing noting that a specific user has cleared the…

Looks like someone came up with the needed OIDs even. Can we get this supported soon? These devices are popping up all over now. This may come in handy... i Use tables to gather these... then you may need to edit or check the RowID's you want (http://thwack.solarwinds.com/ideas/3430) 5K & 2K (5xxx and the FEX'es) info…

Either a multi select or missing the word NOT in the question it seems.

Just catching up and the question Wednesday seems to be taking the incorrect inputs for the answer? Selected all but windows system logs and it still says I selected that option.

That's an excellent work around there! But I do agree that a universal search function would be ideal. Thank you for the information!

I'd really like to see task state awareness in this. That way we can run multiple scheduled tasks without guessing how much time in between will be needed. Like running a task that installs only the non-exclusive updates and then right after running one that installs only the exclusive updates. The task engine could then…

I agree with the rest. I picked the wrong answer because it wasn't exactly the same in the hint but the correct answer was in there verbatim. Not a good question when all your answers appear completely wrong.

Another great option is to do a companion piece of software that looks for the module installs in the same directory and will run them all for you. That way you don't have to make so many packages and it'll unpack and run what is needed for all and then optimize as the last step.

The user was just stepping away from their console but they had a report open that was causing severe blocking in the DB. We wanted to kick him out of his session so that we could kill the SPID and not have it respawn the query. Being able to force a user session to close would have been great. I'll definitely keep this…

You'll probably need to use a custom SWQL variable in your email actions. Something like this: Down for ${N=SWQL;M=SELECT TOP 1000 HOURDIFF(TriggeredDateTime, GETUTCDATE()) as HoursDown FROM Orion.AlertActive AA WHERE AA.AlertObjects.RelatedNodeId = ${N=SwisEntity;M=NodeID} } hours.

I'll take a look at this. Though it appears this is for a Microsoft Cluster and my issue is with a SolarWinds HA cluster specifically.

This is for an HA instance. I've set it to look for SWIS being stopped now which should help but if there was a better way then I'm all eyes.

The original intent of this was to reach out to Linux servers and do some maintenance on them when an issue is detected. Now there's been some talk about Remedy ticket integration so I'm looking into API scripting and this will need to pass along credentials as well. Would prefer this to be handled through SolarWinds…

Just bumping this to see if there's any macros/variables I can use to securely pass the credentials I'm setting over to an external program. This would be great for quite a few things I'd like to do but not finding anything on this myself yet.

I spoke too soon it seems. I wasn't getting any errors but further investigation is showing the same error message as before when attempting to use -Certificate "Get-SwisData : The credentials supplied to the package were not recognized". Haven't put in one of the domain service accounts into Orion as a user yet but that…

I look forward to any additional info you can get on this topic!

I am currently using the -Certificate switch at this time. Seems to be working as intended when pointing at a secondary Orion server.

Didn't seem to help the issue and it turns out that I can run this on an APE from the Main without any issues. Another user confirmed they have the same issue. I'm wondering if this is by design or a bug maybe? Though it brings up questions on how this would work in single poller environments. Forgot to add that I'll open…

That's one way to tackle it for sure. I'll evaluate this against our access controls to see if it's a better path to take than issuing a prompt for credentials each time the script is run by someone manually. Though this is great for automatic remediation!

That did the trick! Though that raises questions on why we're not able to do this on the same server the script is running from. Single poller environments would benefit from this as well as less complexity having to make sure two pollers are always running for this kind of automation. Still a good start for me. Thanks for…

I did run the above query but I'm not sure what I should be looking for here. This is just showing all alerts and actions that have triggered in the past 24 hours right? I haven't set up any GET requests just yet so no error messages or anything like that exist. The scripting idea would be interesting. I know the engine…

Thank you! My thoughts are that the engine executes the trigger query and does a diff of the results against the list of active alerts with the same definition ID. That way they wouldn't need to do any interpretation on the incoming logic in a custom query just reset what doesn't exist in the trigger results. Just not sure…

This brings up another interesting thought that is implemented in the report writer but not available here. The ability to mark a column as hidden in the output. That ugly looking column would be a great candidate for an indicator to leave it's output out of the final table. Maybe another [_SortFor_ColumnName] so when you…