mandevil · Product Manager · ✭✭✭✭✭

I'm going to recommend a different route. Instead of a custom report, try to navigate in the UI like this and see if it gives you what you need. From the home page, click on the instance you are interested in. Click on a day you want to research. Select the interval to be 1 day (defaults to 1 hour). That should present you…

This won't actually add the report to the drop down options under reports. This will only add to to the Database Query Tool under options -> Support tab... There's no way to add a report to the drop down under reports as that's referenced in source code. Sorry about that. My recommendation is to submit a feature request to…

Things may have changed, but I supported Postgres until about 4 years ago and saying it had robust internal monitoring capabilities is a real stretch. Most of the monitoring was tied very closely to the OS. Other than watching OS processes and resource consumption, there was a build in lock procedure you could call.…

SWI has a strict policy to not disclose release dates or even estimate them. Sorry about that!!

You might have to fully qualify that object. Who is the schema owner of that object in your environment? For example, you might have to specify ignite.con_ag_database if you aren't querying as the schema owner...

This has been on our radar for a while now - to give attention to alerts. It is certainly on our roadmap, but I can't give specifics on if/when it will be done. If/when we do get to it though, would you be interested in taking part in our beta program to review the changes and provide feedback?

Do you have the option set to "and above"?

Yep. 8 )

Also keep in mind you may want to set the deadlock trace flags so that the details of the deadlock get written to the error log.

What you are looking at with the Deadlock metric here is specific to another tool within the SolarWinds suite called DPA (Database Performance Analyzer). It's specific to deadlocks within a database instance. I'd recommend you re-post your question under the NPM tool forum as I think it's more specific to that product. Let…

You created this as a document instead of a disd cussion, so I'll just use the comments instead of a reply... You should be able to go into reports and create the one you want and then schedule it to run nightly. Have you played around with the reports at all in DPA? That's the best place to start... Using reports is…

If that works, great. This alert was intended to be used with the DPA application and we *should* be converting the value used for interval for the #frequency# (it gets parsed by the java code and should be in the same increment as how the interval is defined for the alert). Let me know if that's not quite clear...

Hate to do this, but can you repost this under the SAM forum? That community will be able to help you much better than we will. 8 )

I don't. I'd ask the 3rd party provider of the backup tool if they log anything within SQL Server itself. If they update any tables within SQL Server, then we might be able to look at them, but I'd have no idea how their tool is architected.

This is alert is for monitored instances. The results are persisted in our alert tables. Is there a way to query SQL Server for 3rd party backups? If yes, then a 3rd party alert is possible. If not, we're confined to what we can pull from SQL Server directly as we don't have OS access (intentionally).

Were you logged in as the schema owner? Otherwise, you may have to fully qualify object names...

I think your second value returned is not numeric. You will have to pull back a numeric for the 2nd field. The hash should work for the 2nd field as it is a numeric - try reversing the order of return in the select...

You can only select two values (the first can be alpha-numeric, but the second has to be numeric. If you post your query, I can take a look at it.

Did you try using the multi-numeric custom alert template? I believe it allows you to return one non-numeric + one numeric value. Instead of returning the sqlhash, you could join to const table to get the text. You will have to use the substr function or equivalent due to size limitations for alert messages. To get full…

We collect SQL text and store that in our repo, so yes, you can edit this to do a join to the const_#DBID# table and pulling the ST column in the select clause. I haven't played around with it, but you'll want to join con_stats_sum_#DBID#.sqlhash to const_#DBID#.h

I found this one out there: Custom Alert - Execution Plan Changed for SQL

Make sure you are logged in as the schema owner, or fully qualify the object (for example: ignite.cond).

It means that you have 99% of your extents used. Here is a discussion about tempdb: https://support.microsoft.com/en-gb/help/307487/how-to-shrink-the-tempdb-database-in-sql-server Keep in mind that when tempdb allocates an extent, even when done with it, it will not automatically shrink tempdb. However, any currently…

At this point, I'm going to ask that you submit a support case. They can do a screen share with you and hopefully get this worked out...

OK, so it is not liking that to_number function. Can you run a test against the repo directly using a simple query with a to_number function? Something simple like select to_number('10') from dual; Run this from Options -> Support (tab) -> Database Query Tool and make sure if gives you a valid return.

Your repository is Oracle I take it?

For sure. See constraints in my previous post though about outputs.

Go here: https://thwack.solarwinds.com/community/database-management/database-performance-analyzer and scroll down. On the left hand side you will see a link to custom SQL content including reports, alerts and even metrics. You can use them as is and plug them into DPA or you can tweak them per your own requirements.…

Very possible. Whatever TSQL you can come up with will work as long as the output is acceptable. For the single numeric, that one's self-explanatory. For the multi numeric, that one can have one alpha/numeric and one numeric value (the numeric always second). In fact, we've had lots of folks grab some of our custom content…

SPIDs will get reused. Remember, we're displaying what SQL Server is telling us. A SPID could be used by app A between 2:01 and 2:03, then release it and then that same SPID will be used by app B between 2:03:01 and 2:03:05. If applications are using connection pooling (best practice), the app hopping should not be…