m60freeman · Senior Database Administrator · ✭✭✭✭✭

It works if I add "ignite." before each table reference. You need to add this to the inner subquery's WHERE clause: AND a.advice LIKE ''Missing Index:%'' In my database, I also have instances where advice is "Unmatched Index: <IndexName>" because of the use of filtered indexes.

That didn't seem to work, but I put in a hack that does. I added "AND CONVERT(TIME, GETDATE()) > CONVERT(TIME, '00:10')" to the query to suppress the alert for this situation, which always occurs between midnight and 12:10am. I'd prefer something more targeted, but this will get me by until I can install the SP2 for SQL…

Thanks for the quick response! I'm on SQL Server. I can add "AND sw.izho <> 6041431186" to the WHERE clause, for example, but if the hash changes due to a restart or memory flush, etc., this won't help. Also, the hash will be different on each server. Is there a join I can add that will get me the procedure name? I don't…

Every night, SSIS runs a procedure called SSISDB.internal.cleanup_server_retention_window, which blocks executions of SSISDB.internal.append_event_message for several minutes. I would like to add a WHERE clause to this custom alert query to filter out these nightly events. Which table and column name would I need to check…

Looking forward to them!

Great idea! Unfortunately, that search comes up with 240 results on one of my main instances. If there was a way to select jobs instead of job steps, that would help. I have one job with 20 steps, two with 17 each, eleven more with between 2 and 10 steps, and forty-five single-step jobs on that instance. And it isn't…

Great idea! (Sent from my phone)

In the "Improving management of instances in DPA for large environments" category, here is a shameless plug for people to vote for:

The notification text says "The following Ignite monitors are currently down and not collecting data." That has nothing to do with blocking. It was obviously cut and pasted from a different alert. Perhaps something like "These are sessions being blocked:" would be more appropriate. Ideally, the query text would closely…

The notification text doesn't match the alert.

For me, DB_clue is populated for all stored procedures, but not for ad hoc statements. Just FYI.

The example shows a fourth parameter ( 'sqlplus.exe'), but the proc only has three. Mark Freeman Database Administrator | Rogue Fitness 1080 Steelwood Road, Columbus, OH 43212 MFreeman@RogueFitness.com On Wed, Jan 7, 2015 at 6:15 AM, kathleen.gibbs <

I had a similar issue that I had attributed to a version upgrade of DPA. The root cause was that the servers with the high Signal Wait % were very lightly loaded and so although the Percentage was high, the actual time spent in Signal Waits was very low, making the high Percentage irrelevant. Adding a chart for the Signal…

I think that this is a good idea, but I think it needs refinement: 1. Without the query text, there is no way to know which query for a SPID is running long. I can see why trying to include the query text would be a problem, but you could include the handle (r.sql_handle) instead. 2. What is long for one query is not long…

When we set up transaction log replication, jobs and alerts were automatically created that handled this, assuming you have Database Mail configured.

I had no issues with the RC, with a SQL Server repository and only SQL Server and MySQL instances being monitored. Mark Freeman Database Administrator | Rogue Fitness 1080 Steelwood Road, Columbus, OH 43212 MFreeman@RogueFitness.com On Wed, Mar 15, 2017 at 11:57 AM, rharland2012 <communityteam@solarwinds.com

There are feature requests to deal with this as there is no current product feature that will do what you are asking. Please vote for: There may be others that are related. Note that I'm just a user and do not speak for SolarWinds.

Please vote for this feature request (if you haven't already): It's currently the #4 ranked request!

You may want to vote for these related feature requests:

I agree completely. I already have a couple of related feature requests submitted that you can vote on that are small steps toward a more comprehensive way to address the idle blockers mysteries:

You may want to vote for one or more of these feature suggestions: A workaround is to select a Custom Interval on the Trends page, then select the Timeslice tab, then view the Resources section below the main Trends chart. You can change which Resources charts appear there, and in what order, although you are limited in…

Not to derail the conversation, but being able to configure the color used for a hash would be handy. How is that done? The automatic assignment of colors leaves a lot to be desired. See for an example.

You may want to vote for this feature suggestion:

We're running on a VM with no trouble at all. We're only monitoring 7 instances, but there are over a dozen other apps that use the same SQL Server VM for their data store. Mark Freeman Database Administrator | Rogue Fitness 1080 Steelwood Road, Columbus, OH 43212 MFreeman@RogueFitness.com On Tue, Jan 19, 2016 at 11:21 AM,…

The idea of "discoverability" in UI design is great, but still doesn't replace the benefit of a manual. It's not unusual for someone to point out a feature in DPA that I never noticed or knew about before. It would be great to be able to go through a manual and find all of the features.

DPA has really good features in this area, but you have to be on 10.2 to get them. I think you also need to be monitoring a SQL 2012 or higher instance.

You may be interested in voting for these feature requests:

I'd agree with that, specifically when it comes to the colors (and the order in which they are used) for the chart bars. In the example below, it would be much more readable if colors 2 and 3 (counting from the top in the legend) were swapped, 8 and 9 were swapped, and 11 and 13 were swapped.