Comments
-
It's on our roadmap to persist/display the SQL handle which would allow the bridge back to SQL Server or from SQL Server to DPA in this case. However, that hasn't been implemented yet. I respectfully disagree about the "uselessness" of this alert though. I'd assume you are running this to notify you that a SPID is running…
-
That we can't do. The query you are running for the alert as defined runs directly against the monitored instance. The query_plan_hash being selected is directly from the sys.dm_exec_requests DMV and DPA isn't doing it's own hash as it never actually flows through our app (we only show what SQL shows us). Sorry about that!
-
1. The alerts and the calc for the long running list are not kept in sync. I believe the long running average list is updated every 10 minutes (tied to our stats poll where we collect the number of executions). This would make sense as we can't do an average execution time until we know how many times it got executed. If…
-
If you click on the link below, you can submit and it can get voted up. Our product managers monitor these requests. Database Performance Analyzer Feature Requests
-
If you are still running as an alert, unfortunately, we are limited to 100 characters currently for the output. That's part of our data model and probably not easily addressed. Have you submitted this as a feature request here on Thwack?
-
khathaway - can I have you open a support ticket for this? Not aware of any systemic issue in DPA regarding current view. They are likely going to want logs and share the real-time query we run so you can run it outside of DPA to see what might be happening...
-
Ray, couple of thoughts on this: * DPA persists only certain data about session activity currently (select metadata - basically the dimensional tabs in the UI) * You can watch for current sessions and keep refreshing them, but we're only as good as what information is sent from the client application - some send more…
-
Steve, looks like you may be in a position where you'll have to uninstall and reinstall to get the license back. Can you see if that does the trick?
-
Antonio, not sure that will buy you much as we are already summarizing data, so deleting older summarized data won't get back much storage. Can you run this script against your repo DB and see what objects are consuming the most space? -- temporary table to hold results of sp_spacedused for a specific table IF…
-
We have a custom report that you can run against the repo to retrieve the data manually (using SSMS or another client). Custom Report - Top SQL for Database Let me know if that's what you had in mind.
-
As you found, for monitoring, you can remove the sysadmin role, but have to replace them with quite a few. Check it out here: Success Center Keep in mind that *some* functionality may be missing should the monitoring user not have sysadmin. For the repo, sysadmin is used to create the new database and schema owner. After…
-
Set your timeframe to be within an hour or within a day. If on a day, set the interval to be 1 day. Then click on the waits tab in the dimensions listed across the top. Click on the 1 vertical bar. Click into the wait you want to see patterned. Now click back to timeslice. What you will see are the top SQL statements that…
-
In the UI, that specific functionality is limited. What DB are you using as the repo? SQL or Oracle?
-
If you've got the SQL that can pull the numeric, this could easily be built into DPA as a custom metric that can then be alerted on. Got to options -> custom metric and walk through the template. It's pretty self-explanatory, but if you still need assistance, you might create a ticket with support.
-
Unfortunately, at an aggregated level, there is not a report like what you are asking about. You could run a query against the repo to extract that info fairly easily (by hitting our summary tables), but it would not be something within DPA itself. For more info about how to query our repo directly, I'd recommend…
-
What database platform are you monitoring? Also, what is your DPA repo on (Oracle or SQL Server)?
-
Average doesn't adequately address peak needs. Steps I'd take: When in the instance in question, click on the resources view, then CPU tab. Set your timeframe to last 24 hours or 1 week (keep in mind though that the more you zoom out in time, the more averaging will knock off peaks and valleys). Note a couple timeframes…
-
Likely looking at a feature request for this. I know we have our retention settings for data (detail, hourly, daily for wait trends, metrics, etc.). However, currently we don't have a way to tell DPA to get more aggressive in it's retention as it nears a specific storage usage...
-
No report at this time, but you should be able to email the visual you want to show them from DPA. Let me know if that doesn't suffice.
-
Megan, I was reviewing past posts and saw this wasn't answered - did you ever get this resolved?
-
I believe I've used this before. This looks at max bytes for the tablespace and calculates a percent used. You could change the select to just return the bytes used or customize for your purposes. Let us know how it goes! select df.tablespace_name, round(((MAX_BYTES-((MAX_BYTES-BYTES)+FREE_SPACE)) / MAX_BYTES )*100)…
-
If you are on DPA 9.0, the answer is yes. The monitoring user still exists, but no active connections will be established and no queries against that instance if not licensed.
-
Was the issue on the repository Oracle database? Was monitoring running? Were there any blackouts? Did you get any emails during that time? The short answer is no, that is not the normal behavior.
-
If it is still occurring, can you navigate to the current view, active sessions link, and click on the blocking tab? What does that show?
-
Unfortunately no (good feature request). However, never one to settle, I set up a custom metric in my system which did get applied to all Oracle instances. Then I went into the instance(s) that I did NOT want to collect that metric for clicked on settings -> custom -> uncheck the "enable" flag. Not elegant as it will still…
-
When we pull the SQL text from dm_exec_sql_text, we notate where we note where in the procedure we got the actual SQL statement from (character range which represents the actual SQL statement. Since a procedure may be made of a LOT of SQL statements, we let you know if you want to go back to the raw data where to find the…
-
If you are running the latest version, there was an SMTP issue introduced. Here is a fix IF you are running DPA 10.2.573... * Navigate to <install_dir>\iwc\tomcat\webapps\iwc\WEB-INF\classes\spring\iwc-service.xml * Line 355 should contain following"<ref bean="gmailMailServer"/>"This line needs to be deleted. * Restart…
-
Unfortunately, you will have to download the free product again which will automagically create another free key...
-
Yes, that's normal. SQL Server will grow into the max server size you set for the instance and not release the memory (even if not in active use). Remember when setting the max server size to leave some room for the OS and anything else that may be/need to run on that host. If we're talking about a VM, then that brings in…
-
In the admin guide, there is a list of types of alerts: SolarWinds Online Help Here's the section on report types: SolarWinds Online Help Other than that, you can navigate to alerts or reports (upper right) and scan through the options. Not aware of a comprehensive list anywhere that would detail them if that's what your…
