mandevil · Product Manager · ✭✭✭✭✭

There is actually an option that you have to tweak to get to the 120 hours allowed. options -> administration tab -> Advanced Options On the system options tab, check the support options checkbox and look for CUSTOM_TIME_MAX_RANGE_IN_HOURS. I don't think it says, but if my memory serves, I do think it's 120 hours max as…

I'd need to know a little more about your monitoring setup. If you are monitoring using the AG listener, then we'll follow the primary replica. Regardless of which instance it currently lives on. So we'll only "see" activity on the primary, nothing from the secondary. Might be worth opening a case with support. Let them…

My thought exactly. You can't set it up initially using RDS as we do a check if creating the repository that it will trip over. But if the repository already exists and we edit the repo.properties file, don't see why the app would care. If you do try it, let me know how it goes. Again, bear in mind that this would not be a…

Looks like it's time for a GTM. Submit a ticket with support and we'll hook you up.

Try deconstructing it a bit. Start with a simple SQL query which will return just a numeric value. Can you let me know what it is you are trying to achieve/alert on?

That's correct. 9.2 will ship still using the old jTDS driver. However, the jTDS with the fixes mentioned will be included in the bits, so a simple rename will be required and a restart of DPA. There will be a KB article going live with 9.2 with details on how to accomplish this.

The change will only be for future data, that part is correct. To see data prior to 30 days, you can leverage the reports function within the tool. Keep in mind though that historic data older than the detail will be summary data as implied by it not being detail. That means no drill in functionality, just static charts.

That definitely helps - thanks for the reply and thoughts!

When sessions go inactive, they will be excluded from our poll for that second. If they become active again, we'll "see" them again. Most apps use connection pooling (or should). Let's say the app spins up 20 connections in a pool. Of those 20 connections, at any given moment, it may be using say 9 of them for SQL requests…

Just make sure you understand what changed to cause the red (run through the analysis already done by DPA) - then you will look way smart! 8 )

What version are you currently running? I'd like to implement this on DPA 9.2 as we can then go against the summary table which will now hold 10 minute summary data points. Going against detail would be possible, but much more difficult and a lot more overhead. Thoughts?

Feedback I got was if you want to include the drive letter, it has to be part of the name as configured.

For that, I'd recommend opening up a support case. I know we've shared information about the ERD previously...

Another cause would be a cleaners that run nightly removing rows once summarized and exceed the retention value as defined in the CLEAN_DAYS_OF_DETAIL value in options. You can manage the space, but I'd recommend following vendor best practices. Something like this Oracle doc can help with doing online reorgs and then…

OK, yep, some common reasons for lack of stats are: * Rollback (for transactions) * Statement didn't complete within the window or timeframe being viewed * Statistics got reset in the DMVs (recompilation) There are certainly other causes as well. If you do want to dig deeper on this, I'd suggest opening a support ticket…

If we are talking about sharing information about the ERD, can you submit a support case? We can share that portion of the model so that you can see how we're joining the objects in our schema.

Is this host dedicated to SQL Server? I agree though, there *should* be some OS overhead, even if minimal. Do you mind opening a support ticket? We can share our calculations via a support case which may help determine why they are the same (we can go for the raw data directly from SQL Server).

OK, yep, you can do that. Let me know if you don't want to see the repo activity as there's a way to add to the our collection query to exclude things.

What rdbms are you monitoring? SQL, Oracle, ??

If you have an enterprise alerting/notification solution that can consume SNMP traps, that may be the best option for this. DPA can send alerts as an SNMP trap which can be received by the enterprise solution and then dealt with appropriately. I understand that this would be a workaround and not native to DPA, so I only…

I believe in the past we'd tried that solution, but found out that there were other dependencies and it was not as easy as updating that table... Something is getting written on the java side in a config file. Thinking editing the script is the safer way to go.

Let me verify if WRITELOG is included in i/o waits. I think it should be, but want to make sure. This may be easier to discuss on a phone call... Can you open a support case and mention you have been working with me on this? We can coordinate a time next week to hop on a screenshare and investigate more. (I'm currently OOO…

Same questions - can you supply other metadata about this statement (where it's coming from, who's calling it, what program, etc.)?

First, I'd recommend an upgrade. 8 ) B, see if this helps: SolarWinds Knowledge Base :: Increase DPA Memory How many instances are you monitoring, BTW?

Thoughts on the upgrade? New features (may take a couple days for the TTA to kick into gear as the analysis only runs at midnight each 24 hours.

Does anyone know what features the %_cur% tables support? Was thinking this may be a part of real-time polling in PerfStack, but not sure. Have a situation where a number of tables are growing large that are memory optimized causing less and less memory for normal buffer cache. Especially seeing ResponseTime_CS_cur getting…

Any chance you can upgrade your repo version or move to another supported SQL Server version? In DPA 12.0 has a TON of goodness that I don't want you to miss out on.

We'd also like to handle those nicely asked questions via a support ticket as we may get into material that is better not posted to publicly consumable content. 8 )

Recommendation is to use your company email server. Can you log an internal ticket in your company or contact your exchange folks? They should be able to help you with the correct entries to use company email. Using embedded means we'll try to leverage the mail executables on the server where DBA is installed. There may be…

Looks like a permission issue with the credentials provided. Can you check with your SMTP admins on the correct access and permissions? The fact that it's returning permission error is actually a good sign - it means you're getting to the SMTP server.