gcp · Database Administrator · ✭✭✭✭✭

Thanks @jaminsql that is just what I needed to hear

I'd also be interested in any information about the repository tables. There's an option to execute custom alerts against the repository (and it also can be used in the Database Query Tool) which might be useful, but without knowing the repository configuration, it's a bit hit and miss. A simple one I'd like to know right…

Thanks roballen much appreciated. I'll try that out as a workaround, that solution may be ok for a limited set of queries where you can use snapshots. However, as a DPA user, why would I think a custom alert would operate any differently to a normal alert? My expectation is that I provide the custom sql, and that DPA runs…

Thanks for the suggestion. I have tried a multiple numeric alert. The problem is that while I can get the dbname I only get 1 result per instance. So DPA is connecting to a database (possibly every database) but only returning 1 result. e.g. On a server with 6 databases in the instance (let's call them MYDB1 - MYDB6) I…

thomas.nemet wrote: Can´t imagine that i`m the only person that wants to monitor a Sybase DB with DPA. Isn´t there any property i could set up in "Regist instance wizard" to enable password encryption? You really need to lodge a support ticket to get those answers. They don't bite. Much.

I suggest you also put in a Feature Request to have the functionality added to DPA. https://thwack.solarwinds.com/t5/DPA-Feature-Requests/idb-p/dpa-feature-requests If enough people vote on it, it may, eventually, possibly even in your lifetime , be added to the product.

I would like to add Australia to the list of countries that can actually use the THWACK store It looks like you live in a country that we do not currently ship to.We can ship to the following countries:Canada, Germany, United Kingdom, United States. 

When you go to a forum page e.g. https://thwack.solarwinds.com/t5/DPA-Feature-Requests/idb-p/dpa-feature-requests there is an Action drop down menu on the Right Hand side with a subscribe option Hope that helps!

2013 or 2014?

If you have the SQL hash and a sufficiently long history, what about View Historical Charts for SQL from the trend window? It will show you the number of executions per day for the duration of the history. Very easy to see any unusual peaks.

Assuming you have already configured and tested the Mail Server settings (you have, right?), then check the DPA logs (Support -> Web Client Log File I think) for error messages. Also check your Junk mail and Spam folders just in case?

* Do you continue to visit the solarwinds.com website? How often? What kind(s) of content do you normally go looking for on solarwinds.com? Not very often. * How often do you log into the customer portal? Other than product upgrades, what other content do you visit the customer portal for? Maybe monthly or a bit less.…

If you edit the alert (add a comment or whatever) and save it again, it resets the run time to restart from the save time. There is some couple of minutes of lag time between the edit, and the repository tables being updated so it is not precise. If you're feeling brave, have a look at the repository tables con_alert (id,…

Are there any error messages in the SQL Server error logs? Is SQL Server enabled for TCP connections? Is the IP address/Port correct? Is there a firewall rule blocking the connection?

It appears to be something to do with Optimizer Statistics on Linked Servers - this link may help a bit: Optimizer Statistics on Linked Servers | Benjamin Nevarez

You can set up a custom alert (Custom SQL Alert - Multiple Numeric Return) with something like this: SELECT T1.NAME AS DatabaseName, CAST(COALESCE(DATEDIFF(second, MAX(T2.backup_finish_date) , CURRENT_TIMESTAMP)/ 3600.0, 9999.0) as DECIMAL (5,2)) AS [Hours Since Backup] FROM master.sys.databases T1 LEFT OUTER JOIN…

I don't think SolarWinds publish the Data model, but if you ask them nicely they can usually help you with specific questions.

Full Table Scan - as the DPA Advice says, probably first thing to look at is whether there are any suitable indexes? If so, are the table and index statistics up to date? If not, can you add a new index or modify an existing index? Can the SQL be re-written to be more efficient? Have you checked the execution plan? It…

The only time I've seen DPA monitoring show up is in a Dev/Test environment where little or no other activity was happening. In that case the % allocated to DPA was quite high, but the actual wait time etc was quite low, so it was a false positive. If you are seeing a genuine impact, I'd raise a support ticket ASAP.

The query has an unqualified table name, so the Live plan will try to use your DPA user as the default schema name for the table i.e. <DPA_user>.OA_CUSTOMER_DTL which (apparently) does not exist. I don't have Oracle, but when you do a Live Plan for other DBMS you may get prompted for a schema name - use the appropriate one…

As far as I can see (DPA 9.1) the notification is per alert, not per group. (Waits for someone from SolarWinds to point out the overlooked option...) I can see what you want to do, and I think it would be a great Feature Request if it's not currently supported - specify Contacts/Contact Groups for Alert Groups not just at…

Unfortunately DPA does not let you create your own custom reports. The Data Model is not published as far as I know, but the SW support are usually pretty good with helping you to write queries against the repository. You may want to open a support ticket to see if they can assist with the SQL query.

The way we handle at my work this is to include a special string in the alert name (e.g.$SMS$, #TEXT# etc.) and have an email rule set up to forward that via SMS to the relevant area. You will need access to some form of SMS gateway or product.

thesh wrote: Evaluation Frequency of Alert 1 Minute If that is what you defined, I'd guess that is why it runs every minute. Should that be 15 minutes?

A couple of ideas, but please note these are not tested. * Set up an alert on wait time for a single SQL (maybe Average Wait Time for a Single SQL) - but set the High threshold as a low range, rather than a high range. But, being an average, this may not be a timely alert as it will potentially take time for the average to…

If you select a time period (e.g. 8-9pm) for the day/database in question and select the 'DB Users' tab you can see who was active in that time period. if you then click on the Userids listed and it will show you what they were running during that window. You might be able to find them that way, although it is a bit clunky.

I believe the free space is based on the allocated file size, not the amount of free disk space. If you have unlimited growth turned on, then the alert is useless - it always returns 100%. I raised this particular issue (for DB2, but works the same on SQL - I've tried) and got this info: Explanation for DB Freespace Alert:…

Do you mean that the actual SQL Server windows Service is being restarted, not just the DPA monitoring itself? I don't think DPA has that functionality (unless it is new in 9.2). It will attempt to restart monitoring, but as far as I know, there's no option to restart the database services themselves. Are you sure you…

If the alert is set to run every 15 minutes, it will. The SQL construct is not considered as part of the alert schedule, it is just run when the alert fires. If you don't want it to know about it outside of office hours, you will need to set an alert blackout.

skenny wrote: DBI DBD::mysql DBD::Sponge DBD::Proxy DBD::Pg DBD::Gofer DBD::File DBD::DBM I wouldn't think that a MySQL driver would work terribly well with a DB2 database?