Comments
-
What version of Sybase are you monitoring? If you are running 15.7, we are aware of an inconsistency in our calc. We do a regex in java with the results, but pull the raw data from Sybase itself (monEngine table). If within DPA, you drill into a specific day trend, use the interval dropdown and set it to 1 day. Then click…
-
First of all, welcome! A few things jump out at me about this one: * This transaction is waiting on locks to be released (could be causing them also, stepping on its own toes). * It's also waiting a long time writing to the transaction log (look for high write latency for the tran log file under the storage i/o view). *…
-
Maybe go back in and double check that the schedule is working correctly and the contacts assigned are the same as they were. If that doesn't work, then this may be a good candidate to engage support on. If that's the case, you can submit a ticket within customer portal or from this link: Submit a Customer Service or…
-
Interesting - the upgrade, especially to SAM 6.3 should have upgraded the DPA integration module. If you have access to your portal, I'd recommend downloading/installing the latest version of the DPA Orion integration module. If you have any questions, reach out to your account rep and they'll line up a resource to help…
-
This information is sent by the client application to SQL Server. I'd recommend doing an nslookup on that host to see what IP address it references (assuming it is in DNS). Also, do you have anyone that works for you named Kim Grant? 8 ) This would be the same info you'd see if you ran this from SSMS: "select hostname from…
-
We only collect this data per instance currently. I think this would make a good feature request though.
-
Yes, but be careful. See this post from earlier. There are a couple ways to "clean up" DPA trending charts... Re: SQL Server Monitoring
-
Found this script on Stackexchange that might do what you need. You can modify it for your purposes - like do a sum(bs.backup_size) in the select and then use that numeric to compare against some size for when you would like to get notified. When you get something working per how you want to use it, do you mind posting…
-
FYI: DPA now supports monitoring of Azure SQL Database and utilizing it for a repository with DPA 11.
-
I've not actually created an instance of DPA in azure, but no reason you can't as long as your are spinning up a virtual machine. I don't think you can do much with the SQL database offering YET with DPA (but stay tuned). The tricky part of hosting DPA in the cloud and monitoring on premise instances would be network…
-
The best way to answer this is to say that we send things out using SMTP and do not support SMS at this point from DPA. If your carrier can handle the SMTP traffic and convert to SMS, this may be feasible, but testing locally, I found that I could not get it to work. I have t-mobile though and there may be some…
-
Good question. I don't see that option. You can define a rolling set of days (last N), current week (and exclude say Sunday and Saturday if you want), but I don't see a way to have it displayed say Monday to Sunday... I think this would fall into feature request territory.
-
If you are on the latest versions of both Orion and DPA, no worries - they *can* run on the same host. Just keep in mind resource contention. DPA and Orion can also absolutely reside on separate hosts, not problem. Either way, the communication between the two will remain the same.
-
What version of DPA are you currently running? In I believe 9.1 (latest GA version), you can click on alerts and you will see alert blackouts tab. Use that to configure when you DON'T want alerts to run (like during business hours).
-
If the database is the only thing running on that system AND you weren't doing some kind of system backup on the 18th or don't have another cause for the CPU spike, here's what I'd recommend. In DPA, click on the 18th. Set the Interval to be 1 day. Click on the one vertical bar. Select the Waits tab. Click on the…
-
Can you post the specific values you use? For resource alerts, we take the thresholds from the resource metric definition, then apply some logic to those (you don't actually specify a threshold differently within the resource alert). What value is PLE at that you are expecting an alert on? Is it below 600 or below 300?
-
No current way that I know of to do it within the UI besides doing it in the trend view (as you mention). There is a request to include resources in our reporting - would that suffice? If yes, vote up this request: If you are looking for where the data is in the repo, you will likely have to open a support case for that.
-
Currently, we can only send out messages as SMTP. SMS is not supported. However, some carriers can convert the message if sent to something like <number>@att.com. You'd have to check with your carrier to make sure they support this.
-
Here is the calc DPA is using: select cntr_value from sys.dm_os_performance_counters where lower(counter_name) = 'page life expectancy' and object_name like '%Buffer Manager%' As you can see, we're relying on SQL Server for this counter. My question would be are they calculating it correctly? I have not looked into how MS…
-
When you see memory/cpu, that's actually an absence of waits. It means that the query is executing and likely the most significant activity is doing logical reads (cpu fetching data from cache). The reason we track it is that from an end user perspective, wall clock time is still passing, so even though that's an efficient…
-
You are exactly right. Basically the idea is that any data collected while validly licensed is your data (we don't do anything to it except summarize per normal). You can move licenses around as needed. Now, keep in mind that excessive movement of licenses could cause repository bloat as we'll keep 30 days of detail for…
-
Can you tell us what version of DPA you are on?
-
You are talking about DPA, right? If yes, do a query against the repository: select id, name from COND; That will show you what's being displayed. You can update that column directly with what you want to see (use _ or other strategies). update COND set name='INSTNAME' where id = X Shouldn't take more than just a couple…
-
Check out this post. DPA Notification Policy You may need to change the notification policy...
-
Since this involves sharing part of the ERD to get the data out of the repository, can you open a case with support? They should be able to get you pointed in the right direction.
-
select name, id from cond; -- this will give the id for the db you are tracking down the disabled query analysis for select * from con_problem_silence_<Id from above> this will show if there are any sql ids that are silenced and the date of the silencing You can delete the row from con_problem_silence_<id> to get the…
-
We don't persist any of that data currently on the Orion side. All data that you see displayed within Orion on the database tab is pulled via SOAP calls to the DPA web server. You'd have to query against the DPA repo to get that data.
-
To move licenses, you can deactivate them on your current DPA installation by going into license management and following this document: Activate DPA licenses - SolarWinds Worldwide, LLC. Help and Support If you would like to keep your existing history when moving, you can follow this article (the bottom section talks…
-
We run a query against the monitored instance once per minute. We look for any blocked sessions. If we see any, we plot them (per the graph). Since I see Batch Requests/Sec, I'll assume this is a SQL Server instance (correct me if I'm not right about that). I've included the query below. When you zoom out, I believe the…
-
Karthik, we poll or sample once every second. When we do that, we assign 1 second of blocking even if the block actually only took 10 ms. It's just that we "caught" it when it was actively blocking another SPID. Hope that helps explain why you see 1 second or in increments of even seconds.
