FTView Historian SE IncuityDiagnotics file size blowout

[ASF]

Hi all,

Just recently put in my first FTView Historian server. I also installed VantagePoint for the client to use to gather and manipulate the data. Less than a week later, the hard drive on the historian machine was full. It's a virtual server, and only has a 40GB hard drive, so we'll expand that no worries, but I tracked the bulk of the growth down to the IncuityDiagnostics database in the SQL server. It's currently 8GB and growing at a rate of 1-2GB per day. The autogrowth settings are set to 1MB, no limit, so it's obviously filling up with something, but I have no idea why. Google and the Rockwell KB seem to have no ideas, has anyone seen this before? Can anyone give me any idea what sort of data would actually be stored here, so I can work out how it's getting filled up? Can I clear out the data to free up space somehow? I've put in a call to tech support but their track record with callbacks has been pretty woeful of late so I'm not optimistic about learning anything quickly.

Any ideas welcome!

 

[ASF]

Well Rockwell Tech support just made a liar out of me

Their immediate solution was to disable event reporting: VantagePoint Manager>System>Services>Storage right click on Diagnostics and click Edit. Uncheck the "Allow Direct Queries" and "Enabled" boxes.

Apparently I can also use a Size Based Aging Strategy in the same window to set up the diagnostics log to only keep the most recent logs after a certain size. But in any case, the Windows event viewer mirrors all the logs, so it's really not needed in my case. Come to think of it, Windows Event Log has probably just grown by the same amount, which would explain why the hard drive filled up so fast

Now to work out what it is that was actually filling up the database so fast! Any ideas as to how to find out?

 

[Operaghost]

Open your SQL Server Management Studio and navigate to the Databases ==> IncuityDiagnostics. From there expand the Tables and you will see several "dbo.History..." tables. Right-click on one and select Properties.

In the upper left click Storage and then on the right in the general section you will see "Row count" and "Data space" properties. Check those to see which is eating up the space.

You can open the one storing all the data and view the records. Hopefully you can make out from there what is getting logged.

OG

 

[ASF]

OK. So it's all in the dbo.LogEntries table. Attached is a csv of the error list - there seems to be a common thread in there, but I'm not sure what to make of it. There's mention of a failed login, and of a timeout to a service? I've checked that the service is running.

 

[ASF]

So what I'd forgotten is that I attached the Historian to an existing SQL server on another machine. The windows event log on the SQL server machine was reporting a failed login from my computer name at the same frequency my VantagePoint SQL database on the Historian machine was logging errors (19,000 of them, in fact).

I had to add the computer to the SQL server as a login and assign it dbread access to the PIAF database (according to rockwell tech support, the error being thrown up was related to a failed read attempt). Once I did that, the login failure messages stopped straight away.

Now I've just got to empty out the LogEntries table and shrink my database back down to size...which shoudl be straightforward, apart from apparently I don't have enough disk space to run the delete command? Hmm. Got the IT department looking at it for me now...