Universal Pollers data Slow into Database

Hi Guys

Not sure if anybody is experiencing the same issue that we do.

Basically we have a few Custom Pollers setup which retrieves data from devices.

This could be a UPS, Temperature Probes, Generators etc.

We need to identify things like Power Loss, Temperature fluctuations and Generator Failures etc as soon as possible.

We have all the OID values and we created the pollers without any major hiccups.

We then went about to do our alerting based on the pollers, when testing it we noticed huge delays in the alert triggering.

When we investigated we noticed that according to the chart data for the pollers, it may take up to 5 hours before data is written into the DB server.

This explains why the alerts does not trigger.

As for latency and Bandwidth statistics, all seems to be in the DB on time. No issues here.

We have moved the devices around to different Additional Polling engines to see if this makes a difference, it does seem to be in the DB faster but still takes between 30 minutes to 2 hours to go into the DB.

Any Advise on what we could check which could cause specifically the Universal Device Pollers data to be delayed.

This is our Setup.

We have our Core Server Virtualized with more than enough resources, hosted on proper and fast SAN hardware and drives.

Our DB server is hosted on a Physical Server with 16 Core 3.3Ghz CPU, Windows Server 2012, SQL Server 2012.

8x SSD 200GB Drives all in a RAID 10. Partitioned C:OS, D:Data, E: Logs and F: Backups.

We also have 3x APE's running in Virtualized infrastructure also on propper SAN hardware and drives.

We are running wit the following versions

Main Orion Server Details