Community
- Command Central
- MVP Program
- Monthly Mission
- Blogs
- Groups
- Events
- Media Vault
Products
- Observability
- Network Management
- Application Management
- IT Security
- IT Service Management
- System Management
- Database Management
Content Exchange
- SolarWinds Platform
- Server & Application Monitor
- Database Performance Analyzer
- Server Configuration Monitor
- Network Performance Monitor
- Network Configuration Manager
- SQL Sentry
- Web Help Desk
Free Tools & Trials

NPM 12.3 and SQL transactions filling log file

Every since we upgraded to 12.3 NPM i have been having a battle trying to keep the Database transaction log file from filling the entire log disk

I have done a lot of research and found some things.

As far as I can tell there is nothing wrong with the SQL server itself and Microsoft agrees.

Currently our SQL server is SQL server enterprise 2017, and we are running in an AlwaysOn HA pair. So the database is in full recovery mode and has transaction log backups running every 15 minutes.

After troubleshooting using DBCC OPENTRAN and traced back a large number of processes that are keeping the transaction logs from clearing I've found a few things.

using sp_who2 on the processes i see that the ProgramName is causing this every time: SolarWinds.Collector.Service@domain-SolarWinds.HardwareHealth.VMwareJobExtension.HardwareHealthVmwareResultProcessor

These SPID's will run the SQL server out of space in the span of about 2-3 days, and we have a 350GB log disk. This is mainly due to previously it ran out of space and we had to increase the size to allow SQL to continue functioning

I've had to resort to kill the offending processes. Obviously this isn't the ideal way to fix the issue but it was those processes or the entire Solarwinds environment.

Has anyone else had any issues like this?

Find more posts tagged with

database transaction logs

Accepted answers

All comments

bene

YES!!!! We've been fighting what sounds like the same issue. Every few days our log disk fills up and the whole environment blows up. Do you have an active ticket open with Solarwinds regarding this issue. I'd be happy to piggy-back on that ticket to try to increase visibility.

jimspencer

Well i've narrowed it down to solarwinds leaving processes without clearing them up. It wouldn't make much of a deal if you are in simple recovery mode, as it wouldn't do anything with the logs, but with full recovery it's a different story.

I have to go in and use DBCC OPENTRAN and kill off the processes that are more than a few hours old.

Most of the ones i've seen keep the log file full are more than 12 hours old and are suspended. It's not the ideal solution. But will keep SQL from crashing until it's fixed

bene

We reconfigured to simple mode yesterday and managed to not fill up our logs drive overnight, for the first time in several days. We're probably fine with simple recovery mode long term so this may be our solution. Best of luck.

jimspencer

I have a ticket open with Solarwinds, but they are telling me "Working as intended". I don't know how they sell this product as an Enterprise level application when it can't even support any form of High Availability. For SQL any form of HA requires full recovery mode. I'm going to write a script that will kill any process that has open transactions and over 24 hrs old. This isn't a solution, just a work around that allows our solarwinds environment doesn't crash every 2 days.

And what is even better, we have a project to implement solarwinds for a customer, and they already set up their SQL server in an HA pair, as they should. I'm not sure what Solarwinds expects me to tell them.