Question

Photo of Michael Garrison

0

Large number of exceptions and system instability

3

I get what seems to me to be a pretty large number of errors (OTMH I'd estimate it at between 30-50) every day from Rock. There are exceptions, but a vast majority of them include

--SqlException in .Net SqlClient Data Provider

Message
Timeout expired. The timeout period elapsed prior to completion of the operation or the server is not responding.

as well as

---Win32Exception in

Message
The wait operation timed out

These are frequently accompanied by a "Whoops, that wasn't supposed to happen" error page on the user side (not always- if it's an ajax request it just never populates)- reloading the page usually (but not always) works ok. Sometimes a series of these cause some pretty bad slowdowns, and sometimes they are even followed by an apparent complete Rock crash. It got pretty bad this weekend when Rock crashed for 10 minutes during the height of check-in.

I'm hosting locally, on a virtual server running on a physical box with a lot of resources.

So I'm unsure of how to fix this (or whether it's peculiar to my installation). I know others are hosting on virtual servers, so it doesn't seem like that's an issue. I hesitate to grant any kind of remote access to my production AD and file servers, even through such an apparently innocent route as IIS.

Still, I have two physical servers that the virtual server moves back and forth between, and so it should be possible to configure replication, failover, or some kind of load-sharing configuration between the virtual server and my physical servers so that I can have real resiliance for Rock. Is this possible with Sql Server Express or would I need to upgrade to the full version? (I'm somewhat experienced with SQL, but this is my first use of SQL Server). And would that be likely to help this particular issue? It seems like SQL Server keeps choking on requests so I'm trying to figure out how to give it whatever it needs.

  • Photo of Rock RMS

    0

    Any chance you could email the contents of your exception table to us (info(at)sparkdevnetwork.org)?  For performance we would would recommend running your SQL Server on a separate virtual machine from IIS (if your not) if possible.

  • Photo of Michael Garrison

    0

    E-mailed. I'll move our SQL installation off of our IIS virtual server. I just re-perused the "Installing" manual and have discovered how much larger our database is than I thought initially- yes, I imagine I'm running in a pretty undersized manner. Hopefully that's all it is.

    SQL Server 2014 express with tools is out- I assume I should probably go ahead and use this latest version of the software instead of 2012 as the manual says?

    • Rock RMS

      Rock RMS

      After looking into the logs we think this might be caused by a previous issue in the workflow processing. This has been fixed and will be in the next release.

      • Michael Garrison

        Michael Garrison

        OK, I've moved my SQL installation onto 2014 on another server installation. I'll let you know if I see a difference either way in the number of errors in the meantime.

      • Photo of Michael Garrison

        0

        I have not gotten a single timeout exception since Wednesday morning, when i pulled the SQL Server installation off of my virtual IIS server and put it on a physical internal-facing server. Thanks for your comment, making me reread the installation chapter; we're a stone's throw.from 10k records so we are well past the point where I should have put those functions on separate server instances. I think it's safe to say that the performance bottleneck was the cause of these evolutions