Troubleshooting the Scheduler
We have seen a influx of cases regarding the Scheduler service, so I thought it would be a good time to write about how to troubleshoot it yourself!
The scheduling service run jobs you define for it and also EPiServer's own shipped jobs, such as archiving pages, subscription and delete wastebasket pages.
But sometimes it goes wrong and the scheduler doesn't automatically run the jobs anymore. It's stuck.
We have a few tactics we use. First one is to delete the, possibly corrupted, config file and there's a FAQ item describing how to do that. If that doesn't work we usually suggest to the client to delete all the rows in the two tables the scheduler uses, tblScheduledItem and tblScheduledItemLog. To be able to do that you must first deactivate the foreign key defined on tblScheduledItemLog to tblScheduledItem. This can be achieved using the first line below.
ALTER TABLE tblScheduledItemLog NOCHECK CONSTRAINT ALL
ALTER TABLE tblScheduledItemLog CHECK CONSTRAINT ALL
Don't worry about truncating these two tables as they are recreated by the Scheduler service. But as always when making changes directly in the database you should take a backup. Safety first, don't forget that!
The third way is to figure it out yourself by logging the Scheduler service. The scheduler service can use the log4net logging framework (as all of our services), so you only need to drop a log4net.dll and a log4net.config (see the fileAppender section) in its folder for it to start logging. This way you can see what's causing the scheduler to fail and hopefully you can fix it yourself.
Example of a log4net.config file
<?xml version="1.0" encoding="utf-8" ?>
<appender name="ErrorAppender" type="log4net.Appender.FileAppender" >
<file value="C:\temp\LogServerError.log" />
<appendToFile value="true" />
<conversionPattern value="%d [%t] %l - %m%n" />
<level value="ALL" />
<appender-ref ref="ErrorAppender" />