SearchStax Managed Search service users sometimes encounter 500 Internal Server errors. This is a generic HTTP message, given when an unexpected condition was encountered and no more specific message could be provided. Many users have contacted the SearchStax support desk looking for an explanation.
Although the problem is officially “unknown,” a review of case histories reveals some “usual suspects” to consider.
Common Causes of 500 Errors
- A 500 Server Error often means that the deployment has run out of disk space. Zookeeper stopped running or became corrupted. A clean restart may fix Zookeeper, but you should also consider adding disk space or upgrading to a larger deployment.
- A similar situation, caused by low disk space or by a spike in system load, can stop the Solr IndexWriter process. Subsequent attempts to copy index segments to a replica result in Server 500 messages. Restart Solr.
- In Sitecore deployments, a mistake in the connection string can produce Server 500 messages. See the notes under Sitecore Connection Errors.
- In deployments that use Sitecore’s switchOnRebuild feature, a mistake in setting up the active and rebuild aliases can send updates to a collection that doesn’t exist. We sometimes see this in xDB indexes that have been created using the default Sitecore configset instead of the special configset provided with xDB.
- We have seen Server 500 errors when attempting to use an APIkey with a SearchStax API function that does not accept APIkeys.
Questions?
Do not hesitate to contact the SearchStax Support Desk.