MVEL infinite loop causing cluster performance degradation

Hi all,

I wonder if anyone else has seen an error similar to this:

Basically, MVEL gets into an infinite loop in its,
presumably when trying to report an error. This causes the affected node to
hang completely, which in turn causes a pretty serious performance
degradation in the cluster, manifested as cascading timed out requests
until the affected node is restarted.

I've already shared some logs with Shay, and I'm still working to identify
a request that can consistently duplicate this issue. But in the mean time
I thought I'd also mention it here to see if anyone else has seen a similar
error and might be able to share their observations.

Nick Zadrozny

Hosted full-text search, with Solr and Elasticsearch.

Let's talk in real time:

You received this message because you are subscribed to the Google Groups "elasticsearch" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
For more options, visit