On 4/13/2015 10:11 PM, vsilgalis wrote: > just a couple of notes: > this a 2 shard setup with 2 nodes per shard. > > Currently these are on VMs with 8 cores and 8GB of ram each (java max heap > is ~5588mb but we usually never even get that high) backed by a NFS file > store which we store the indexes on (netapp SAN with nfs exports on SAS > disk).
Broken pipe errors usually indicate that the client gave up waiting for the server and disconnected the TCP connection before the server completed processing and sent a response. This is frequently because of configured timeouts on the client. If reasonable timeouts are being exceeded, it's usually a performance problem. You haven't indicated how much disk space is occupied by the index data on each of these servers. There are also several other things that would be helpful to know. Please read this wiki page, then come back with any questions you might have, and I may also ask a question or two: http://wiki.apache.org/solr/SolrPerformanceProblems My immediate suspects are an OS disk cache that is too small, and/or problems with garbage collection pauses. These are two of the issues discussed on that wiki page. Thanks, Shawn
