Hi Jörg,
thanks a lot for taking the time to answer! Of course I messed up my original message, and meant to say that the embedded version is actually 0.9.19 (missed out on the "1", shame on me!). The same is happening on Mac OS X 10.8 (Mountain Lion) as well as RedHat 6.0. Since this is a single node "cluster", it is not possible that different versions of the JRE are being used (at least I can't see how that would be possible, given that the client (content management system) and Elasticsearch are both running in the same VM).
Tobias
On 08.11.2012, at 18:57, Jörg Prante joergprante@gmail.com wrote:
Hi,
some thoughts about possible causes:
weird JRE class loader issue when using class loader transport.settings().getClassLoader()?
different JREs (JRE 6, JRE 7) in one cluster?
Netty object deserialization issue? Possibly related to Netty 3.x : A java.util.Locale encoded with ObjectEncoder in Java 1.6 cannot be decoded with ObjectDecoder in Java 1.7 · Issue #603 · netty/netty · GitHub
Ensuring a common JRE version in the cluster and upgrading to ES >= 0.19.10 may help.
Would be nice to get more setup parameters of the cluster (os, JVM), to get a chance to reproduce the issue.
The "Message not fully read" warning might be harmless.
Jörg
On Thursday, November 8, 2012 2:52:28 PM UTC+1, Tobias Wunden wrote:
The same happens to me (I see the same stack trace when using the code example that has been posted earlier) on Mac OS X as well as RedHat EL 6. Elasticsearch 0.19.9 is embedded, and I am using the following configuration (development setup):cluster.name: weblounge
node.name: "Localhost"
node.master: true
node.data: true
index.number_of_shards: 1
index.number_of_replicas: 0
http.enabled: trueEverything else is commented out (defaults). The error appears every time I restart. I am using elasticsearch inside an OSGi bundle (embeded), so it seems strange to me that the bundle classloader should not be able to find an elasticsearch class. I am creating the client within the same bundle.
I also see this in the logs (right after the stack trace):
14:28:19 WARN (Slf4jESLogger:104) [Localhost] Message not fully read (response) for [8] handler org.elasticsearch.action.support.master.TransportMasterNodeOperationAction$4@14ecf6c5, error [true], resetting.
Any help is appreciated, please let me know if I can be of assistance, i. e providing more information.
--
--