Re: [Neo4j] neo4j-enterprise-1.8 on linux : HA startup problem : org.neo4j.com.IllegalProtocolVersionException: Unexpected application protocol version 1, expected 4

159 views
Skip to first unread message

Mattias Persson

unread,
Oct 10, 2012, 8:48:12 AM10/10/12
to ne...@googlegroups.com
This could be a race condition where the first instance gets downgraded from master to slave somehow. The port for receiving communication from slaves and receiving communication from the master (where it's pushing transactions to slaves) are the same for any given instances, switching with the role of the database instance. Are the two databases empty?

2012/10/3 christophe <can...@googlemail.com>
hi,
i'm trying to configure a simple 2-nodes cluster (linux) using same version of neo4j-enterprise (1.8)
i've followed ha setup tutorial to configure coord.cfg neo4j.properties and neo4j-server.properties correctly
in server properties, i've uncommented the following properties
org.neo4j.server.webserver.address=0.0.0.0
org.neo4j.server.webserver.port=7474
org.neo4j.server.webserver.https.port=7473
org.neo4j.server.database.mode=HA
 
to start, i'm doing :
coordinator start on server1 -> OK
coordinator start on server2 -> OK
neo4j server start on server1 -> OK -> the webadmin console says it is master and there is 0 slave, which is normal
neo4j server start on server2 -> KO
the log gives:
10/4/12 12:59:39 AM org.neo4j.server.database.Database SEVERE: Failed to start database.
10/4/12 12:59:39 AM org.neo4j.server.CommunityBootstrapper SEVERE:
org.neo4j.com.IllegalProtocolVersionException: Unexpected application protocol version 1, expected 4. Header:[00001000][00000001]
        at org.neo4j.com.DechunkingChannelBuffer.assertSameProtocolVersion(DechunkingChannelBuffer.java:142)
        at org.neo4j.com.DechunkingChannelBuffer.readNextChunk(DechunkingChannelBuffer.java:104)
        at org.neo4j.com.DechunkingChannelBuffer.<init>(DechunkingChannelBuffer.java:59)
        at org.neo4j.com.Client.sendRequest(Client.java:205)
        at org.neo4j.kernel.ha.MasterClient18.getMasterIdForCommittedTx(MasterClient18.java:264)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.ensureDataConsistencyWithMaster(HighlyAvailableGraphDatabase.java:1151)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.reevaluateMyself(HighlyAvailableGraphDatabase.java:1011)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.newMaster(HighlyAvailableGraphDatabase.java:1365)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.start(HighlyAvailableGraphDatabase.java:622)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.<init>(HighlyAvailableGraphDatabase.java:279)
        at org.neo4j.kernel.HighlyAvailableGraphDatabase.<init>(HighlyAvailableGraphDatabase.java:209)
        at org.neo4j.server.enterprise.EnterpriseDatabase$DatabaseMode$2.createDatabase(EnterpriseDatabase.java:52)
        at org.neo4j.server.enterprise.EnterpriseDatabase.start(EnterpriseDatabase.java:73)
        at org.neo4j.server.AbstractNeoServer.start(AbstractNeoServer.java:103)
        at org.neo4j.server.Bootstrapper.start(Bootstrapper.java:79)
        at org.neo4j.server.Bootstrapper.main(Bootstrapper.java:46)
10/4/12 12:59:39 AM org.neo4j.server.CommunityBootstrapper SEVERE: Failed to start Neo Server on port [7474]
I'm wondering I've got such an error as the tar.gz used on both servers is exactly the same..
Any help would be appreciated.
TIA

--
 
 



--
Mattias Persson, [mat...@neotechnology.com]
Hacker, Neo Technology
www.neotechnology.com

christophe

unread,
Oct 30, 2012, 5:01:49 PM10/30/12
to ne...@googlegroups.com
Yes, the databases are empty. This is a fresh install on both sides. Therefore no object is created yet.

Mattias Persson

unread,
Nov 5, 2012, 12:42:38 PM11/5/12
to Neo4j Development
Any progress on this? Do you mind including your configuration files for your instances, as well as data/graph.db/messages.log?


2012/10/30 christophe <can...@googlemail.com>

Peter Neubauer

unread,
Nov 21, 2012, 3:53:16 AM11/21/12
to Neo4j User
Oskar, Christophe,
any chance to get some logs form the instances in order to be able to pinpoint the problem better?


/peter


Cheers,

/peter neubauer

G:  neubauer.peter
S:  peter.neubauer
P:  +46 704 106975
L:   http://www.linkedin.com/in/neubauer
T:   @peterneubauer

Neo4j 1.8 GA - http://www.dzone.com/links/neo4j_18_release_fluent_graph_literacy.html


On Wed, Nov 21, 2012 at 5:00 AM, Oskar Levin <oskar...@gmail.com> wrote:
Any progress on this issue? I have the same problem here, configured two embedded Neo4j servers for HA and used the 1.8_M06 package of Neo4j to create two coordinator databases. Are you running this on two separate machines or on the same machine?
--
 
 

Oskar Levin

unread,
Nov 22, 2012, 1:19:58 PM11/22/12
to ne...@googlegroups.com
Never mind about me, I managed to solve the issue. I configured ha.server on both the master and slave to an IP address the slave couldn't bind to. Thanks anyway!

2012/11/21 Peter Neubauer <peter.n...@neotechnology.com>
--
 
 



--
Med vänliga hälsningar
Oskar Levin

Peter Neubauer

unread,
Nov 23, 2012, 9:21:46 AM11/23/12
to Neo4j User
Nice.

Hope things are going well for you now.

/peter


Cheers,

/peter neubauer

G:  neubauer.peter
S:  peter.neubauer
P:  +46 704 106975
L:   http://www.linkedin.com/in/neubauer
T:   @peterneubauer

Neo4j 1.8 GA - http://www.dzone.com/links/neo4j_18_release_fluent_graph_literacy.html


--
 
 

Reply all
Reply to author
Forward
0 new messages