Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
No, once a cluster is up, it will be able to recover after being down for a while. Cheers, /peter neubauer Google:neubauer.peter Skype:peter.neubauer Phone: +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter @peterneubauer brew install neo4j && neo4j start heroku addons:add neo4j On Fri, Dec 23, 2011 at 1:09 PM, Anirudha Ravindra Ghante wrote: > Hi Peter, > > We are not trying to run multiple neo4j clusters on same machine. > Configuration is : > 3 machines : each running a ZK and neo4j server. > > So, does this mean that if one of the nodes goes down, we have to delete > the graph.db before restarting the neo4j server again (even with the above > configuration)? > > > On Fri, Dec 23, 2011 at 6:38 AM, Peter Neubauer < > peter.neuba...@neotechnology.com> wrote: > >> Anirudha, >> I think this might be a bug that is fixed in Master. Now, the Neo4j >> Coordinator ZK cluster can manage several HA clusters and should not >> give that exception. However, do you want to run multiple clusters >> there? Otherwise, removing the DB and starting fresh so it gets copied >> over from the master is the way to go. >> >> Cheers, >> >> /peter neubauer >> >> Google:neubauer.peter >> Skype:peter.neubauer >> Phone: +46 704 106975 >> LinkedIn http://www.linkedin.com/in/neubauer >> Twitter @peterneubauer >> >> brew install neo4j && neo4j start >> heroku addons:add neo4j >> >> >> >> On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: >> > Hi, >> > >> > We are trying to setup neo4j in high-availability mode with 3 nodes in >> the >> > cluster. >> > When we restart any of the neo4j servers, it fails. We get following >> > exception: >> > >> > ==> neo4j-zookeeper.log <== >> > ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while >> calling >> > watcher >> > java.lang.RuntimeException: There's already an HA cluster managed by this >> > ZooKeeper cluster >> > at >> org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) >> > at >> > >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) >> > at >> > >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) >> > at >> org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) >> > at >> > >> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) >> > at >> org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) >> > >> > >> > If we delete the database folder "graph.db" and then start the server, it >> > starts with no exception and syncs the data. >> > Am I missing any configuration? >> > >> > Thanks. >> > >> > -- >> > View this message in context: >> http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html >> > Sent from the Neo4j Community Discussions mailing list archive at >> Nabble.com. >> > ___ >> > Neo4j mailing list >> > User@lists.neo4j.org >> > https://lists.neo4j.org/mailman/listinfo/user >> ___ >> NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please >> register and consider posting at >> https://groups.google.com/forum/#!forum/neo4j >> >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user >> > ___ > NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register > and consider posting at https://groups.google.com/forum/#!forum/neo4j > > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Hi Peter, We are not trying to run multiple neo4j clusters on same machine. Configuration is : 3 machines : each running a ZK and neo4j server. So, does this mean that if one of the nodes goes down, we have to delete the graph.db before restarting the neo4j server again (even with the above configuration)? On Fri, Dec 23, 2011 at 6:38 AM, Peter Neubauer < peter.neuba...@neotechnology.com> wrote: > Anirudha, > I think this might be a bug that is fixed in Master. Now, the Neo4j > Coordinator ZK cluster can manage several HA clusters and should not > give that exception. However, do you want to run multiple clusters > there? Otherwise, removing the DB and starting fresh so it gets copied > over from the master is the way to go. > > Cheers, > > /peter neubauer > > Google:neubauer.peter > Skype:peter.neubauer > Phone: +46 704 106975 > LinkedIn http://www.linkedin.com/in/neubauer > Twitter @peterneubauer > > brew install neo4j && neo4j start > heroku addons:add neo4j > > > > On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: > > Hi, > > > > We are trying to setup neo4j in high-availability mode with 3 nodes in > the > > cluster. > > When we restart any of the neo4j servers, it fails. We get following > > exception: > > > > ==> neo4j-zookeeper.log <== > > ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while > calling > > watcher > > java.lang.RuntimeException: There's already an HA cluster managed by this > > ZooKeeper cluster > >at > org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) > >at > > > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) > >at > > > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) > >at > org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) > >at > > > org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) > >at > org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) > > > > > > If we delete the database folder "graph.db" and then start the server, it > > starts with no exception and syncs the data. > > Am I missing any configuration? > > > > Thanks. > > > > -- > > View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html > > Sent from the Neo4j Community Discussions mailing list archive at > Nabble.com. > > ___ > > Neo4j mailing list > > User@lists.neo4j.org > > https://lists.neo4j.org/mailman/listinfo/user > ___ > NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please > register and consider posting at > https://groups.google.com/forum/#!forum/neo4j > > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user > ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Anirudha, I think this might be a bug that is fixed in Master. Now, the Neo4j Coordinator ZK cluster can manage several HA clusters and should not give that exception. However, do you want to run multiple clusters there? Otherwise, removing the DB and starting fresh so it gets copied over from the master is the way to go. Cheers, /peter neubauer Google:neubauer.peter Skype:peter.neubauer Phone: +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter @peterneubauer brew install neo4j && neo4j start heroku addons:add neo4j On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: > Hi, > > We are trying to setup neo4j in high-availability mode with 3 nodes in the > cluster. > When we restart any of the neo4j servers, it fails. We get following > exception: > > ==> neo4j-zookeeper.log <== > ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while calling > watcher > java.lang.RuntimeException: There's already an HA cluster managed by this > ZooKeeper cluster > at org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) > at > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) > at > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) > at org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) > at > org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) > at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) > > > If we delete the database folder "graph.db" and then start the server, it > starts with no exception and syncs the data. > Am I missing any configuration? > > Thanks. > > -- > View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html > Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Hey Peter, This link should work : https://gist.github.com/1436975 Thanks On Fri, Dec 16, 2011 at 11:41 AM, Peter Neubauer wrote: > Anirudha, > do you have a working link to the logs/gist? > > Cheers, > > /peter neubauer > > TC CEO of the year - vote for Emil Eifrém! > http://crunchies2011.techcrunch.com/nominate/ > > Google:neubauer.peter > Skype:peter.neubauer > Phone: +46 704 106975 > LinkedIn http://www.linkedin.com/in/neubauer > Twitter @peterneubauer > > brew install neo4j && neo4j start > heroku addons:add neo4j > > > > On Wed, Dec 14, 2011 at 2:12 PM, Michael Hunger > wrote: >> But it is a cute octocat github 404 :) >> >> the part w/o the ;cid... works >> >> https://gist.github.com/1433340 >> >> Cheers >> >> Michael >> >> Am 14.12.2011 um 13:18 schrieb Mattias Persson: >> >>> I'm getting a 404 on that link. >>> >>> 2011/12/6 Michael Hunger >>> Answer from Mahesh: Hi, Anirudha is a part of our team. We are using 1.5 in a 3 server setup on different physical machines. The messages.log from the server failing to start is as follows. (https://gist.github.com/1433340;cid=1323211777524-299) Here is the messages.log file on the server that is failing Hope this helps you help us.. Cheers... Mahesh Am 05.12.2011 um 09:03 schrieb Peter Neubauer: > Hi there, > what version are you running, any chance you could collect the > messages.log from the instances and send them over? > > Cheers, > > /peter neubauer > > GTalk: neubauer.peter > Skype peter.neubauer > Phone +46 704 106975 > LinkedIn http://www.linkedin.com/in/neubauer > Twitter http://twitter.com/peterneubauer > > brew install neo4j && neo4j start > heroku addons:add neo4j > > > > On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: >> Hi, >> >> We are trying to setup neo4j in high-availability mode with 3 nodes in the >> cluster. >> When we restart any of the neo4j servers, it fails. We get following >> exception: >> >> ==> neo4j-zookeeper.log <== >> ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while calling >> watcher >> java.lang.RuntimeException: There's already an HA cluster managed by this >> ZooKeeper cluster >> at org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) >> at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) >> at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) >> at org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) >> at >> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) >> at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) >> >> >> If we delete the database folder "graph.db" and then start the server, it >> starts with no exception and syncs the data. >> Am I missing any configuration? >> >> Thanks. >> >> -- >> View this message in context: http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html >> Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. >> ___ >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user >>> >>> >>> >>> -- >>> Mattias Persson, [matt...@neotechnology.com] >>> Hacker, Neo Technology >>> www.neotechnology.com >>> ___ >>> NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please >>> register and consider posting at >>> https://groups.google.com/forum/#!forum/neo4j >>> >>> Neo4j mailing list >>> User@lists.neo4j.org >>> https://lists.neo4j.org/mailman/listinfo/user >> >> ___ >> NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please >> register and consider posting at >> https://groups.google.com/forum/#!forum/neo4j >> >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user > ___ > NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register > and consider posting at https://groups.google.com/forum/#!forum/neo4j > > Neo
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Anirudha, do you have a working link to the logs/gist? Cheers, /peter neubauer TC CEO of the year - vote for Emil Eifrém! http://crunchies2011.techcrunch.com/nominate/ Google:neubauer.peter Skype:peter.neubauer Phone: +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter @peterneubauer brew install neo4j && neo4j start heroku addons:add neo4j On Wed, Dec 14, 2011 at 2:12 PM, Michael Hunger wrote: > But it is a cute octocat github 404 :) > > the part w/o the ;cid... works > > https://gist.github.com/1433340 > > Cheers > > Michael > > Am 14.12.2011 um 13:18 schrieb Mattias Persson: > >> I'm getting a 404 on that link. >> >> 2011/12/6 Michael Hunger >> >>> Answer from Mahesh: >>> >>> Hi, >>> >>> Anirudha is a part of our team. >>> >>> We are using 1.5 in a 3 server setup on different physical machines. >>> >>> The messages.log from the server failing to start is as follows. >>> (https://gist.github.com/1433340;cid=1323211777524-299) >>> Here is the messages.log file on the server that is failing >>> >>> Hope this helps you help us.. >>> >>> Cheers... >>> >>> Mahesh >>> >>> Am 05.12.2011 um 09:03 schrieb Peter Neubauer: >>> Hi there, what version are you running, any chance you could collect the messages.log from the instances and send them over? Cheers, /peter neubauer GTalk: neubauer.peter Skype peter.neubauer Phone +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter http://twitter.com/peterneubauer brew install neo4j && neo4j start heroku addons:add neo4j On Fri, Dec 2, 2011 at 3:08 PM, anirudha >>> wrote: > Hi, > > We are trying to setup neo4j in high-availability mode with 3 nodes in >>> the > cluster. > When we restart any of the neo4j servers, it fails. We get following > exception: > > ==> neo4j-zookeeper.log <== > ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while >>> calling > watcher > java.lang.RuntimeException: There's already an HA cluster managed by >>> this > ZooKeeper cluster > at >>> org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) > at > >>> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) > at > >>> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) > at >>> org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) > at > >>> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) > at >>> org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) > > > If we delete the database folder "graph.db" and then start the server, >>> it > starts with no exception and syncs the data. > Am I missing any configuration? > > Thanks. > > -- > View this message in context: >>> http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html > Sent from the Neo4j Community Discussions mailing list archive at >>> Nabble.com. > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user >>> >>> ___ >>> Neo4j mailing list >>> User@lists.neo4j.org >>> https://lists.neo4j.org/mailman/listinfo/user >>> >> >> >> >> -- >> Mattias Persson, [matt...@neotechnology.com] >> Hacker, Neo Technology >> www.neotechnology.com >> ___ >> NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please >> register and consider posting at >> https://groups.google.com/forum/#!forum/neo4j >> >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user > > ___ > NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register > and consider posting at https://groups.google.com/forum/#!forum/neo4j > > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
But it is a cute octocat github 404 :) the part w/o the ;cid... works https://gist.github.com/1433340 Cheers Michael Am 14.12.2011 um 13:18 schrieb Mattias Persson: > I'm getting a 404 on that link. > > 2011/12/6 Michael Hunger > >> Answer from Mahesh: >> >> Hi, >> >> Anirudha is a part of our team. >> >> We are using 1.5 in a 3 server setup on different physical machines. >> >> The messages.log from the server failing to start is as follows. >> (https://gist.github.com/1433340;cid=1323211777524-299) >> Here is the messages.log file on the server that is failing >> >> Hope this helps you help us.. >> >> Cheers... >> >> Mahesh >> >> Am 05.12.2011 um 09:03 schrieb Peter Neubauer: >> >>> Hi there, >>> what version are you running, any chance you could collect the >>> messages.log from the instances and send them over? >>> >>> Cheers, >>> >>> /peter neubauer >>> >>> GTalk: neubauer.peter >>> Skype peter.neubauer >>> Phone +46 704 106975 >>> LinkedIn http://www.linkedin.com/in/neubauer >>> Twitter http://twitter.com/peterneubauer >>> >>> brew install neo4j && neo4j start >>> heroku addons:add neo4j >>> >>> >>> >>> On Fri, Dec 2, 2011 at 3:08 PM, anirudha >> wrote: Hi, We are trying to setup neo4j in high-availability mode with 3 nodes in >> the cluster. When we restart any of the neo4j servers, it fails. We get following exception: ==> neo4j-zookeeper.log <== ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while >> calling watcher java.lang.RuntimeException: There's already an HA cluster managed by >> this ZooKeeper cluster at >> org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) at >> org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) at >> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) at >> org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) If we delete the database folder "graph.db" and then start the server, >> it starts with no exception and syncs the data. Am I missing any configuration? Thanks. -- View this message in context: >> http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html Sent from the Neo4j Community Discussions mailing list archive at >> Nabble.com. ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user >>> ___ >>> Neo4j mailing list >>> User@lists.neo4j.org >>> https://lists.neo4j.org/mailman/listinfo/user >> >> ___ >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user >> > > > > -- > Mattias Persson, [matt...@neotechnology.com] > Hacker, Neo Technology > www.neotechnology.com > ___ > NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register > and consider posting at https://groups.google.com/forum/#!forum/neo4j > > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
I'm getting a 404 on that link. 2011/12/6 Michael Hunger > Answer from Mahesh: > > Hi, > > Anirudha is a part of our team. > > We are using 1.5 in a 3 server setup on different physical machines. > > The messages.log from the server failing to start is as follows. > (https://gist.github.com/1433340;cid=1323211777524-299) > Here is the messages.log file on the server that is failing > > Hope this helps you help us.. > > Cheers... > > Mahesh > > Am 05.12.2011 um 09:03 schrieb Peter Neubauer: > > > Hi there, > > what version are you running, any chance you could collect the > > messages.log from the instances and send them over? > > > > Cheers, > > > > /peter neubauer > > > > GTalk: neubauer.peter > > Skype peter.neubauer > > Phone +46 704 106975 > > LinkedIn http://www.linkedin.com/in/neubauer > > Twitter http://twitter.com/peterneubauer > > > > brew install neo4j && neo4j start > > heroku addons:add neo4j > > > > > > > > On Fri, Dec 2, 2011 at 3:08 PM, anirudha > wrote: > >> Hi, > >> > >> We are trying to setup neo4j in high-availability mode with 3 nodes in > the > >> cluster. > >> When we restart any of the neo4j servers, it fails. We get following > >> exception: > >> > >> ==> neo4j-zookeeper.log <== > >> ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while > calling > >> watcher > >> java.lang.RuntimeException: There's already an HA cluster managed by > this > >> ZooKeeper cluster > >>at > org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) > >>at > >> > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) > >>at > >> > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) > >>at > org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) > >>at > >> > org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) > >>at > org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) > >> > >> > >> If we delete the database folder "graph.db" and then start the server, > it > >> starts with no exception and syncs the data. > >> Am I missing any configuration? > >> > >> Thanks. > >> > >> -- > >> View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html > >> Sent from the Neo4j Community Discussions mailing list archive at > Nabble.com. > >> ___ > >> Neo4j mailing list > >> User@lists.neo4j.org > >> https://lists.neo4j.org/mailman/listinfo/user > > ___ > > Neo4j mailing list > > User@lists.neo4j.org > > https://lists.neo4j.org/mailman/listinfo/user > > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user > -- Mattias Persson, [matt...@neotechnology.com] Hacker, Neo Technology www.neotechnology.com ___ NOTICE: THIS MAILING LIST IS BEING SWITCHED TO GOOGLE GROUPS, please register and consider posting at https://groups.google.com/forum/#!forum/neo4j Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Answer from Mahesh: Hi, Anirudha is a part of our team. We are using 1.5 in a 3 server setup on different physical machines. The messages.log from the server failing to start is as follows. (https://gist.github.com/1433340;cid=1323211777524-299) Here is the messages.log file on the server that is failing Hope this helps you help us.. Cheers... Mahesh Am 05.12.2011 um 09:03 schrieb Peter Neubauer: > Hi there, > what version are you running, any chance you could collect the > messages.log from the instances and send them over? > > Cheers, > > /peter neubauer > > GTalk: neubauer.peter > Skype peter.neubauer > Phone +46 704 106975 > LinkedIn http://www.linkedin.com/in/neubauer > Twitter http://twitter.com/peterneubauer > > brew install neo4j && neo4j start > heroku addons:add neo4j > > > > On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: >> Hi, >> >> We are trying to setup neo4j in high-availability mode with 3 nodes in the >> cluster. >> When we restart any of the neo4j servers, it fails. We get following >> exception: >> >> ==> neo4j-zookeeper.log <== >> ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while calling >> watcher >> java.lang.RuntimeException: There's already an HA cluster managed by this >> ZooKeeper cluster >>at org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) >>at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) >>at >> org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) >>at org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) >>at >> org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) >>at >> org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) >> >> >> If we delete the database folder "graph.db" and then start the server, it >> starts with no exception and syncs the data. >> Am I missing any configuration? >> >> Thanks. >> >> -- >> View this message in context: >> http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html >> Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. >> ___ >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Answer from Mahesh (not getting through nabble): Hi, The issue is with any server that is a part of the cluster. It is reproducible at will. The method we were following was - 1) start all three co-ordinators 2) start all three servers 3) shut down one server 4) start the same server again - it fails You can take a look at all the three messages.log files here. The file called slave2_messages_failing.log is the server that fails to start. The other two instances' messages.log files are also present in the same location. I tried the scenario outlined -> - Create the cluster from scratch - Restart one cluster member - It will fail as you have described - Wipe that member's database, start it - It will successfully join the cluster, as you have described. - What happens if you restart it again? Even in the above case, the server fails to start. The issue can be reproduced on any machine - not necessarily a slave or a master. Mahesh Am 05.12.2011 um 15:52 schrieb Chris Gioran: > Hi, > > Two questions, two requests: > > 1) Does this happen only on one machine or any member of the cluster? > Is it reproducible at will? > 1) Could you please provide the messages.log for the other two machines as > well? > 2) Can you please tell me the result of this scenario: >- Create the cluster from scratch >- Restart one cluster member > - It will fail as you have described >- Wipe that member's database, start it > - It will successfully join the cluster, as you have described. >- What happens if you restart it again? > > thank you, > CG > > On Mon, Dec 5, 2011 at 2:26 PM, anirudha wrote: >> Link for the messages.log file : https://gist.github.com/1433340 >> >> -- >> View this message in context: >> http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3561268.html >> Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. >> ___ >> Neo4j mailing list >> User@lists.neo4j.org >> https://lists.neo4j.org/mailman/listinfo/user > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Hi, Two questions, two requests: 1) Does this happen only on one machine or any member of the cluster? Is it reproducible at will? 1) Could you please provide the messages.log for the other two machines as well? 2) Can you please tell me the result of this scenario: - Create the cluster from scratch - Restart one cluster member - It will fail as you have described - Wipe that member's database, start it - It will successfully join the cluster, as you have described. - What happens if you restart it again? thank you, CG On Mon, Dec 5, 2011 at 2:26 PM, anirudha wrote: > Link for the messages.log file : https://gist.github.com/1433340 > > -- > View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3561268.html > Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Link for the messages.log file : https://gist.github.com/1433340 -- View this message in context: http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3561268.html Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
Re: [Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Hi there, what version are you running, any chance you could collect the messages.log from the instances and send them over? Cheers, /peter neubauer GTalk: neubauer.peter Skype peter.neubauer Phone +46 704 106975 LinkedIn http://www.linkedin.com/in/neubauer Twitter http://twitter.com/peterneubauer brew install neo4j && neo4j start heroku addons:add neo4j On Fri, Dec 2, 2011 at 3:08 PM, anirudha wrote: > Hi, > > We are trying to setup neo4j in high-availability mode with 3 nodes in the > cluster. > When we restart any of the neo4j servers, it fails. We get following > exception: > > ==> neo4j-zookeeper.log <== > ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while calling > watcher > java.lang.RuntimeException: There's already an HA cluster managed by this > ZooKeeper cluster > at org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) > at > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) > at > org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) > at org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) > at > org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) > at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) > > > If we delete the database folder "graph.db" and then start the server, it > starts with no exception and syncs the data. > Am I missing any configuration? > > Thanks. > > -- > View this message in context: > http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html > Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. > ___ > Neo4j mailing list > User@lists.neo4j.org > https://lists.neo4j.org/mailman/listinfo/user ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
[Neo4j] There's already an HA cluster managed by this ZooKeeper cluster
Hi, We are trying to setup neo4j in high-availability mode with 3 nodes in the cluster. When we restart any of the neo4j servers, it fails. We get following exception: ==> neo4j-zookeeper.log <== ERROR main-EventThread org.apache.zookeeper.ClientCnxn - Error while calling watcher java.lang.RuntimeException: There's already an HA cluster managed by this ZooKeeper cluster at org.neo4j.kernel.ha.zookeeper.ZooClient.getRoot(ZooClient.java:370) at org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getAllMachines(AbstractZooKeeperManager.java:194) at org.neo4j.kernel.ha.zookeeper.AbstractZooKeeperManager.getMasterFromZooKeeper(AbstractZooKeeperManager.java:134) at org.neo4j.kernel.ha.zookeeper.ZooClient.process(ZooClient.java:122) at org.apache.zookeeper.ClientCnxn$EventThread.processEvent(ClientCnxn.java:530) at org.apache.zookeeper.ClientCnxn$EventThread.run(ClientCnxn.java:506) If we delete the database folder "graph.db" and then start the server, it starts with no exception and syncs the data. Am I missing any configuration? Thanks. -- View this message in context: http://neo4j-community-discussions.438527.n3.nabble.com/There-s-already-an-HA-cluster-managed-by-this-ZooKeeper-cluster-tp3554731p3554731.html Sent from the Neo4j Community Discussions mailing list archive at Nabble.com. ___ Neo4j mailing list User@lists.neo4j.org https://lists.neo4j.org/mailman/listinfo/user
