Hi Colin, I checked the latency prior to posting and between regions it is about 65ms on average. What should I set the latency to for Hazelcast?
Amir. On Tuesday, March 24, 2015 at 11:49:25 AM UTC-5, Colin wrote: > > Hi Amir, > > You might also do a ping and a traceroute between the machines and see > what kind of latency you're getting, just in case it's a timeout issue with > Hazelcast. > > -Colin > > On Tuesday, March 24, 2015 at 11:32:21 AM UTC-5, Amir Khawaja wrote: >> >> Hi Colin, >> >> Thank you for the prompt response. >> >> I'm a little confused as you say "the US West node will not come online >>> telling me that the database is not yet online. At that point, I kill the >>> process and then eventually the database comes online." >> >> Do you mean you kill the database process and then restart it and then it >>> starts communicating? >> >> >> Yes. I kill the database process on the cluster node where the OrientDB >> is not coming online. >> >> Can you see on each machine when Hazelcast 'sees' all the members? Are >>> all the members showing up? >> >> >> Yes. I see the databases are talking to each other as the IP address of >> the nodes show up in the log of each database server. >> >> I will try setting hotAlignment to false and report my results on this >> thread. >> >> Amir. >> >> >> On Tuesday, March 24, 2015 at 11:25:16 AM UTC-5, Colin wrote: >>> >>> Hi Amir, >>> >>> Is it consistently a problem between the same machines not seeing each >>> other? >>> >>> I'm a little confused as you say "the US West node will not come online >>> telling me that the database is not yet online. At that point, I kill the >>> process and then eventually the database comes online." >>> >>> Do you mean you kill the database process and then restart it and then >>> it starts communicating? >>> >>> In your distributed json file, try setting "hotAlignment" to false. >>> >>> Can you see on each machine when Hazelcast 'sees' all the members? Are >>> all the members showing up? >>> >>> -Colin >>> >>> Orient Technologies >>> >>> The Company behind OrientDB >>> >>> On Tuesday, March 24, 2015 at 11:19:05 AM UTC-5, Amir Khawaja wrote: >>>> >>>> Greetings, everyone. Has anyone had much success running an OrientDB >>>> 2.0.5 cluster in Azure? I created a cluster in Windows Azure with 4 nodes >>>> using CentOS 7 and OrientDB Community 2.0.4 -- 2 nodes in US East2 and 2 >>>> nodes in US West. There is a Site-to-Site VPN connection between the two >>>> regions in Azure and data is flowing between machines across the network. >>>> I >>>> have three databases that I have currently deployed and testing. I find >>>> that many times the synchronization between databases does not occur. For >>>> instance, if I startup the first node in US East2 and once that comes >>>> online, fire up the second node in US West, the US West node will not come >>>> online telling me that the database is not yet online. At that point, I >>>> kill the process and then eventually the database comes online. I even >>>> have >>>> to go so far as to delete the databases in the database path folder. I do >>>> this a few times and eventually the server may startup. Sometimes, I will >>>> have three of the four nodes working and the fourth just refuses to come >>>> online. >>>> >>>> The VM size selected for each node in the cluster is a D4 (4 cores, >>>> 28GB RAM). This should be more than sufficient to handle most loads. >>>> Surely, I must be missing something as this is not acceptable production >>>> behavior. For reference, I am pasting the hazelcast.xml and >>>> default-distributed-db-config.json files here in hopes that someone has >>>> some pointers for me. >>>> >>>> *** hazelcast.xml *** >>>> >>>> <?xml version="1.0" encoding="UTF-8"?> >>>> <!-- ~ Copyright (c) 2008-2012, Hazel Bilisim Ltd. All Rights Reserved. >>>> ~ >>>> ~ Licensed under the Apache License, Version 2.0 (the "License"); ~ you >>>> may >>>> not use this file except in compliance with the License. ~ You may >>>> obtain >>>> a copy of the License at ~ ~ http://www.apache.org/licenses/LICENSE-2.0 >>>> ~ >>>> ~ Unless required by applicable law or agreed to in writing, software ~ >>>> distributed >>>> under the License is distributed on an "AS IS" BASIS, ~ WITHOUT >>>> WARRANTIES >>>> OR CONDITIONS OF ANY KIND, either express or implied. ~ See the License >>>> for >>>> the specific language governing permissions and ~ limitations under the >>>> License. --> >>>> >>>> <hazelcast >>>> xsi:schemaLocation="http://www.hazelcast.com/schema/config >>>> hazelcast-config-3.0.xsd" >>>> xmlns="http://www.hazelcast.com/schema/config"; xmlns:xsi=" >>>> http://www.w3.org/2001/XMLSchema-instance";> >>>> <group> >>>> <name>[name]</name> >>>> <password>[password]</password> >>>> </group> >>>> <network> >>>> <port auto-increment="true">2434</port> >>>> <join> >>>> <multicast enabled="false"> >>>> <multicast-group>235.1.1.1</multicast-group> >>>> <multicast-port>2434</multicast-port> >>>> </multicast> >>>> <tcp-ip enabled="true"> >>>> <member>10.0.0.4</member> >>>> <member>10.0.0.5</member> >>>> <member>10.1.0.4</member> >>>> <member>10.1.0.5</member> >>>> </tcp-ip> >>>> </join> >>>> </network> >>>> <executor-service> >>>> <pool-size>16</pool-size> >>>> </executor-service> >>>> </hazelcast> >>>> >>>> >>>> *** default-distributed-db-config.json *** >>>> >>>> { >>>> "autoDeploy": true, >>>> "hotAlignment": true, >>>> "executionMode": "synchronous", >>>> "readQuorum": 1, >>>> "writeQuorum": 3, >>>> "failureAvailableNodesLessQuorum": false, >>>> "readYourWrites": true, >>>> "clusters": { >>>> "internal": { >>>> }, >>>> "index": { >>>> }, >>>> "*": { >>>> "servers" : [ "<NEW_NODE>" ] >>>> } >>>> } >>>> } >>>> >>>> Thank you for any assistance you can offer. >>>> >>>> Amir. >>>> >>> -- --- You received this message because you are subscribed to the Google Groups "OrientDB" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
