incubator-cloudstack-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Alex Mathiasen (JIRA)" <j...@apache.org>
Subject [jira] [Created] (CLOUDSTACK-525) Cloudstack Redundant Virtual Router issues both MASTER.
Date Wed, 21 Nov 2012 08:21:57 GMT
Alex Mathiasen created CLOUDSTACK-525:
-----------------------------------------

             Summary: Cloudstack Redundant Virtual Router issues both MASTER. 
                 Key: CLOUDSTACK-525
                 URL: https://issues.apache.org/jira/browse/CLOUDSTACK-525
             Project: CloudStack
          Issue Type: Bug
          Components: Install and Setup, KVM, Management Server, Network Controller, Network
Devices
    Affects Versions: 4.0.0
         Environment: Two servers with same hardware running Ubuntu 12.04 LTS. 
            Reporter: Alex Mathiasen
            Priority: Critical


Steps:

- Two hosts added to cluster
- Add network offering with RVR set, 
- Create instance using that network offering
- Look at the state of both virtual routers on each host being MASTER

Due to the state above, you will recieve errors with the following message:

More than one redundant virtual router is in MASTER state! Router r-15-VM and router r-14-VM
- Virtual router (name: r-15-VM, id: 15 and router (name: r-14-VM, id: 15) are both in MASTER
state! If the problem persist, restart both of routers. 

No errors in the logfiles, I have tried rebooting the virtual routers. Migrating the second
virtual router onto the same host will fix the state to MASTER and BACKUP, but it's not the
point - As the virtual router is supposed to run on each host. 

Looking at the console on both virtual routers, you will discover that keepalived, conntrackd
and haproxy failed to start. 

Trying to ping the virtual routers by their local-link address. from management-server, you
will discover the following result:

This is the server running on same host as the instance.
root@cloudsrv1:/# ping 169.254.0.169
PING 169.254.0.169 (169.254.0.169) 56(84) bytes of data.
64 bytes from 169.254.0.169: icmp_req=1 ttl=64 time=0.193 ms

This is the server running on the second host, which should behave as BACKUP.
root@cloudsrv1:/mnt/mysql# ping 169.254.3.121
PING 169.254.3.121 (169.254.3.121) 56(84) bytes of data.
>From 169.254.0.1 icmp_seq=1 Destination Host Unreachable

View from management-server.log:

2012-11-21 09:19:42,066 DEBUG [agent.transport.Request] (RouterStatusMonitor-1:null) Seq 3-2029663984:
Sending  { Cmd , MgmtId: 279278805451002, via: 3, Ver: v1, Flags: 100011, [{"CheckRouterCommand":{"accessDetails":{"router.ip":"169.254.0.169","router.name":"r-12-VM"},"wait":60}}]
}
2012-11-21 09:19:42,212 DEBUG [agent.transport.Request] (AgentManager-Handler-8:null) Seq
3-2029663984: Processing:  { Ans: , MgmtId: 279278805451002, via: 3, Ver: v1, Flags: 10, [{"CheckRouterAnswer":{"state":"MASTER","isBumped":false,"result":true,"details":"Status:
MASTER&Bumped: NO","wait":0}}] }
2012-11-21 09:19:42,212 DEBUG [agent.transport.Request] (RouterStatusMonitor-1:null) Seq 3-2029663984:
Received:  { Ans: , MgmtId: 279278805451002, via: 3, Ver: v1, Flags: 10, { CheckRouterAnswer
} }
2012-11-21 09:19:42,213 DEBUG [agent.manager.AgentManagerImpl] (RouterStatusMonitor-1:null)
Details from executing class com.cloud.agent.api.CheckRouterCommand: Status: MASTER&Bumped:
NO
2012-11-21 09:19:42,267 DEBUG [agent.transport.Request] (RouterStatusMonitor-1:null) Seq 5-1237057653:
Sending  { Cmd , MgmtId: 279278805451002, via: 5, Ver: v1, Flags: 100011, [{"CheckRouterCommand":{"accessDetails":{"router.ip":"169.254.3.121","router.name":"r-13-VM"},"wait":60}}]
}
2012-11-21 09:19:42,421 DEBUG [agent.transport.Request] (AgentManager-Handler-14:null) Seq
5-1237057653: Processing:  { Ans: , MgmtId: 279278805451002, via: 5, Ver: v1, Flags: 10, [{"CheckRouterAnswer":{"state":"MASTER","isBumped":false,"result":true,"details":"Status:
MASTER&Bumped: NO","wait":0}}] }
2012-11-21 09:19:42,421 DEBUG [agent.transport.Request] (RouterStatusMonitor-1:null) Seq 5-1237057653:
Received:  { Ans: , MgmtId: 279278805451002, via: 5, Ver: v1, Flags: 10, { CheckRouterAnswer
} }
2012-11-21 09:19:42,421 DEBUG [agent.manager.AgentManagerImpl] (RouterStatusMonitor-1:null)
Details from executing class com.cloud.agent.api.CheckRouterCommand: Status: MASTER&Bumped:
NO

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Mime
View raw message