[
https://issues.apache.org/jira/browse/CLOUDSTACK-8851?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14768923#comment-14768923
]
ASF GitHub Bot commented on CLOUDSTACK-8851:
--------------------------------------------
Github user remibergsma commented on the pull request:
https://github.com/apache/cloudstack/pull/839#issuecomment-140743242
- Deployed Advanced zone with two KVM clusters
- Created redundant router offering
- Created network
- Created VM in network
Two routers were started, each on one cluster:
```
MariaDB [cloud]> select name,host_id from vm_instance;
+---------+---------+
| name | host_id |
+---------+---------+
| s-1-VM | 1 |
| v-2-VM | 1 |
| testvm1 | 4 |
| r-4-VM | 4 |
| r-5-VM | 1 |
+---------+---------+
5 rows in set (0.00 sec)
MariaDB [cloud]> select name,id,cluster_id from host;
+--------+----+------------+
| name | id | cluster_id |
+--------+----+------------+
| kvm3 | 1 | 1 |
| v-2-VM | 2 | NULL |
| s-1-VM | 3 | NULL |
| kvm1 | 4 | 2 |
+--------+----+------------+
4 rows in set (0.00 sec)
MariaDB [cloud]> select id,name from cluster;
+----+------------+
| id | name |
+----+------------+
| 2 | MCCT-KVM-1 |
| 1 | MCCT-KVM-3 |
+----+------------+
2 rows in set (0.00 sec)
```
Log message:
```
2015-09-16 13:17:49,543 DEBUG [c.c.v.VirtualMachineManagerImpl]
(Work-Job-Executor-3:ctx-fbf1d659 job-16/job-17 ctx-a4481789) Deploy avoids
pods: null, clusters: [2], hosts: null
```
So, the second router (r-5-VM) avoids cluster 2 (because r-4-VM runs on
host 4 which is part of cluster 2).
Seems perfectly fine: LGTM
Thanks @bvbharat
> Redundant VR getting started in the same cluster or host even when there are
> suitable hosts available.
> ------------------------------------------------------------------------------------------------------
>
> Key: CLOUDSTACK-8851
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-8851
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Reporter: Bharat Kumar
> Assignee: Bharat Kumar
>
> Issue
> =====
> Redundant VR getting started in the same cluster or some time in same host
> CloudStack often starts both of redundant VRs in same Cluster and what is
> worse, Cloudstack sometimes starts both of redundant VRs in same host. if due
> to some reason if this host gets effected both the redundant routes get
> effected.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
