[
https://issues.apache.org/jira/browse/CLOUDSTACK-3904?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13722753#comment-13722753
]
Min Chen commented on CLOUDSTACK-3904:
--------------------------------------
After investigation, this is not a list api issue. It seems like a regression
introduced by this storage motion commit
21ce3befc8ea9e1a6de449a21499a50ff141a183. On VMware 5.0, Storage motion is
supported in our DB, so based on this piece of code:
if (canMigrateWithStorage) {
allHostsPair = searchForServers(startIndex, pageSize, null,
hostType, null, srcHost.getDataCenterId(), null,
null, null, null, null, null, srcHost.getHypervisorType(),
srcHost.getHypervisorVersion());
allHosts = allHostsPair.first();
allHosts.remove(srcHost);
// Check if the host has storage pools for all the volumes of the
vm to be migrated.
for (Iterator<HostVO> iterator = allHosts.iterator();
iterator.hasNext();) {
Host host = iterator.next();
Map<Volume, List<StoragePool>> volumePools =
findSuitablePoolsForVolumes(vmProfile, host);
if (volumePools.isEmpty()) {
iterator.remove();
} else {
if (!host.getClusterId().equals(srcHost.getClusterId()) ||
usesLocal) {
requiresStorageMotion.put(host, true);
}
}
}
plan = new DataCenterDeployment(srcHost.getDataCenterId(), null,
null, null, null, null);
We will create a Deployment plan with null podId and null clusterId, which
caused this NPE exception in listHostsCmd in invoking
FirstFitAllocator.reorderHostsByNumberOfVms.
> listHosts API fails in VMware setup when virtualmachineid parameter is passed
> -----------------------------------------------------------------------------
>
> Key: CLOUDSTACK-3904
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-3904
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Components: Automation, VMware
> Affects Versions: 4.2.0
> Environment: VMware setup
> Reporter: Gaurav Aradhye
> Assignee: Alena Prokharchyk
> Priority: Critical
>
> listHosts API failing in VMware setup (observed through test cases and also
> manually hitting the API through URL) when virtualmachineid parameter is
> passed.
> ErrorText: SELECT pod.id, SUM(IF(vm.state='Running' AND vm.account_id = ?, 1,
> 0)) FROM `cloud`.`host_pod_ref` pod LEFT JOIN `cloud`.`vm_instance` vm ON
> pod.id = vm.pod_id WHERE pod.data_center_id = ? GROUP BY pod.id ORDER BY 2
> ASC
> Steps to reproduce:
> Hit listHosts URL throgh url. Suppose setup is hosted at ip x and port y.
> Create an instance. suppose virtualmachineid of instance is z.
> Hit URL:
> http://x:y/client/api?command=listHosts&&virtualmachineid=z
--
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
