[
https://issues.apache.org/jira/browse/CLOUDSTACK-2823?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13676840#comment-13676840
]
Wei Zhou commented on CLOUDSTACK-2823:
--------------------------------------
Testing result for new systemvm-kvm:
(1) systemvm can not get message from /dev/vport0p1 on its first run.
(2) systemvm can get messages from /dev/vport0p1 on later run.
I am not sure whether it is a CentOS 6.4 issue, as the kernel version of CentOS
is 2.6.32 and some articles on Internet mentioned that virtio serial was
supported from 2.6.34.
Workaround is sending the message some times, like this:
diff --git
a/plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/LibvirtComputingResource.java
b/plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/LibvirtComputingResource.java
index bab53bc..62bb5eb 100755
---
a/plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/LibvirtComputingResource.java
+++
b/plugins/hypervisors/kvm/src/com/cloud/hypervisor/kvm/resource/LibvirtComputingResource.java
@@ -3374,7 +3374,14 @@ ServerResource {
// pass cmdline info to system vms
if (vmSpec.getType() != VirtualMachine.Type.User) {
- passCmdLine(vmName, vmSpec.getBootArgs() );
+ for (int count = 0; count < 10; count ++) {
+ try {
+ Thread.sleep(5000);
+ } catch (InterruptedException e) {
+ s_logger.trace("Ignoring InterruptedException.", e);
+ }
+ passCmdLine(vmName, vmSpec.getBootArgs());
+ }
}
state = State.Running;
> SystemVMs start fail and have no HA
> -----------------------------------
>
> Key: CLOUDSTACK-2823
> URL: https://issues.apache.org/jira/browse/CLOUDSTACK-2823
> Project: CloudStack
> Issue Type: Bug
> Security Level: Public(Anyone can view this level - this is the
> default.)
> Affects Versions: 4.2.0
> Environment: CentOS 6.4, KVM
> CloudStack master
> Reporter: Wei Zhou
> Assignee: Marcus Sorensen
>
> Host:
> [root@cs-kvm015 ~]# virsh version
> Compiled against library: libvirt 0.10.2
> Using library: libvirt 0.10.2
> Using API: QEMU 0.10.2
> Running hypervisor: QEMU 0.12.1
> Network:
> em0 -> cloudbr0 (Guest, Public), Guest: 172.16.61.0/24, Public: 10.11.11.0/24
> (Can not connect outside)
> em1.103 -> cloudbr1 (Management, Storage) , IP: 192.168.103.*
> Issue descripion
> (1) SystemVMs (SSVM, CPVM) start fails , and have no IP (Public, guest,
> management)
> (2) After I destroyed SystemVM, no new SystemVM will be created automatically.
> This issue only exists on master branch
> Deploy 4.1 successfully, and SystemVms work well.
--
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
