etan39
/
cloudstack
mirror of https://github.com/apache/cloudstack.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

CLOUDSTACK-7182: NPE while trying to deploy VMs in parallel in isolated network 

The following changes are made:
- Check to see if network is implemented changed from 'state == Implementing||Implemented' to 'state == Implemented'.
The earlier check was a hack to prevent the issue described below.
- At the time of implementing network (using implementNetwork() method), if the VR needs to be deployed then
it follows the same path of regular VM deployment. This leads to a nested call to implementNetwork() while
preparing VR nics. This flow creates issues in dealing with network state transitions. The original call
puts network in "Implementing" state and then the nested call again tries to put it into same state resulting
in issues. In order to avoid it, implementNetwork() call for VR is replaced with below code.
This commit is contained in:
Koushik Das 2014-07-25 15:17:35 +05:30
parent ea740aa0e1
commit 47d6a64b31
1 changed files with 14 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

16
engine/orchestration/src/org/apache/cloudstack/engine/orchestration/NetworkOrchestrator.java
View File

@ -925,7 +925,7 @@ public class NetworkOrchestrator extends ManagerBase implements NetworkOrchestra
boolean isNetworkImplemented(NetworkVO network) {
Network.State state = network.getState();
if (state == Network.State.Implemented || state == Network.State.Implementing) {
if (state == Network.State.Implemented) {
return true;
} else if (state == Network.State.Setup) {
DataCenterVO zone = _dcDao.findById(network.getDataCenterId());
@ -1277,7 +1277,19 @@ public class NetworkOrchestrator extends ManagerBase implements NetworkOrchestra
});
for (NicVO nic : nics) {
Pair<NetworkGuru, NetworkVO> implemented = implementNetwork(nic.getNetworkId(), dest, context);
Pair<NetworkGuru, NetworkVO> implemented = null;
if (vmProfile.getVirtualMachine().getType() != Type.DomainRouter) {
implemented = implementNetwork(nic.getNetworkId(), dest, context);
} else {
// At the time of implementing network (using implementNetwork() method), if the VR needs to be deployed then
// it follows the same path of regular VM deployment. This leads to a nested call to implementNetwork() while
// preparing VR nics. This flow creates issues in dealing with network state transitions. The original call
// puts network in "Implementing" state and then the nested call again tries to put it into same state resulting
// in issues. In order to avoid it, implementNetwork() call for VR is replaced with below code.
NetworkVO network = _networksDao.findById(nic.getNetworkId());
NetworkGuru guru = AdapterBase.getAdapterByName(networkGurus, network.getGuruName());
implemented = new Pair<NetworkGuru, NetworkVO>(guru, network);
}
if (implemented == null || implemented.first() == null) {
s_logger.warn("Failed to implement network id=" + nic.getNetworkId() + " as a part of preparing nic id=" + nic.getId());
throw new CloudRuntimeException("Failed to implement network id=" + nic.getNetworkId() + " as a part preparing nic id=" + nic.getId());