Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Many jobs based on the same agent template produce many failed deployment #102

Merged
merged 2 commits into from
Feb 14, 2022
Merged

Many jobs based on the same agent template produce many failed deployment #102

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8ab0ba9
#101: Workaround for unexpected NullPointerException
sparsick Feb 14, 2022
2f9ac3c
fix typo
sparsick Feb 14, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
52 changes: 36 additions & 16 deletions src/main/java/com/microsoft/jenkins/containeragents/aci/AciCloud.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,7 +131,7 @@ public Collection<NodeProvisioner.PlannedNode> provision(CloudState cloudState,
return agent;
} catch (Exception e) {
LOGGER.log(Level.WARNING, "AciCloud: Provision agent {0} failed: {1}",
new Object[] {agent.getNodeName(), e.getMessage()});
new Object[] {agent.getNodeName(), e});

agent.terminate();

Expand Down Expand Up @@ -177,8 +177,22 @@ public AciContainerTemplate getFirstTemplate(Label label) {
public void addIpEnv(AciAgent agent) throws Exception {
AzureResourceManager azureResourceManager = getAzureClient();

String ip = azureResourceManager.containerGroups()
.getByResourceGroup(resourceGroup, agent.getNodeName()).ipAddress();
// Workaround for https://github.com/Azure/azure-sdk-for-java/issues/27083
String ip = null;
boolean nullIsThrown;
do {
try {
ip = azureResourceManager.containerGroups()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could we check the ip address for null instead and then retry rather than handling a null pointer?
is it possible the IP hasn't been allocated yet? seems quite weird though

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the NullPointerException is coming from Resource Manager internally

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more. 

java.lang.NullPointerException
        at com.azure.resourcemanager.containerinstance.implementation.ContainerGroupImpl.initializeChildrenFromInner(ContainerGroupImpl.java:217)

.getByResourceGroup(resourceGroup, agent.getNodeName()).ipAddress();
nullIsThrown = false;
} catch (NullPointerException e) {
LOGGER.log(Level.WARNING, "During asking for IP address of Agent {0} NullPointerException is thrown,"
+ "but it is ignored.", agent.getNodeName());
nullIsThrown = true;
final int retryInterval = 5 * 1000;
Thread.sleep(retryInterval);
}
} while (nullIsThrown);

EnvironmentVariablesNodeProperty ipEnv = new EnvironmentVariablesNodeProperty(
new EnvironmentVariablesNodeProperty.Entry("IP", ip)
Expand Down Expand Up @@ -215,19 +229,25 @@ private void waitToOnline(AciAgent agent, int startupTimeout, StopWatch stopWatc
if (computer == null) {
throw new IllegalStateException("Agent node has been deleted");
}
ContainerGroup containerGroup =
azureResourceManager.containerGroups().getByResourceGroup(resourceGroup, agent.getNodeName());

if (containerGroup.containers().containsKey(agent.getNodeName())
&& containerGroup.containers().get(agent.getNodeName()).instanceView().currentState().state()
.equals("Terminated")) {

// there doesn't seem to be anyway to get debug information with the current API version in the SDK
// logs and events just return nothing
// while debugging with the CLI the best way I could find was 'attaching' to the container
// see https://github.com/Azure/azure-libraries-for-java/issues/1379
throw new IllegalStateException("ACI container terminated, see the Azure portal / "
+ "CLI for more information");
try {
ContainerGroup containerGroup =
azureResourceManager.containerGroups().getByResourceGroup(resourceGroup, agent.getNodeName());

if (containerGroup.containers().containsKey(agent.getNodeName())
&& containerGroup.containers().get(agent.getNodeName()).instanceView().currentState().state()
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

could we check the container for null instead and then retry rather than handling a null pointer?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the NullPointerException is coming from Resource Manager internally

.equals("Terminated")) {

// there doesn't seem to be anyway to get debug information with the current API version in the SDK
// logs and events just return nothing
// while debugging with the CLI the best way I could find was 'attaching' to the container
// see https://github.com/Azure/azure-libraries-for-java/issues/1379
throw new IllegalStateException("ACI container terminated, see the Azure portal / "
+ "CLI for more information");
}
} catch (NullPointerException e) {
// workaround for https://github.com/Azure/azure-sdk-for-java/issues/27083
LOGGER.log(Level.WARNING, "Waiting for Agent {0} produces a NullPointerException, "
+ "but it is ignored.", agent.getNodeName());
}

if (computer.isOnline()) {
Expand Down