You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cloudstack.apache.org by pe...@apache.org on 2026/01/12 16:47:11 UTC
(cloudstack) branch ghi12249-cleanupImplementingNetworks created (now f59ff96998e)
This is an automated email from the ASF dual-hosted git repository.
pearl11594 pushed a change to branch ghi12249-cleanupImplementingNetworks
in repository https://gitbox.apache.org/repos/asf/cloudstack.git
at f59ff96998e Allow cleaning up of networks stuck in Implementing state
This branch includes the following new commits:
new f59ff96998e Allow cleaning up of networks stuck in Implementing state
The 1 revisions listed above as "new" are entirely new to this
repository and will be described in separate emails. The revisions
listed as "add" were already present in the repository and have only
been added to this reference.
(cloudstack) 01/01: Allow cleaning up of networks stuck in Implementing state
Posted by pe...@apache.org.
This is an automated email from the ASF dual-hosted git repository.
pearl11594 pushed a commit to branch ghi12249-cleanupImplementingNetworks
in repository https://gitbox.apache.org/repos/asf/cloudstack.git
commit f59ff96998efb269e1d4fb069f64ea37fb86fac2
Author: Pearl Dsilva <pe...@gmail.com>
AuthorDate: Mon Jan 12 11:46:50 2026 -0500
Allow cleaning up of networks stuck in Implementing state
---
.../framework/jobs/impl/AsyncJobManagerImpl.java | 76 ++++++++++++++++++++++
tools/marvin/setup.py | 2 +-
2 files changed, 77 insertions(+), 1 deletion(-)
diff --git a/framework/jobs/src/main/java/org/apache/cloudstack/framework/jobs/impl/AsyncJobManagerImpl.java b/framework/jobs/src/main/java/org/apache/cloudstack/framework/jobs/impl/AsyncJobManagerImpl.java
index 80140b0d950..d11ecbdc124 100644
--- a/framework/jobs/src/main/java/org/apache/cloudstack/framework/jobs/impl/AsyncJobManagerImpl.java
+++ b/framework/jobs/src/main/java/org/apache/cloudstack/framework/jobs/impl/AsyncJobManagerImpl.java
@@ -969,6 +969,13 @@ public class AsyncJobManagerImpl extends ManagerBase implements AsyncJobManager,
}
logger.trace("End cleanup expired async-jobs");
+
+ // 3) Cleanup orphaned networks stuck in Implementing state without async jobs
+ try {
+ cleanupOrphanedNetworks();
+ } catch (Throwable e) {
+ logger.error("Unexpected exception when trying to cleanup orphaned networks", e);
+ }
} catch (Throwable e) {
logger.error("Unexpected exception when trying to execute queue item, ", e);
}
@@ -1284,6 +1291,75 @@ public class AsyncJobManagerImpl extends ManagerBase implements AsyncJobManager,
}
}
+ /**
+ * Cleanup networks that are stuck in Implementing state without associated async jobs.
+ * Similar to how expired unfinished jobs are cleaned up, this only processes networks
+ * that have been stuck for longer than the job expiration threshold.
+ */
+ private void cleanupOrphanedNetworks() {
+ try {
+ SearchCriteria<NetworkVO> sc = networkDao.createSearchCriteria();
+ sc.addAnd("state", SearchCriteria.Op.EQ, Network.State.Implementing);
+ sc.addAnd("removed", SearchCriteria.Op.NULL);
+ List<NetworkVO> implementingNetworks = networkDao.search(sc, null);
+
+ if (implementingNetworks == null || implementingNetworks.isEmpty()) {
+ return;
+ }
+
+ logger.debug("Found {} networks in Implementing state, checking for orphaned networks", implementingNetworks.size());
+
+ final long expireMinutes = JobExpireMinutes.value();
+ final Date cutoffTime = new Date(System.currentTimeMillis() - (expireMinutes * 60 * 1000));
+
+ for (NetworkVO network : implementingNetworks) {
+ if (network.getCreated().after(cutoffTime)) {
+ logger.trace("Network {} in Implementing state is only {} minutes old (threshold: {} minutes), skipping cleanup",
+ network.getId(),
+ (System.currentTimeMillis() - network.getCreated().getTime()) / 60000,
+ expireMinutes);
+ continue;
+ }
+
+ List<AsyncJobVO> jobs = _jobDao.findInstancePendingAsyncJobs("Network", network.getAccountId());
+ boolean hasActiveJob = false;
+ for (AsyncJobVO job : jobs) {
+ if (job.getInstanceId() != null && job.getInstanceId().equals(network.getId())) {
+ hasActiveJob = true;
+ break;
+ }
+ }
+
+ if (hasActiveJob) {
+ logger.debug("Network {} in Implementing state has active async job, skipping cleanup", network.getId());
+ continue;
+ }
+
+ logger.warn("Found orphaned network {} in Implementing state without async job. " +
+ "Network created: {}, age: {} minutes, expiration threshold: {} minutes. Transitioning to Shutdown state.",
+ network.getId(), network.getCreated(),
+ (System.currentTimeMillis() - network.getCreated().getTime()) / 60000,
+ expireMinutes);
+ updateNetworkState(network);
+
+ }
+ } catch (Exception e) {
+ logger.error("Error while cleaning up orphaned networks", e);
+ }
+ }
+
+ private void updateNetworkState(NetworkVO network) {
+ try {
+ networkOrchestrationService.stateTransitTo(network, Network.Event.OperationFailed);
+ logger.info("Successfully transitioned orphaned network {} to Shutdown state using state machine", network.getId());
+ } catch (final NoTransitionException e) {
+ logger.debug("State transition failed for orphaned network {}, forcing state update", network.getId());
+ network.setState(Network.State.Shutdown);
+ networkDao.update(network.getId(), network);
+ logger.info("Successfully forced orphaned network {} to Shutdown state", network.getId());
+ }
+ }
+
@Override
public void onManagementNodeJoined(List<? extends ManagementServerHost> nodeList, long selfNodeId) {
}
diff --git a/tools/marvin/setup.py b/tools/marvin/setup.py
index f01863d6989..1ceff3dc204 100644
--- a/tools/marvin/setup.py
+++ b/tools/marvin/setup.py
@@ -27,7 +27,7 @@ except ImportError:
raise RuntimeError("python setuptools is required to build Marvin")
-VERSION = "4.20.3.0-SNAPSHOT"
+VERSION = "4.20.3.0"
setup(name="Marvin",
version=VERSION,