devmadhuu commented on code in PR #6178:
URL: https://github.com/apache/ozone/pull/6178#discussion_r1480900070
##########
hadoop-ozone/recon/src/main/java/org/apache/hadoop/ozone/recon/tasks/ContainerSizeCountTask.java:
##########
@@ -132,36 +138,40 @@ private void process(ContainerInfo container,
/**
* The process() function is responsible for updating the counts of
* containers being tracked in a containerSizeCountMap based on the
- * ContainerInfo objects in the list containers.It then iterates through
+ * ContainerInfo objects in the list containers. It then iterates through
* the list of containers and does the following for each container:
*
- * 1) If the container is not present in processedContainers,
- * it is a new container, so it is added to the processedContainers map
- * and the count for its size in the containerSizeCountMap is incremented
- * by 1 using the handlePutKeyEvent() function.
- * 2) If the container is present in processedContainers but its size has
- * been updated to the new size then the count for the old size in the
- * containerSizeCountMap is decremented by 1 using the
- * handleDeleteKeyEvent() function. The count for the new size is then
- * incremented by 1 using the handlePutKeyEvent() function.
- * 3) If the container is not present in containers list, it means the
- * container has been deleted.
- * The remaining containers inside the deletedContainers map are the ones
- * that are not in the cluster and need to be deleted. Finally, the counts in
- * the containerSizeCountMap are written to the database using the
- * writeCountsToDB() function.
+ * 1) If the container's state is not "deleted," it will be processed:
+ * - If the container is not present in processedContainers, it is a new
+ * container. Therefore, it is added to the processedContainers map, and
+ * the count for its size in the containerSizeCountMap is incremented by
+ * 1 using the handlePutKeyEvent() function.
+ * - If the container is present in processedContainers but its size has
+ * been updated to a new size, the count for the old size in the
+ * containerSizeCountMap is decremented by 1 using the
+ * handleDeleteKeyEvent() function. Subsequently, the count for the new
+ * size is incremented by 1 using the handlePutKeyEvent() function.
+ *
+ * 2) If the container's state is "deleted," it is skipped, as deleted
+ * containers are not processed.
+ *
+ * After processing, the remaining containers inside the deletedContainers
map
+ * are those that are not in the cluster and need to be deleted from the
total
+ * size counts. Finally, the counts in the containerSizeCountMap are written
+ * to the database using the writeCountsToDB() function.
*/
public void process(List<ContainerInfo> containers) {
lock.writeLock().lock();
try {
final Map<ContainerSizeCountKey, Long> containerSizeCountMap
= new HashMap<>();
- final Map<ContainerID, Long> deletedContainers
- = new HashMap<>(processedContainers);
-
+ final Map<ContainerID, Long> deletedContainers =
+ new HashMap<>(processedContainers);
// Loop to handle container create and size-update operations
for (ContainerInfo container : containers) {
- // The containers present in the cache hence it is not yet deleted
+ if (container.getState().equals(DELETED)) {
Review Comment:
This will have lesser impact because Recon doesn't update the DELETED state
of newly added containers, while startup only, DELETED containers gets synced
from SCM when Recon takes full SCM DB snapshot.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
