avijayanhwx commented on a change in pull request #194: HDDS-2241. Optimize the 
refresh pipeline logic used by KeyManagerImpl…
URL: https://github.com/apache/hadoop-ozone/pull/194#discussion_r347661288
 
 

 ##########
 File path: 
hadoop-ozone/ozone-manager/src/main/java/org/apache/hadoop/ozone/om/KeyManagerImpl.java
 ##########
 @@ -688,6 +663,48 @@ public OmKeyInfo lookupKey(OmKeyArgs args, String 
clientAddress)
     }
   }
 
+  /**
+   * Refresh pipeline info in OM by asking SCM.
+   * @param value OmKeyInfo
+   */
+  @VisibleForTesting
+  protected void refreshPipeline(OmKeyInfo value) {
+    // Refresh container pipeline info from SCM
+    // based on OmKeyArgs.refreshPipeline flag
+    // 1. Client send initial read request OmKeyArgs.refreshPipeline = false
+    // and uses the pipeline cached in OM to access datanode
+    // 2. If succeeded, done.
+    // 3. If failed due to pipeline does not exist or invalid pipeline state
+    //    exception, client should retry lookupKey with
+    //    OmKeyArgs.refreshPipeline = true
+    Map<Long, ContainerWithPipeline> containerWithPipelineMap = new 
HashMap<>();
+    for (OmKeyLocationInfoGroup key : value.getKeyLocationVersions()) {
+      key.getLocationList().forEach(k -> {
+        // TODO: fix Some tests that may not initialize container client
+        // The production should always have containerClient initialized.
+        if (scmClient.getContainerClient() != null) {
+          ContainerWithPipeline cp = containerWithPipelineMap
+              .computeIfAbsent(k.getContainerID(), cId -> {
+                try {
+                  return scmClient.getContainerClient()
+                      .getContainerWithPipeline(cId);
+                } catch (IOException e) {
+                  LOG.error("Unable to update pipeline for container:{}",
+                      k.getContainerID());
+                  return null;
+                }
+              });
+          if (cp == null) {
+            return;
 
 Review comment:
   Refactored the logic to throw exception when the call to SCM fails.

----------------------------------------------------------------
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.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to