adoroszlai opened a new pull request #3223:
URL: https://github.com/apache/ozone/pull/3223


   ## What changes were proposed in this pull request?
   
   Smoketest for `read-replicas` 
[assumes](https://github.com/apache/ozone/pull/3187#discussion_r825649889) 
`chunkinfo` and `read-replicas` return datanodes in the same order.  This 
turned out to be false, causing intermittent failures.
   
   This PR eliminates this assumption, improving the test to not rely on 
replica order.
   
   https://issues.apache.org/jira/browse/HDDS-6488
   
   ## How was this patch tested?
   
   Temporarily changed `execute_debug_tests` to corrupt replica=2 instead of 
replica=0.
   
   Excerpt from `read-replicas` manifest:
   
   ```
   "replicas": [
     {
       "hostname": "ozone_datanode_1.ozone_default",
       "uuid": "00fab883-5d74-403c-ad54-2d31da07e445"
     },
     {
       "hostname": "ozone_datanode_5.ozone_default",
       "uuid": "ff218756-b571-41e9-9d4b-65f79dee30f0"
     },
     {
       "hostname": "ozone_datanode_3.ozone_default",
       "uuid": "5025a57f-8178-42d5-b18c-5731b71cc678",
       "exception": "java.util.concurrent.ExecutionException: 
org.apache.ratis.thirdparty.io.grpc.StatusRuntimeException: UNAVAILABLE: io 
exception"
     }
   ]
   ```
   
   Order of datanodes in `chunkinfo` output:
   
   ```
   $ cat result/testfile-blocks-10827 | jq -r 
'.KeyLocations[0][]["Datanode-HostName"]'
   ozone_datanode_1.ozone_default
   ozone_datanode_5.ozone_default
   ozone_datanode_3.ozone_default
   ```
   
   Test still passed.
   
   CI with only this test:
   https://github.com/adoroszlai/hadoop-ozone/runs/5649813090#step:5:127
   
   Regular CI:
   https://github.com/adoroszlai/hadoop-ozone/runs/5650040811#step:5:887


-- 
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]

Reply via email to