Hong Chen created HDFS-16578: -------------------------------- Summary: Missing blocks appeared after snn has transitioned to active state Key: HDFS-16578 URL: https://issues.apache.org/jira/browse/HDFS-16578 Project: Hadoop HDFS Issue Type: Bug Components: datanode, namenode Affects Versions: 2.9.2 Reporter: Hong Chen
Scenario: There is no missing blocks at NN1(ann), after NN2 has transitioned to active state by stopping zkfc at NN1 manually. we have found some missing blocks at NN2. {panel:title=/user/xxx/d=2020-01-03/000154_0.lzo file fscklog at NN1} /user/xxx/d=2020-01-03/000154_0.lzo 1555552 bytes, 1 block(s): OK 0. BP-459146894-xxx-1581848181424:{color:#172b4d}blk_1081077638_7337053{color} len=1555552 {color:#172b4d}Live_repl=2{color} DatanodeInfoWithStorage[{color:#172b4d}datanode1{color}:1004,DS-3236bdbc-8af9-4d3a-8bc8-c921b3a8862b,DISK]], [DatanodeInfoWithStorage[{color:#172b4d}datanode2{color}:1004,DS-84b0a3be-5aec-4850-ba71-ed348b94e7c0,DISK] Status: HEALTHY Total size: 1555552 B Total dirs: 0 Total files: 1 Total symlinks: 0 Total blocks (validated): 1 (avg. block size 1555552 B) Minimally replicated blocks: 1 (100.0 %) Over-replicated blocks: 0 (0.0 %) Under-replicated blocks: 0 (0.0 %) Mis-replicated blocks: 0 (0.0 %) Default replication factor: 3 Average block replication: 2.0 Corrupt blocks: 0 Missing replicas: 0 (0.0 %) Number of data-nodes: 2400 Number of racks: 90 FSCK ended at Thu May 12 17:50:37 CST 2022 in 49 milliseconds {panel} then we we check the {color:#172b4d}blk_1081077638_7337053{color} in datanodes log {panel:title=datanode1} 2022-05-10 12:00:42,984 [12699841344] - INFO [BP-459146894-xxx-1581848181424 heartbeating to xxx/xxx:8021:FsDatasetAsyncDiskService@217] - Scheduling {color:#172b4d}blk_1081077638_7337053{color} file /mnt/dfs/9/data/current/BP-459146894-xxx-1581848181424/current/finalized/subdir15/subdir15/blk_1081077638 for deletion 2022-05-10 12:00:44,409 [12699842769] - INFO [Async disk worker #46179 for volume /mnt/dfs/9/data/current:FsDatasetAsyncDiskService$ReplicaFileDeleteTask@321] - Deleted BP-459146894-xxx-1581848181424 blk_1081077638_7337053 file /mnt/dfs/9/data/current/BP-459146894-xxx-1581848181424/current/finalized/subdir15/subdir15/blk_1081077638 {panel} {panel:title=datanode2} 2021-11-29 16:27:07,411 [2765933340] - INFO [BP-459146894-xxx-1581848181424 heartbeating to xxx/xxx:8021:FsDatasetAsyncDiskService@217] - Scheduling {color:#172b4d}blk_1081077638_7337053{color} file /mnt/dfs/5/data/current/BP-459146894-xxx-1581848181424/current/finalized/subdir15/subdir15/blk_1081077638 for deletion 2021-11-29 16:27:08,587 [2765934516] - INFO [Async disk worker #10145 for volume /mnt/dfs/5/data/current:FsDatasetAsyncDiskService$ReplicaFileDeleteTask@321] - Deleted BP-459146894-xxx-1581848181424 blk_1081077638_7337053 file /mnt/dfs/5/data/current/BP-459146894-xxx-1581848181424/current/finalized/subdir15/subdir15/blk_1081077638 {panel} {panel:title=Exception} hadoop fs -get /user/xxx/d=2020-01-03/000154_0.lzo 1555552 . get: Could not obtain block: BP-459146894-xxx-1581848181424:{color:#172b4d}blk_1081077638_7337053{color} file=/user/xxx/d=2020-01-03/000154_0.lzo {panel} -- This message was sent by Atlassian Jira (v8.20.7#820007) --------------------------------------------------------------------- To unsubscribe, e-mail: hdfs-dev-unsubscr...@hadoop.apache.org For additional commands, e-mail: hdfs-dev-h...@hadoop.apache.org