hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Matteo Bertozzi (JIRA)" <j...@apache.org>
Subject [jira] [Comment Edited] (HBASE-14240) Unactionable WARN messages in wal.ProcedureWALFormatReader
Date Tue, 18 Aug 2015 17:33:46 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14240?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14701650#comment-14701650
] 

Matteo Bertozzi edited comment on HBASE-14240 at 8/18/15 5:33 PM:
------------------------------------------------------------------

is the master shutdown clean? i'm not able to understand it from the logs.
in theory on a clean master shutdown the trailer should be written to the wal and the file
closed. which seems not to be happened here. so you get the warning that is basically saying
that that wal is from a non clean shutdown.

{noformat}
----- MASTER RESTARTED (first run with 1.1) ----
2015-08-18 00:26:15,093 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Starting WAL Procedure Store lease recovery
2015-08-18 00:26:15,095 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Log directory not found: File hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs
does not exist.
2015-08-18 00:26:15,107 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Roll new state log: 1
----- MASTER RESTARTED ----
2015-08-18 00:42:59,647 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Starting WAL Procedure Store lease recovery
2015-08-18 00:42:59,676 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Unable to read tracker for hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
- Missing trailer: size=9 startPos=9
2015-08-18 00:42:59,703 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Roll new state log: 2
2015-08-18 00:42:59,705 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Lease acquired for flushLogId: 2
2015-08-18 00:42:59,711 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
nothing left to decode. exiting with missing EOF
2015-08-18 00:42:59,711 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
No active entry found in state log hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log.
removing it
{noformat}


was (Author: mbertozzi):
is the master shutdown clean? i'm not able to understand it from the logs.
in theory on a clean master shutdown the trailer should be written to the wal and the file
closed. which seems not to be happened here. so you get the warning that is basically saying
that that wal is from a non clean shutdown.

{noformat}
2015-08-18 00:26:15,093 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Starting WAL Procedure Store lease recovery
2015-08-18 00:26:15,095 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Log directory not found: File hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs
does not exist.
2015-08-18 00:26:15,107 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Roll new state log: 1
----- MASTER RESTARTED ----
2015-08-18 00:42:59,647 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Starting WAL Procedure Store lease recovery
2015-08-18 00:42:59,676 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Unable to read tracker for hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
- Missing trailer: size=9 startPos=9
2015-08-18 00:42:59,703 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Roll new state log: 2
2015-08-18 00:42:59,705 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Lease acquired for flushLogId: 2
2015-08-18 00:42:59,711 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
nothing left to decode. exiting with missing EOF
2015-08-18 00:42:59,711 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
No active entry found in state log hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log.
removing it
{noformat}

> Unactionable WARN messages in wal.ProcedureWALFormatReader
> ----------------------------------------------------------
>
>                 Key: HBASE-14240
>                 URL: https://issues.apache.org/jira/browse/HBASE-14240
>             Project: HBase
>          Issue Type: Bug
>          Components: master, proc-v2
>    Affects Versions: 1.1.2
>            Reporter: Nick Dimiduk
>            Priority: Minor
>         Attachments: ndimiduk-rc-dist-1.tgz
>
>
> While testing 1.1.2RC0, I notice the following messages in my master log.
> {noformat}
> 2015-08-18 00:42:59,646 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] procedure2.ProcedureExecutor:
Starting procedure executor threads=5
> 2015-08-18 00:42:59,647 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Starting WAL Procedure Store lease recovery
> 2015-08-18 00:42:59,652 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] util.FSHDFSUtils:
Recovering lease on dfs file hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
> 2015-08-18 00:42:59,664 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] util.FSHDFSUtils:
recoverLease=true, attempt=0 on file=hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
after 10ms
> 2015-08-18 00:42:59,665 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Opening state-log: FileStatus{path=hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log;
isDirectory=false; length=9; replication=3; blocksize=134217728; modification_time=1439858084806;
access_time=1439857575101; owner=hbase; group=hdfs; permission=rw-r--r--; isSymlink=false}
> 2015-08-18 00:42:59,676 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Unable to read tracker for hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
- Missing trailer: size=9 startPos=9
> 2015-08-18 00:42:59,703 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Roll new state log: 2
> 2015-08-18 00:42:59,705 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Lease acquired for flushLogId: 2
> 2015-08-18 00:42:59,711 WARN  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
nothing left to decode. exiting with missing EOF
> 2015-08-18 00:42:59,711 INFO  [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.ProcedureWALFormatReader:
No active entry found in state log hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log.
removing it
> 2015-08-18 00:42:59,711 DEBUG [ndimiduk-rc-dist-1:60000.activeMasterManager] wal.WALProcedureStore:
Remove log: hdfs://ndimiduk-rc-dist-1.openstacklocal:8020/apps/hbase/data/MasterProcWALs/state-00000000000000000001.log
> {noformat}
> These log lines are raised immediately after performing an in-place master upgrade from
0.98.0. Master initialized successfully and with no further errors.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message