hive-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Rajkumar Singh (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HIVE-20442) Hive stale lock when the hiveserver2 background thread died with NPE
Date Fri, 26 Jul 2019 06:28:00 GMT

     [ https://issues.apache.org/jira/browse/HIVE-20442?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Rajkumar Singh updated HIVE-20442:
----------------------------------
    Attachment: HIVE-20442.01.branch-2.patch
        Status: Patch Available  (was: Open)

> Hive stale lock when the hiveserver2 background thread died with NPE
> --------------------------------------------------------------------
>
>                 Key: HIVE-20442
>                 URL: https://issues.apache.org/jira/browse/HIVE-20442
>             Project: Hive
>          Issue Type: Bug
>          Components: Hive, Transactions
>    Affects Versions: 2.1.1
>         Environment: Hive-2.1
>            Reporter: Rajkumar Singh
>            Assignee: Rajkumar Singh
>            Priority: Major
>         Attachments: HIVE-20442.01.branch-2.patch
>
>
> this look like a race condition where background thread is not able to release the lock
it aquired.
> 1. hiveserver2 background thread request for lock
> {code}
> 2018-08-20T14:13:38,813 INFO  [HiveServer2-Background-Pool: Thread-XXXXX]: lockmgr.DbLockManager
(DbLockManager.java:lock(100)) - Requesting: queryId=hive_xxxxxxx LockRequest(component:[LockComponent(type:SHARED_READ,
level:TABLE, dbname:testdb, tablename:test_table, operationType:SELECT)], txnid:0, user:hive,
hostname:HOSTNAME, agentInfo:hive_xxxxxxx)
> {code}
> 2. acquired the lock and start heartbeating
> {code}
> 2018-08-20T14:36:30,233 INFO  [HiveServer2-Background-Pool: Thread-XXXXX]: lockmgr.DbTxnManager
(DbTxnManager.java:startHeartbeat(517)) - Started heartbeat with delay/interval = 150000/150000
            MILLISECONDS for query: agentInfo:hive_xxxxxxx
> {code}
> 3. during time between event #1 and #2, client disconnected and deleteContext cleanup
the session dir
> {code}
> 2018-08-21T15:39:57,820 INFO  [HiveServer2-Handler-Pool: Thread-XXX]: thrift.ThriftCLIService
(ThriftBinaryCLIService.java:deleteContext(136)) - Session disconnected without closing properly.
> 2018-08-21T15:39:57,820 INFO  [HiveServer2-Handler-Pool: Thread-XXXX]: thrift.ThriftCLIService
(ThriftBinaryCLIService.java:deleteContext(140)) - Closing the session: SessionHandle [3be07faf-5544-4178-8b50-8173002b171a]
> 2018-08-21T15:39:57,820 INFO  [HiveServer2-Handler-Pool: Thread-XXXX]: service.CompositeService
(SessionManager.java:closeSession(363)) - Session closed, SessionHandle [xxxxxxxxxxxxxxxxxxxxxxx],
current sessions:2
> {code}
> 4. background thread died with NPE while trying to get the queryid 
> {code}
> java.lang.NullPointerException: null
>         at org.apache.hadoop.hive.ql.Driver.execute(Driver.java:1568) ~[hive-exec-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hadoop.hive.ql.Driver.runInternal(Driver.java:1414) ~[hive-exec-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hadoop.hive.ql.Driver.run(Driver.java:1211) ~[hive-exec-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hadoop.hive.ql.Driver.run(Driver.java:1204) ~[hive-exec-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hive.service.cli.operation.SQLOperation.runQuery(SQLOperation.java:242)
[hive-service-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hive.service.cli.operation.SQLOperation.access$800(SQLOperation.java:91)
[hive-service-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at org.apache.hive.service.cli.operation.SQLOperation$BackgroundWork$1.run(SQLOperation.java:336)
[hive-service-2.1.0.2.6.5.0-292.jar:2.1.0.2.6.5.0-292]
>         at java.security.AccessController.doPrivileged(Native Method) [?:1.8.0_77]
>         at javax.security.auth.Subject.doAs(Subject.java:422) [?:1.8.0_77]
> {code}
> did not get a chance to release the lock and heartbeater thread continue heartbeat indefinately.



--
This message was sent by Atlassian JIRA
(v7.6.14#76016)

Mime
View raw message