hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "takeshi.miao (JIRA)" <j...@apache.org>
Subject [jira] [Created] (HBASE-10379) Turn the msg "Request is a replay (34) - PROCESS_TGS" from logging level ERROR to WARN
Date Mon, 20 Jan 2014 02:39:19 GMT
takeshi.miao created HBASE-10379:
------------------------------------

             Summary: Turn the msg "Request is a replay (34) - PROCESS_TGS" from logging level
ERROR to WARN
                 Key: HBASE-10379
                 URL: https://issues.apache.org/jira/browse/HBASE-10379
             Project: HBase
          Issue Type: Improvement
    Affects Versions: 0.94.16
            Reporter: takeshi.miao
            Assignee: takeshi.miao
            Priority: Minor


Hi All,

Recently we got the error msg "Request is a replay (34) - PROCESS_TGS" while we are using
the HBase client API to put data into HBase-0.94.16 with krb5-1.6.1 enabled. The related msg
as follows...
{code}
[2014-01-15 09:40:38,452][hbase-tablepool-1-thread-3][ERROR][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation.doAs(UserGroupInformation.java:1124)):
PriviledgedActionException as:takeshi_miao@LAB cause:javax.security.sasl.SaslException: GSS
initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level: Request
is a replay (34) - PROCESS_TGS)]
[2014-01-15 09:40:38,453][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation.logPriviledgedAction(UserGroupInformation.java:1143)):
PriviledgedAction as:takeshi_miao@LAB from:sun.reflect.NativeMethodAccessorImpl.invoke0(Native
Method)                                                                                  
       
[2014-01-15 09:40:38,453][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.ipc.SecureClient](org.apache.hadoop.hbase.ipc.SecureClient$SecureConnection$1.run(SecureClient.java:213)):
Exception encountered while connecting to the server : javax.security.sasl.SaslException:
GSS initiate failed [Caused by GSSException: No valid credentials provided (Mechanism level:
Request is a replay (34) - PROCESS_TGS)]
[2014-01-15 09:40:38,454][hbase-tablepool-1-thread-3][INFO ][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation.reloginFromTicketCache(UserGroupInformation.java:657)):
Initiating logout for takeshi_miao@LAB
[2014-01-15 09:40:38,454][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation$HadoopLoginModule.logout(UserGroupInformation.java:154)):
hadoop logout
[2014-01-15 09:40:38,454][hbase-tablepool-1-thread-3][INFO ][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation.reloginFromTicketCache(UserGroupInformation.java:667)):
Initiating re-login for takeshi_miao@LAB
[2014-01-15 09:40:38,455][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation$HadoopLoginModule.login(UserGroupInformation.java:146)):
hadoop login
[2014-01-15 09:40:38,456][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation$HadoopLoginModule.commit(UserGroupInformation.java:95)):
hadoop login commit
[2014-01-15 09:40:38,456][hbase-tablepool-1-thread-3][DEBUG][org.apache.hadoop.security.UserGroupInformation](org.apache.hadoop.security.UserGroupInformation$HadoopLoginModule.commit(UserGroupInformation.java:100)):
using existing subject:[takeshi_miao@LAB, UnixPrincipal: takeshi_miao, UnixNumericUserPrincipal:
501, UnixNumericGroupPrincipal [Primary Group]: 501, UnixNumericGroupPrincipal [Supplementary
Group]: 502, takeshi_miao@LAB]
{code}

At the beginning, we were worry about the "data loss" occurring while we found the "Request
is a replay (34) - PROCESS_TGS" (especially it is the ERROR level) in log, but after code
study, this is basically *NOT* a data loss issue due to HBase client API would try 5 times
internally (o.a.h.hbase.ipc.SecureClient, L#296, a one thread) and also 10 times of retry
externally (o.a.h.hbase.client.HConnectionManager, L#1661, for all failed thread), The HTable
API would also throw IOEcxeption to client code if any thread still fail after these retries.

Based on HBase users' viewpoint as us, we think this is better to change the logging level
from 'ERROR' to 'WARN', due to the 'ERROR' level had been confused us for a while...But this
code change may need to change boht HBase code and Hadoop code as well; so I am wondering
how community think about this small thing but may be important to the pure HBase users.

mailing list
http://mail-archives.apache.org/mod_mbox/hbase-user/201401.mbox/%3CCADcMMgGiEyho0HGwgbfOUS78ymDpCo5Q0PStWAPUk40W%3DPfcFQ%40mail.gmail.com%3E




--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Mime
View raw message