hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Purtell <apurt...@apache.org>
Subject Re: Occasional GSSException that brings down region server
Date Tue, 11 Mar 2014 23:37:44 GMT
If you might have more than one security enabled Java process running under
the same UNIX user on the servers, then this and other weirdness can happen
because they share the same ticket cache file in /tmp. Someone does a
(re)login and another process concurrently accessing the cache gets garbage
or unexpected state.


On Mon, Mar 10, 2014 at 2:26 PM, Wei Tan <wtan@us.ibm.com> wrote:

> Hi,
>
>   We are running a HBase cluster in these settings and with kerberos
> enabled.
> HBase: 0.96.1.1
> Zookeeper: 3.4.5
> Hadoop: 1.1.1
>
>
> We constantly put data into HBase and every several hours we get the error
> below on a random region server; this error arises and the region server
> kills itself.
>
> ERROR:
> 2014-02-28 09:32:39,755 ERROR [hconnection-0x116987ad-shared--pool1378-t9]
> security.UserGroupInformation: PriviledgedActionException
> as:XXXXXXXX@DOMAIN cause:javax.security.sasl.SaslException: GSS initiate
> failed [Caused by GSSException: No valid credentials provided (Mechanism
> level: The ticket isn't for us (35) - BAD TGS SERVER NAME)]
>
>
>
> We also tried with multiple version of kdc - all the way up to latest
> 1.12.1 - still see this error. What is weird is that most put gets
> processed successfully until this error occurs and kills the RS.
>
> Thanks,
> Wei
> ---------------------------------
> Wei Tan, PhD
> Research Staff Member
> IBM T. J. Watson Research Center
> http://researcher.ibm.com/person/us-wtan




-- 
Best regards,

   - Andy

Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message