You are viewing a plain text version of this content. The canonical link for it is here.
Posted to hdfs-issues@hadoop.apache.org by "Plamen Jeliazkov (JIRA)" <ji...@apache.org> on 2018/12/05 18:27:00 UTC
[jira] [Commented] (HDFS-14059) Test reads from standby on a secure
cluster with Configured failover
[ https://issues.apache.org/jira/browse/HDFS-14059?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16710454#comment-16710454 ]
Plamen Jeliazkov commented on HDFS-14059:
-----------------------------------------
Apologies for the delay on this -- I ran into some difficulty in setup -- anyway, here is my testing experience:
I tested in GCP with 6 nodes, all CentOS 7, 3 NameNodes (active + standby + obs), 3 DNs/NMs, in secure mode with 1 RM, 1 ZK, and (initially) ZKFCs on each NN that I eventually terminated.
I tested initially as the hdfs user and then switched over to my own user and principal once things were stable:
I ran the following simple scenarios:
TestDFSIO - writing 3 x 5GB files. With a manual failover between the active and standby.
TestDFSIO - appending 3 x 1GB files. With a manual failover between the active and standby.
TestDFSIO - reading 3 x 5GB files. With 2 standby NNs and only the observer active during the read mappers -- it completed and went to reduce phase and needed active NN to complete.
TeraGen/TeraSort/TeraValidate - no failover. During TeraValidate phase I ran with 2 observers and no active.
So from my end things look good as far as ConfiguredFailoverProxy is concerned. I never experienced anything concerning around DelegationTokens and could confirm seeing them being generated and used by jobs.
> Test reads from standby on a secure cluster with Configured failover
> --------------------------------------------------------------------
>
> Key: HDFS-14059
> URL: https://issues.apache.org/jira/browse/HDFS-14059
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: test
> Reporter: Konstantin Shvachko
> Assignee: Plamen Jeliazkov
> Priority: Major
>
> Run standard HDFS tests to verify reading from ObserverNode on a secure HA cluster with {{ConfiguredFailoverProxyProvider}}.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: hdfs-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: hdfs-issues-help@hadoop.apache.org