FAQ
Hi,

We had Namenode HA configured in CDH. After a Cloudera Manager 4.6 upgrade,
both the namenodes are moving into the standby mode.We tried doing the
manual failover but that did not work.The error thrownin
hadoop-cmf-hdfs1-NAMENODE-hdfs01.log.out was :

Unable to trigger a roll of the active NN
org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.ipc.StandbyException):
Operation category JOURNAL is not supported in state standby


Appreciate any help regarding this.

--
Thanks & Regards
Himanish

Search Discussions

  • Himanish Kushary at Jun 12, 2013 at 6:39 pm
    Initially I got around the issue by disabling "Automatic Failover" for the
    nameservice and restarting the "hdfs" service.Whenever i enabled Automatic
    Failover both the namenode would go to standby mode with the error

    *"Failed to initialize High Availability state in ZooKeeper. This might be
    because a ZNode for this nameservice is already created. Either remove the
    ZNode, or to reuse the ZNode skip this step and simply start the NameNodes
    and Failover Controllers. To retry, use the "Initialize High Availability
    state in ZooKeeper" command available as a Failover Controller action."*

    This gave me the clue.On deleting the ZNode from Zookeeper using the
    zkCli.sh , i was able to fix the issue and enable high availability.

    On Wed, Jun 12, 2013 at 11:01 AM, Himanish Kushary wrote:

    Hi,

    We had Namenode HA configured in CDH. After a Cloudera Manager 4.6
    upgrade, both the namenodes are moving into the standby mode.We tried doing
    the manual failover but that did not work.The error thrownin
    hadoop-cmf-hdfs1-NAMENODE-hdfs01.log.out was :

    Unable to trigger a roll of the active NN
    org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.ipc.StandbyException): Operation category JOURNAL is not supported in state standby


    Appreciate any help regarding this.

    --
    Thanks & Regards
    Himanish


    --
    Thanks & Regards
    Himanish
  • Vinithra Varadharajan at Jun 12, 2013 at 8:18 pm
    Himanish,
    Glad you got this sorted. Would you mind explaining whether this was a
    matter of stale data or unexpected behavior from CM. In other words, did
    you move from a non-CM managed CDH cluster that had CDH to using CM4.6? Or
    did you have some earlier version of CM and then upgrade to CM4.6 and then
    things broke?

    Thanks,
    Vinithra

    On Wed, Jun 12, 2013 at 11:39 AM, Himanish Kushary wrote:

    Initially I got around the issue by disabling "Automatic Failover" for the
    nameservice and restarting the "hdfs" service.Whenever i enabled Automatic
    Failover both the namenode would go to standby mode with the error

    *"Failed to initialize High Availability state in ZooKeeper. This might
    be because a ZNode for this nameservice is already created. Either remove
    the ZNode, or to reuse the ZNode skip this step and simply start the
    NameNodes and Failover Controllers. To retry, use the "Initialize High
    Availability state in ZooKeeper" command available as a Failover Controller
    action."*

    This gave me the clue.On deleting the ZNode from Zookeeper using the
    zkCli.sh , i was able to fix the issue and enable high availability.

    On Wed, Jun 12, 2013 at 11:01 AM, Himanish Kushary wrote:

    Hi,

    We had Namenode HA configured in CDH. After a Cloudera Manager 4.6
    upgrade, both the namenodes are moving into the standby mode.We tried doing
    the manual failover but that did not work.The error thrownin
    hadoop-cmf-hdfs1-NAMENODE-hdfs01.log.out was :


    Unable to trigger a roll of the active NN
    org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.ipc.StandbyException): Operation category JOURNAL is not supported in state standby



    Appreciate any help regarding this.

    --
    Thanks & Regards
    Himanish


    --
    Thanks & Regards
    Himanish
  • Himanish Kushary at Jun 13, 2013 at 9:00 am
    We had a CM 4.5.3 managed environment with CDH 4.2.1.

    On Wed, Jun 12, 2013 at 4:17 PM, Vinithra Varadharajan wrote:

    Himanish,
    Glad you got this sorted. Would you mind explaining whether this was a
    matter of stale data or unexpected behavior from CM. In other words, did
    you move from a non-CM managed CDH cluster that had CDH to using CM4.6? Or
    did you have some earlier version of CM and then upgrade to CM4.6 and then
    things broke?

    Thanks,
    Vinithra

    On Wed, Jun 12, 2013 at 11:39 AM, Himanish Kushary wrote:

    Initially I got around the issue by disabling "Automatic Failover" for
    the nameservice and restarting the "hdfs" service.Whenever i enabled
    Automatic Failover both the namenode would go to standby mode with the
    error

    *"Failed to initialize High Availability state in ZooKeeper. This might
    be because a ZNode for this nameservice is already created. Either remove
    the ZNode, or to reuse the ZNode skip this step and simply start the
    NameNodes and Failover Controllers. To retry, use the "Initialize High
    Availability state in ZooKeeper" command available as a Failover Controller
    action."*

    This gave me the clue.On deleting the ZNode from Zookeeper using the
    zkCli.sh , i was able to fix the issue and enable high availability.

    On Wed, Jun 12, 2013 at 11:01 AM, Himanish Kushary wrote:

    Hi,

    We had Namenode HA configured in CDH. After a Cloudera Manager 4.6
    upgrade, both the namenodes are moving into the standby mode.We tried doing
    the manual failover but that did not work.The error thrownin
    hadoop-cmf-hdfs1-NAMENODE-hdfs01.log.out was :


    Unable to trigger a roll of the active NN
    org.apache.hadoop.ipc.RemoteException(org.apache.hadoop.ipc.StandbyException): Operation category JOURNAL is not supported in state standby



    Appreciate any help regarding this.

    --
    Thanks & Regards
    Himanish


    --
    Thanks & Regards
    Himanish

    --
    Thanks & Regards
    Himanish

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupscm-users @
categorieshadoop
postedJun 12, '13 at 3:01p
activeJun 13, '13 at 9:00a
posts4
users2
websitecloudera.com
irc#hadoop

People

Translate

site design / logo © 2022 Grokbase