FAQ
How to restart cluster if manager does not come up?
I have 3 node cluster. Recently I added hive and was working with oozie.
For soe reason, cdh manger does not load when I go to http://host:7180
Is there any way I can get back cdh manger ui back?
Regards
-Mac

Search Discussions

  • Mark Schnegelberger at Dec 26, 2012 at 3:19 pm
    Hi Mac,

    It will be useful to establish what has happened so you can address the
    exact cause.

    During configuration of oozie did you stop or impact the Cloudera Manager
    service in some way? On the node where Cloudera Manager is running, what
    does 'service cloudera-scm-server status' show? If stopped, perhaps it's
    somehow been disabled from starting at boot, and should be changed. If
    started, check its logs in /var/log/cloudera-scm-server.log and see if
    there are any errors. Cloudera Manager also depends on a backing database
    (cloudera-scm-server-db by default). Is that backing database running
    correctly?

    Regards,
    --
    *Mark*
    *
    *

    On Sun, Dec 23, 2012 at 2:35 AM, hud wrote:

    How to restart cluster if manager does not come up?
    I have 3 node cluster. Recently I added hive and was working with oozie.
    For soe reason, cdh manger does not load when I go to http://host:7180
    Is there any way I can get back cdh manger ui back?
    Regards
    -Mac
  • Hud at Dec 31, 2012 at 7:21 pm
    Scm manager log shows following error

    2012-12-30 09:06:10,102 I2012-12-30 13:03:34,700 WARN
    [com.mchange.v2.async.ThreadPoolAsynchronousRunner$PoolThread-#0:resourcepool.BasicResourcePool@1841]
    com.mchange.v2.resourcepool.BasicResourcePool$AcquireTask@40e98b7 --
    Acquisition Attempt Failed!!! Clearing pending acquires. While trying to
    acquire a needed new resource, we failed to succeed more than the maximum
    number of allowed acquisition attempts (5). Last acquisition attempt
    exception:
    org.postgresql.util.PSQLException: FATAL: could not access status of
    transaction 0
    Detail: could not write to file "pg_subtrans/002D" at offset 24576: No
    space left on device
    at
    org.postgresql.core.v3.ConnectionFactoryImpl.readStartupMessages(ConnectionFactoryImpl.java:469)
    What happened yesterday was: I started map/r job with 13 maps and it
    stopped in between. I canceled the job from command line. I want to restart
    all the services again but I get exception as:

    A server error has occurred. Send the following information to Cloudera.

    *Path: http://cloudera-cdh-1.etouch.net:7180/cmf/clusters/2/do*
    Cloudera Manager (Free Edition) Version: 4.0.4 (#79 built by jenkins on
    20120821-1040 git: 119fe7452f0c356dcde763a1d22eda9363d24ad4)
    *javax.persistence.RollbackException*:Error while committing the
    transaction
    at *TransactionImpl.java line 93*
    in org.hibernate.ejb.TransactionImpl commit()

    Please, advise
    Thanks
    -Mac

    On Saturday, December 22, 2012 11:35:40 PM UTC-8, hud wrote:

    How to restart cluster if manager does not come up?
    I have 3 node cluster. Recently I added hive and was working with oozie.
    For soe reason, cdh manger does not load when I go to http://host:7180
    Is there any way I can get back cdh manger ui back?
    Regards
    -Mac
  • Adam Smieszny at Jan 1, 2013 at 5:10 pm
    Hi Mac,

    What hardware are you using for your Cloudera Manager server?
    Can you list the space on your devices? I see the following in that message:
    *Detail: could not write to file "pg_subtrans/002D" at offset 24576: No
    space left on device*

    please run a 'df -h' on your cloudera manager host

    Thanks,
    Adam
    On Mon, Dec 31, 2012 at 2:21 PM, hud wrote:

    Detail: could not write to file "pg_subtrans/002D" at offset 24576: No
    space left on device



    --
    Adam Smieszny
    Cloudera | Systems Engineer | http://www.linkedin.com/in/adamsmieszny
    917.830.4156
  • Philip Langdale at Jan 2, 2013 at 9:27 pm
    Hi Mac,

    As Adam suggests, you've probably run out of space on /var, which prevents
    the database from being written to, which will
    prevent the CM server from starting - and probably break many other things
    running on that machine.

    --phil

    On 1 January 2013 09:10, Adam Smieszny wrote:

    Hi Mac,

    What hardware are you using for your Cloudera Manager server?
    Can you list the space on your devices? I see the following in that
    message:
    *Detail: could not write to file "pg_subtrans/002D" at offset 24576: No
    space left on device*

    please run a 'df -h' on your cloudera manager host

    Thanks,
    Adam

    On Mon, Dec 31, 2012 at 2:21 PM, hud wrote:

    Detail: could not write to file "pg_subtrans/002D" at offset 24576: No
    space left on device



    --
    Adam Smieszny
    Cloudera | Systems Engineer | http://www.linkedin.com/in/adamsmieszny
    917.830.4156

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupscm-users @
categorieshadoop
postedDec 23, '12 at 7:35a
activeJan 2, '13 at 9:27p
posts5
users4
websitecloudera.com
irc#hadoop

People

Translate

site design / logo © 2022 Grokbase