Grokbase › Groups › Hadoop › scm-users › June 2012
FAQ
Greetings, I had a power failure over the weekend which resulted in
one of my hadoop nodes not coming back as the file system on the one
server is corrupt. The rest of the cluster is fine and healthy.
However, I need to rebuild the server and add it back to the pool but
hadoop has it stuck in an unknown state so I cannot stop or delete the
server out. How can I get rid of it from the system since it thinks it
is still running?

Thanks,
Joe

Search Discussions

  • Joe Coleman at Jun 13, 2012 at 2:13 pm ⇧
    I should also note I am on Version: 3.7.3 (built by jenkins on
    20120210-1325 git: 7726f1e28fc509d36ea6ad7549c2749179c56138)
    On Jun 13, 8:53 am, Joe Coleman wrote:
    Greetings, I had a power failure over the weekend which resulted in
    one of my hadoop nodes not coming back as the file system on the one
    server is corrupt. The rest of the cluster is fine and healthy.
    However, I need to rebuild the server and add it back to the pool but
    hadoop has it stuck in an unknown state so I cannot stop or delete the
    server out. How can I get rid of it from the system since it thinks it
    is still running?

    Thanks,
    Joe
  • Philip Zeyliger at Jun 13, 2012 at 2:18 pm ⇧
    Hi Joe,

    We've fixed this in a more recent version. In the meanwhile, there's a
    workaround described at
    https://groups.google.com/a/cloudera.org/forum/?fromgroups#!topic/scm-users/DzlVg8WPQ8I.
    Alternately, you can upgrade your server to the latest version.

    Cheers,

    -- Philip
    On Wed, Jun 13, 2012 at 7:13 AM, Joe Coleman wrote:

    I should also note I am on Version: 3.7.3 (built by jenkins on
    20120210-1325 git: 7726f1e28fc509d36ea6ad7549c2749179c56138)
    On Jun 13, 8:53 am, Joe Coleman wrote:
    Greetings, I had a power failure over the weekend which resulted in
    one of my hadoop nodes not coming back as the file system on the one
    server is corrupt. The rest of the cluster is fine and healthy.
    However, I need to rebuild the server and add it back to the pool but
    hadoop has it stuck in an unknown state so I cannot stop or delete the
    server out. How can I get rid of it from the system since it thinks it
    is still running?

    Thanks,
    Joe
  • Joe Coleman at Jun 13, 2012 at 2:29 pm ⇧
    Thanks. I am having trouble connecting to the postrgres db. I am
    trying to use tuples. Is there a better way to do it, I am new with
    postgres?

    could not connect to server: No such file or directory
    Is the server running locally and accepting
    connections on Unix domain socket "/tmp/.s.PGSQL.5432"?

    Thanks,
    Joe
    On 13 June, 09:17, Philip Zeyliger wrote:
    Hi Joe,

    We've fixed this in a more recent version.  In the meanwhile, there's a
    workaround described athttps://groups.google.com/a/cloudera.org/forum/?fromgroups#!topic/scm....
    Alternately, you can upgrade your server to the latest version.

    Cheers,

    -- Philip






    On Wed, Jun 13, 2012 at 7:13 AM, Joe Coleman wrote:
    I should also note I am on Version: 3.7.3 (built by jenkins on
    20120210-1325 git: 7726f1e28fc509d36ea6ad7549c2749179c56138)
    On Jun 13, 8:53 am, Joe Coleman wrote:
    Greetings, I had a power failure over the weekend which resulted in
    one of my hadoop nodes not coming back as the file system on the one
    server is corrupt. The rest of the cluster is fine and healthy.
    However, I need to rebuild the server and add it back to the pool but
    hadoop has it stuck in an unknown state so I cannot stop or delete the
    server out. How can I get rid of it from the system since it thinks it
    is still running?
    Thanks,
    Joe
  • Philip Zeyliger at Jun 13, 2012 at 2:38 pm ⇧
    I tend to do something like:

    psql -h localhost -U cmf

    You need the "-h localhost" to instruct PostgreSQL to use TCP sockets and
    not unix domain sockets, because that's where it's listening, I think.

    -- Philip
    On Wed, Jun 13, 2012 at 7:29 AM, Joe Coleman wrote:

    Thanks. I am having trouble connecting to the postrgres db. I am
    trying to use tuples. Is there a better way to do it, I am new with
    postgres?

    could not connect to server: No such file or directory
    Is the server running locally and accepting
    connections on Unix domain socket "/tmp/.s.PGSQL.5432"?

    Thanks,
    Joe
    On 13 June, 09:17, Philip Zeyliger wrote:
    Hi Joe,

    We've fixed this in a more recent version. In the meanwhile, there's a
    workaround described athttps://
    groups.google.com/a/cloudera.org/forum/?fromgroups#!topic/scm....
    Alternately, you can upgrade your server to the latest version.

    Cheers,

    -- Philip






    On Wed, Jun 13, 2012 at 7:13 AM, Joe Coleman wrote:
    I should also note I am on Version: 3.7.3 (built by jenkins on
    20120210-1325 git: 7726f1e28fc509d36ea6ad7549c2749179c56138)
    On Jun 13, 8:53 am, Joe Coleman wrote:
    Greetings, I had a power failure over the weekend which resulted in
    one of my hadoop nodes not coming back as the file system on the one
    server is corrupt. The rest of the cluster is fine and healthy.
    However, I need to rebuild the server and add it back to the pool but
    hadoop has it stuck in an unknown state so I cannot stop or delete
    the
    server out. How can I get rid of it from the system since it thinks
    it
    is still running?
    Thanks,
    Joe
  • Joe Coleman at Jun 13, 2012 at 5:36 pm ⇧
    worked like a charm thanks!

    On 13 June, 09:37, Philip Zeyliger wrote:
    I tend to do something like:

    psql -h localhost -U cmf

    You need the "-h localhost" to instruct PostgreSQL to use TCP sockets and
    not unix domain sockets, because that's where it's listening, I think.

    -- Philip






    On Wed, Jun 13, 2012 at 7:29 AM, Joe Coleman wrote:
    Thanks. I am having trouble connecting to the postrgres db. I am
    trying to use tuples. Is there a better way to do it, I am new with
    postgres?
    could not connect to server: No such file or directory
    Is the server running locally and accepting
    connections on Unix domain socket "/tmp/.s.PGSQL.5432"?
    Thanks,
    Joe
    On 13 June, 09:17, Philip Zeyliger wrote:
    Hi Joe,
    We've fixed this in a more recent version.  In the meanwhile, there's a
    workaround described athttps://
    groups.google.com/a/cloudera.org/forum/?fromgroups#!topic/scm....
    Alternately, you can upgrade your server to the latest version.
    Cheers,
    -- Philip
    On Wed, Jun 13, 2012 at 7:13 AM, Joe Coleman <coleman....@gmail.com>
    wrote:
    I should also note I am on Version: 3.7.3 (built by jenkins on
    20120210-1325 git: 7726f1e28fc509d36ea6ad7549c2749179c56138)
    On Jun 13, 8:53 am, Joe Coleman wrote:
    Greetings, I had a power failure over the weekend which resulted in
    one of my hadoop nodes not coming back as the file system on the one
    server is corrupt. The rest of the cluster is fine and healthy.
    However, I need to rebuild the server and add it back to the pool but
    hadoop has it stuck in an unknown state so I cannot stop or delete
    the
    server out. How can I get rid of it from the system since it thinks
    it
    is still running?
    Thanks,
    Joe

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
groupscm-users @
categorieshadoop
postedJun 13, '12 at 1:53p
activeJun 13, '12 at 5:36p
posts6
users2
websitecloudera.com
irc#hadoop

2 users in discussion

Joe Coleman: 4 posts Philip Zeyliger: 2 posts

Content

People

Support

Translate

site design / logo © 2022 Grokbase