FAQ
I'm getting this error in Hue when I click on Check for misconfiguration.


hadoop.hdfs_clusters.default.webhdfs_urlCurrent value: http://<namenode/hueserver>:50070/webhdfs/v1
Failed to create temporary file "/tmp/hue_config_validation.4893624165474554515"

When I look @ the error log I see this:

WebHdfs at http://<namenode/hueserver>:50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



I'm on CDH4 and CentOS 6.


Looking @ COnfig through CDM:

dfs.webhdfs.enabled is set to true


The config params:

hadoop.proxyuser.hue.hosts

hadoop.proxyuser.hue.groups

hadoop.proxyuser.httpfs.hosts

hadoop.proxyuser.httpfs.groups


All have an aterisk [*] ...


When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?

Search Discussions

  • Romain Rigaux at Oct 1, 2012 at 4:55 pm
    It seems that the NameNode could not be contacted (the cause is probably in
    the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner should fail
    too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:


    I'm getting this error in Hue when I click on Check for misconfiguration.


    hadoop.hdfs_clusters.default.webhdfs_urlCurrent value: http://<namenode/hueserver>:50070/webhdfs/v1

    Failed to create temporary file "/tmp/hue_config_validation.4893624165474554515"


    When I look @ the error log I see this:


    WebHdfs at http://<namenode/hueserver>:50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;




    I'm on CDH4 and CentOS 6.



    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true


    The config params:

    hadoop.proxyuser.hue.hosts

    hadoop.proxyuser.hue.groups

    hadoop.proxyuser.httpfs.hosts

    hadoop.proxyuser.httpfs.groups


    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Anson Abraham at Oct 1, 2012 at 5:50 pm
    I don't see any errors in the NN logs. The other services FB, Beeswax and
    JD does fail too. Through CDM I resart the hue and beeswax service, which
    I assume excompasses all of these services as well?
    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is probably
    in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner should
    fail too.

    Are you restarting the Hue service only or all of them?

    Romain

    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham <anson....@gmail.com<javascript:>
    wrote:
    I'm getting this error in Hue when I click on Check for misconfiguration.



    hadoop.hdfs_clusters.default.webhdfs_url
    Current value: http://<namenode/hueserver>:50070/webhdfs/v1


    Failed to create temporary file "/tmp/hue_config_validation.4893624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true



    The config params:

    hadoop.proxyuser.hue.hosts

    hadoop.proxyuser.hue.groups

    hadoop.proxyuser.httpfs.hosts

    hadoop.proxyuser.httpfs.groups


    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Romain Rigaux at Oct 1, 2012 at 11:27 pm
    No, the other services should be 'hdfs', 'mapreduce'. Hue is just a UI on
    top of these services.

    Is your cluster up and running?

    https://ccp.cloudera.com/display/CDH4DOC/Installing+CDH4+on+a+Single+Linux+Node+in+Pseudo-distributed+Mode#InstallingCDH4onaSingleLinuxNodeinPseudo-distributedMode-StartingHadoopandVerifyingitisWorkingProperly
    :

    Romain

    On Mon, Oct 1, 2012 at 10:50 AM, Anson Abraham wrote:

    I don't see any errors in the NN logs. The other services FB, Beeswax and
    JD does fail too. Through CDM I resart the hue and beeswax service, which
    I assume excompasses all of these services as well?

    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is probably
    in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner should
    fail too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:

    I'm getting this error in Hue when I click on Check for misconfiguration.




    hadoop.hdfs_clusters.default.**webhdfs_url


    Current value: http://<namenode/**hueserver>:50070/webhdfs/v1




    Failed to create temporary file "/tmp/hue_config_validation.**4893624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:**50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true







    The config params:

    hadoop.proxyuser.hue.hosts



    hadoop.proxyuser.hue.groups



    hadoop.proxyuser.httpfs.hosts



    hadoop.proxyuser.httpfs.groups




    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Anson Abraham at Oct 5, 2012 at 4:04 pm
    Well yeah, hdfs and m/r are running as well as hive and pig. All the
    services are running w/ no issues. it's just hue(beeswax) where
    sporadically I'll get that error in the logs.
    On Monday, October 1, 2012 7:27:03 PM UTC-4, Romain Rigaux wrote:

    No, the other services should be 'hdfs', 'mapreduce'. Hue is just a UI on
    top of these services.

    Is your cluster up and running?


    https://ccp.cloudera.com/display/CDH4DOC/Installing+CDH4+on+a+Single+Linux+Node+in+Pseudo-distributed+Mode#InstallingCDH4onaSingleLinuxNodeinPseudo-distributedMode-StartingHadoopandVerifyingitisWorkingProperly
    :

    Romain


    On Mon, Oct 1, 2012 at 10:50 AM, Anson Abraham <anson....@gmail.com<javascript:>
    wrote:
    I don't see any errors in the NN logs. The other services FB, Beeswax
    and JD does fail too. Through CDM I resart the hue and beeswax service,
    which I assume excompasses all of these services as well?

    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is probably
    in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner should
    fail too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:

    I'm getting this error in Hue when I click on Check for misconfiguration.





    hadoop.hdfs_clusters.default.**webhdfs_url



    Current value: http://<namenode/**hueserver>:50070/webhdfs/v1





    Failed to create temporary file "/tmp/hue_config_validation.**4893624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:**50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true









    The config params:

    hadoop.proxyuser.hue.hosts




    hadoop.proxyuser.hue.groups




    hadoop.proxyuser.httpfs.hosts




    hadoop.proxyuser.httpfs.groups





    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Romain Rigaux at Oct 5, 2012 at 5:38 pm
    Ok, so you get:


    WebHdfs at http://<namenode/hueserver>:50070/webhdfs/v1 -- Validation
    error: ConnectException: Call From <datanode>/<IP address of
    server/datanode> to 0.0.0.0:8020 failed on connection exception:
    java.net.ConnectException: Connection refused;"


    from time to time and need to restart the Hue service in order to solve the
    problem?

    Do you have more Hue and NN logs?

    Romain
    On Fri, Oct 5, 2012 at 9:04 AM, Anson Abraham wrote:

    Well yeah, hdfs and m/r are running as well as hive and pig. All the
    services are running w/ no issues. it's just hue(beeswax) where
    sporadically I'll get that error in the logs.
    On Monday, October 1, 2012 7:27:03 PM UTC-4, Romain Rigaux wrote:

    No, the other services should be 'hdfs', 'mapreduce'. Hue is just a UI on
    top of these services.

    Is your cluster up and running?

    https://ccp.cloudera.com/**display/CDH4DOC/Installing+**
    CDH4+on+a+Single+Linux+Node+**in+Pseudo-distributed+Mode#**
    InstallingCDH4onaSingleLinuxNo**deinPseudo-distributedMode-**
    StartingHadoopandVerifyingitis**WorkingProperly<https://ccp.cloudera.com/display/CDH4DOC/Installing+CDH4+on+a+Single+Linux+Node+in+Pseudo-distributed+Mode#InstallingCDH4onaSingleLinuxNodeinPseudo-distributedMode-StartingHadoopandVerifyingitisWorkingProperly>
    :

    Romain

    On Mon, Oct 1, 2012 at 10:50 AM, Anson Abraham wrote:

    I don't see any errors in the NN logs. The other services FB, Beeswax
    and JD does fail too. Through CDM I resart the hue and beeswax service,
    which I assume excompasses all of these services as well?

    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is
    probably in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner
    should fail too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:

    I'm getting this error in Hue when I click on Check for misconfiguration.






    hadoop.hdfs_clusters.default.**w**ebhdfs_url




    Current value: http://<namenode/**hueser**ver>:50070/webhdfs/v1






    Failed to create temporary file "/tmp/hue_config_validation.**48**93624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:**50**070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true











    The config params:

    hadoop.proxyuser.hue.hosts





    hadoop.proxyuser.hue.groups





    hadoop.proxyuser.httpfs.hosts





    hadoop.proxyuser.httpfs.groups






    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Anson Abraham at Oct 5, 2012 at 6:07 pm
    I'm not sure if it's related to to restarting the hue service though. B/c
    every now and then, when using Hue, i get an Exception communication issue.
    and looking @ the Hue logs, this what I see. When I restart it, hue
    starts working again. Where as hdfs m/r, hive, etc...they all are still
    running w no restart needed (where m/r jobs still run w/ no issues).
    On Friday, October 5, 2012 1:38:18 PM UTC-4, Romain Rigaux wrote:

    Ok, so you get:

    WebHdfs at http://<namenode/hueserver>:50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;"


    from time to time and need to restart the Hue service in order to solve
    the problem?

    Do you have more Hue and NN logs?

    Romain

    On Fri, Oct 5, 2012 at 9:04 AM, Anson Abraham <anson....@gmail.com<javascript:>
    wrote:
    Well yeah, hdfs and m/r are running as well as hive and pig. All the
    services are running w/ no issues. it's just hue(beeswax) where
    sporadically I'll get that error in the logs.
    On Monday, October 1, 2012 7:27:03 PM UTC-4, Romain Rigaux wrote:

    No, the other services should be 'hdfs', 'mapreduce'. Hue is just a UI
    on top of these services.

    Is your cluster up and running?

    https://ccp.cloudera.com/**display/CDH4DOC/Installing+**
    CDH4+on+a+Single+Linux+Node+**in+Pseudo-distributed+Mode#**
    InstallingCDH4onaSingleLinuxNo**deinPseudo-distributedMode-**
    StartingHadoopandVerifyingitis**WorkingProperly<https://ccp.cloudera.com/display/CDH4DOC/Installing+CDH4+on+a+Single+Linux+Node+in+Pseudo-distributed+Mode#InstallingCDH4onaSingleLinuxNodeinPseudo-distributedMode-StartingHadoopandVerifyingitisWorkingProperly>
    :

    Romain

    On Mon, Oct 1, 2012 at 10:50 AM, Anson Abraham wrote:

    I don't see any errors in the NN logs. The other services FB, Beeswax
    and JD does fail too. Through CDM I resart the hue and beeswax service,
    which I assume excompasses all of these services as well?

    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is
    probably in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner
    should fail too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:

    I'm getting this error in Hue when I click on Check for misconfiguration.







    hadoop.hdfs_clusters.default.**w**ebhdfs_url





    Current value: http://<namenode/**hueser**ver>:50070/webhdfs/v1







    Failed to create temporary file "/tmp/hue_config_validation.**48**93624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:**50**070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true













    The config params:

    hadoop.proxyuser.hue.hosts






    hadoop.proxyuser.hue.groups






    hadoop.proxyuser.httpfs.hosts






    hadoop.proxyuser.httpfs.groups







    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?
  • Romain Rigaux at Oct 5, 2012 at 10:57 pm
    Ok, so next time it happens could we get a full trace of the logs?

    Romain
    On Fri, Oct 5, 2012 at 11:01 AM, Anson Abraham wrote:

    I'm not sure if it's related to to restarting the hue service though. B/c
    every now and then, when using Hue, i get an Exception communication issue.
    and looking @ the Hue logs, this what I see. When I restart it, hue
    starts working again. Where as hdfs m/r, hive, etc...they all are still
    running w no restart needed (where m/r jobs still run w/ no issues).

    On Friday, October 5, 2012 1:38:18 PM UTC-4, Romain Rigaux wrote:

    Ok, so you get:

    WebHdfs at http://<namenode/hueserver>:**50070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;"


    from time to time and need to restart the Hue service in order to solve
    the problem?

    Do you have more Hue and NN logs?

    Romain
    On Fri, Oct 5, 2012 at 9:04 AM, Anson Abraham wrote:

    Well yeah, hdfs and m/r are running as well as hive and pig. All the
    services are running w/ no issues. it's just hue(beeswax) where
    sporadically I'll get that error in the logs.
    On Monday, October 1, 2012 7:27:03 PM UTC-4, Romain Rigaux wrote:

    No, the other services should be 'hdfs', 'mapreduce'. Hue is just a UI
    on top of these services.

    Is your cluster up and running?

    https://ccp.cloudera.com/**displ**ay/CDH4DOC/Installing+**CDH4+on+**
    a+Single+Linux+Node+**in+Pseudo-**distributed+Mode#**InstallingCDH**
    4onaSingleLinuxNo**deinPseudo-**distributedMode-**StartingHadoop**
    andVerifyingitis**WorkingProperl**y<https://ccp.cloudera.com/display/CDH4DOC/Installing+CDH4+on+a+Single+Linux+Node+in+Pseudo-distributed+Mode#InstallingCDH4onaSingleLinuxNodeinPseudo-distributedMode-StartingHadoopandVerifyingitisWorkingProperly>
    :

    Romain

    On Mon, Oct 1, 2012 at 10:50 AM, Anson Abraham wrote:

    I don't see any errors in the NN logs. The other services FB, Beeswax
    and JD does fail too. Through CDM I resart the hue and beeswax service,
    which I assume excompasses all of these services as well?

    On Monday, October 1, 2012 12:55:41 PM UTC-4, Romain Rigaux wrote:

    It seems that the NameNode could not be contacted (the cause is
    probably in the NN logs). If it happens, FileBrowser, Beeswax, JobDesigner
    should fail too.

    Are you restarting the Hue service only or all of them?

    Romain
    On Mon, Oct 1, 2012 at 8:53 AM, Anson Abraham wrote:

    I'm getting this error in Hue when I click on Check for misconfiguration.








    hadoop.hdfs_clusters.default.**w****ebhdfs_url






    Current value: http://<namenode/**hueser****ver>:50070/webhdfs/v1








    Failed to create temporary file "/tmp/hue_config_validation.**48****93624165474554515"

    When I look @ the error log I see this:

    WebHdfs at http://<namenode/hueserver>:**50****070/webhdfs/v1 -- Validation error: ConnectException: Call From <datanode>/<IP address of server/datanode> to 0.0.0.0:8020 failed on connection exception: java.net.ConnectException: Connection refused;



    I'm on CDH4 and CentOS 6.


    Looking @ COnfig through CDM:

    dfs.webhdfs.enabled is set to true















    The config params:

    hadoop.proxyuser.hue.hosts







    hadoop.proxyuser.hue.groups







    hadoop.proxyuser.httpfs.hosts







    hadoop.proxyuser.httpfs.groups








    All have an aterisk [*] ...


    When i restart the hue service this misconfiguration message is gone. And Hue users can resume normally what they want to do. Any ideas as to what could be the issue?

Related Discussions

Discussion Navigation
viewthread | post
Discussion Overview
grouphue-user @
categorieshadoop
postedOct 1, '12 at 3:54p
activeOct 5, '12 at 10:57p
posts8
users2
websitecloudera.com
irc#hadoop

2 users in discussion

Anson Abraham: 4 posts Romain Rigaux: 4 posts

People

Translate

site design / logo © 2022 Grokbase