Support Questions

Find answers, ask questions, and share your expertise
Announcements
Celebrating as our community reaches 100,000 members! Thank you!

CDH installation failed on all machines except cloudera manager host

avatar
Champion Alumni

I am installing cdh thorugh cloudera manager UI  using parcels.The installation was succesful in  only in the cloudera manager host.It failed on all other machines in the host.It appears that the machines are trying to connect to the cloudera repo in internet but I don not undertand why.I have given the custom parcel repo link but other hosts do  not recognize it I guess.There is absloutely no errior in the /var/log/cloudera-scm-server/cloudera-scm-server.log.Can some one help?.\

 

 

From logs I can see that authentication from cm host to other mahcines were successful.This is the error message that I see in CM GUI.

Do not unserstand why it is going to internet when I have local repo configured.

 

BEGIN yum makecache 
Loaded plugins: rhnplugin 
This system is receiving updates from RHN Classic or RHN Satellite. 
http://archive.cloudera.com/search/redhat/6/x86_64/search/1.2.0/repodata/repomd.xml: [Errno 12] Timeout on http://archive.cloudera.com/search/redhat/6/x86_64/search/1.2.0/repodata/repomd.xml: (28, 'connect() timed out!') 
Trying other mirror. 
Error: Cannot retrieve repository metadata (repomd.xml) for repository: cdh-solr. Please verify its path and try again
END (1)

1 ACCEPTED SOLUTION

avatar
Champion Alumni

figured this out.These machine where subscribed to statellite yum repo I had configured earlier.Even though I had the new local yum repos these machines where still pointing to the old ones.I cleared the cache and it worked.Thanks all

View solution in original post

5 REPLIES 5

avatar
Champion Alumni

I found an exceptiion and I think I know the problem. I have entered an invalid url  in the add parecel repository section.How can i overwrite this.I tried adding the correct location but it is not getting reflected.Is there some fiel which I can modify.

 

Error while attempting to retrieve repository info for repo kkl
java.net.MalformedURLException: no protocol: kkl/
at java.net.URL.<init>(URL.java:585)
at java.net.URL.<init>(URL.java:482)
at java.net.URL.<init>(URL.java:431)
at com.cloudera.parcel.components.ParcelDownloaderImpl.urljoin(ParcelDownloaderImpl.java:390)
at com.cloudera.parcel.components.ParcelDownloaderImpl.getRepositoryInfoFuture(ParcelDownloaderImpl.java:443)
at com.cloudera.parcel.components.ParcelDownloaderImpl.getRepositoryInfo(ParcelDownloaderImpl.java:404)
at com.cloudera.parcel.components.ParcelDownloaderImpl.syncRemoteRepos(ParcelDownloaderImpl.java:303)
at com.cloudera.parcel.components.ParcelManagerImpl.syncRemoteRepo(ParcelManagerImpl.java:585)
at com.cloudera.parcel.components.ParcelManagerImpl.access$300(ParcelManagerImpl.java:62)
at com.cloudera.parcel.components.ParcelManagerImpl$3.run(ParcelManagerImpl.java:596)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:744)

avatar

Hi Nishan,

 

Get out of the wizard, either by opening a new window or clicking on the Cloudera Manager logo in the upper right. Go to Administration -> Settings, click on Parcels on the left, and click the minus ( - ) sign next to the bad URL to remove it. Then click Save in the upper right.

 

Then try the Add Cluster wizard again from the home page, in the dropdown near the word Status on the left.

 

Thanks,

Darren

avatar
Champion Alumni

I corrected this in the parcels page in admin tab.My local repository has all parcels and my internal remote repo has also correct parcels.All permissions are correct.I do not see any errors in the cloudera-scm-server.log.Cloudera-scm-server.out is empty.

 

Here is the last line from the logs;

 

2014-03-20 17:35:12,930 INFO [NodeConfiguratorThread-0-2:node.NodeConfiguratorProgress@477] xx.yy.01.net: New state is a backward state. Storing failed state
2014-03-20 17:35:12,931 INFO [NodeConfiguratorThread-0-2:node.NodeConfiguratorProgress@512] xx.yy.02.net: Transitioning from PACKAGE_INSTALL jdk (PT30.011S) to WAITING_FOR_ROLLBACK
2014-03-20 17:35:13,992 INFO [NodeConfiguratorThread-0-3:node.NodeConfiguratorProgress@477] xx.yy.02.net: New state is a backward state. Storing failed state
2014-03-20 17:35:13,993 INFO [NodeConfiguratorThread-0-3:node.NodeConfiguratorProgress@512] xx.yy.zz.net: Transitioning from PACKAGE_INSTALL jdk (PT31.013S) to WAITING_FOR_ROLLBACK
2014-03-20 17:35:14,021 INFO [NodeConfiguratorThread-0-1:node.NodeConfiguratorProgress@477] xx.yy.zz.net: New state is a backward state. Storing failed state
2014-03-20 17:35:14,021 INFO [NodeConfiguratorThread-0-1:node.NodeConfiguratorProgress@512] xx.yy.04.net: Transitioning from PACKAGE_INSTALL jdk (PT31.009S) to WAITING_FOR_ROLLBACK
2014-03-20 17:35:14,031 INFO [NodeConfiguratorThread-0-4:node.NodeConfiguratorProgress@477] xx.yy.05.net: New state is a backward state. Storing failed state
2014-03-20 17:35:14,032 INFO [NodeConfiguratorThread-0-4:node.NodeConfiguratorProgress@512] xx.yy.zz.net: Transitioning from PACKAGE_INSTALL jdk (PT31.009S) to WAITING_FOR_ROLLBACK
2014-03-20 17:35:14,036 INFO [NodeConfiguratorThread-0-0:node.NodeConfiguratorProgress@477] xx.yy.zz.net: New state is a backward state. Storing failed state
2014-03-20 17:35:14,036 INFO [NodeConfiguratorThread-0-0:node.NodeConfiguratorProgress@512] xx.yy.zz.net: Transitioning from PACKAGE_INSTALL jdk (PT31.010S) to WAITING_FOR_ROLLBACK
2014-03-20 17:35:55,720 INFO [Parcel Update Service:parcel.ParcelUpdateService@59] Updating parcel state based on local filesystem state and remote repos
2014-03-20 17:35:55,725 INFO [Parcel Update Service:components.LocalParcelManagerImpl@105] Found files CDH-4.6.0-1.cdh4.6.0.p0.26-el6.parcel,SOLR-1.2.0-1.cdh4.5.0.p0.4-el6.parcel under /opt/cloudera/parcel-repo

 

avatar
Champion Alumni

When I  Click on  Parcels tab in Administration it Says 'NO clusters found'.I am just in the first step of my  installation.

avatar
Champion Alumni

figured this out.These machine where subscribed to statellite yum repo I had configured earlier.Even though I had the new local yum repos these machines where still pointing to the old ones.I cleared the cache and it worked.Thanks all