Try not building as root - that can throw things off. A.
On Mon, Aug 5, 2013 at 5:44 PM, a b <autohan...@yahoo.com> wrote: > as for the java 7 problem - i found this suggestion: > > > http://stackoverflow.com/questions/13225149/how-to-install-jdk-7-on-ec2-cluster-via-whirr > > i tried to download whirr - as suggested here: > > https://cwiki.apache.org/confluence/display/WHIRR/How+To+Contribute > > o i did: git clone ... > o i modified: core/src/main/resources/functions/... > o i did: mvn eclipse:eclipse ... > o i skipped: eclipse import > o i ran: mvn install > > it fails in the "mvn install" during test > > Tests in error: > > testActionIsExecutedOnAllRelevantNodes(org.apache.whirr.actions.StopServicesActionTest): > cluster-user != root or do not run as root > > testFilterScriptExecutionByRole(org.apache.whirr.actions.StopServicesActionTest): > cluster-user != root or do not run as root > > testFilterScriptExecutionByInstanceId(org.apache.whirr.actions.StopServicesActionTest): > cluster-user != root or do not run as root > > testFilterScriptExecutionByRoleAndInstanceId(org.apache.whirr.actions.StopServicesActionTest): > cluster-user != root or do not run as root > > testNoScriptExecutionsForNoop(org.apache.whirr.actions.StopServicesActionTest): > cluster-user != root or do not run as root > [..] > > testExecuteOnlyBootstrapForNoop(org.apache.whirr.service.DryRunModuleTest): > cluster-user != root or do not run as root > > testExecuteOnlyBootstrapForNoopWithListener(org.apache.whirr.service.DryRunModuleTest): > cluster-user != root or do not run as root > > testNoInitScriptsAfterConfigurationStartedAndNoConfigScriptsAfterDestroy(org.apache.whirr.service.DryRunModuleTest): > cluster-user != root or do not run as root > testOverrides(org.apache.whirr.command.AbstractClusterCommandTest): > cluster-user != root or do not run as root > > Tests run: 99, Failures: 0, Errors: 49, Skipped: 0 > > [INFO] > ------------------------------------------------------------------------ > [INFO] Reactor Summary: > [INFO] > [INFO] Apache Whirr Build Tools .......................... SUCCESS [2.612s] > [INFO] Whirr ............................................. SUCCESS [4.488s] > [INFO] Apache Whirr Core ................................. FAILURE > [17.113s] > [INFO] Apache Whirr Cassandra ............................ SKIPPED > [INFO] Apache Whirr Ganglia .............................. SKIPPED > [INFO] Apache Whirr Hadoop ............................... SKIPPED > > is there a better way to try this suggestion? > > ------------------------------ > *From:* Andrew Bayer <andrew.ba...@gmail.com> > > *To:* a b <autohan...@yahoo.com> > *Cc:* "user@whirr.apache.org" <user@whirr.apache.org> > *Sent:* Monday, August 5, 2013 4:48 PM > > *Subject:* Re: cannot find hadoop-1.2.1.tar.gz.md5 - file suffix is mds > > Looks like it's the Oracle JDK7 download that's failing - > http://download.oracle.com/otn-pub/java/jdk/7/jdk-7-linux-x64.tar.gzisn't > actually there. I don't know if they actually have a consistent link > to get the JDK7 tarball regardless of version. I'd just use OpenJDK for > now, if you can. > > A. > > On Mon, Aug 5, 2013 at 4:33 PM, a b <autohan...@yahoo.com> wrote: > > Get:19 http://us-east-1.ec2.archive.ubuntu.com precise/main amd64 > Packages [1,273 kB]^M > Get:20 http://us-east-1.ec2.archive.ubuntu.com precise/universe amd64 > Packages [4,786 kB]^M > Get:21 http://us-east-1.ec2.archive.ubuntu.com precise/main i386 Packages > [1,274 kB]^M > Get:22 http://us-east-1.ec2.archive.ubuntu.com precise/universe i386 > Packages [4,796 kB]^M > Get:23 http://us-east-1.ec2.archive.ubuntu.com precise/main > TranslationIndex [3,706 B]^M > Get:24 http://us-east-1.ec2.archive.ubuntu.com precise/universe > TranslationIndex [2,922 B]^M > Get:25 http://us-east-1.ec2.archive.ubuntu.com precise-updates/main > Sources [412 kB]^M > Get:26 http://us-east-1.ec2.archive.ubuntu.com precise-updates/universe > Sources [93.1 kB]^M > Get:27 http://us-east-1.ec2.archive.ubuntu.com precise-updates/main amd64 > Packages [672 kB]^M > Get:28 http://us-east-1.ec2.archive.ubuntu.com precise-updates/universe > amd64 Packages [210 kB]^M > Get:29 http://us-east-1.ec2.archive.ubuntu.com precise-updates/main i386 > Packages [692 kB]^M > Get:30 http://us-east-1.ec2.archive.ubuntu.com precise-updates/universe > i386 Packages [214 kB]^M > Get:31 http://us-east-1.ec2.archive.ubuntu.com precise-updates/main > TranslationIndex [3,564 B]^M > Get:32 http://us-east-1.ec2.archive.ubuntu.com precise-updates/universe > TranslationIndex [2,850 B]^M > Get:33 http://us-east-1.ec2.archive.ubuntu.com precise/main > Translation-en [726 kB]^M > Get:34 http://us-east-1.ec2.archive.ubuntu.com precise/universe > Translation-en [3,341 kB]^M > Get:35 http://us-east-1.ec2.archive.ubuntu.com precise-updates/main > Translation-en [298 kB]^M > Get:36 http://us-east-1.ec2.archive.ubuntu.com precise-updates/universe > Translation-en [123 kB]^M > Fetched 26.1 MB in 21s (1,241 kB/s)^M > Reading package lists...^M > Could not download > http://apache.osuosl.org/hadoop/common/hadoop-1.2.1/hadoop-1.2.1.tar.gz.md5. > Continuing.^M > , error=^M > gzip: stdin: not in gzip format^M > tar: Child returned status 1^M > tar: Error is not recoverable: exiting now^M > mv: cannot stat `jdk1*': No such file or directory^M > update-alternatives: error: alternative path > /usr/lib/jvm/java-7-oracle/bin/java doesn't exist.^M > > ------------------------------ > *From:* Andrew Bayer <andrew.ba...@gmail.com> > *To:* user@whirr.apache.org; a b <autohan...@yahoo.com> > *Sent:* Monday, August 5, 2013 4:22 PM > > *Subject:* Re: cannot find hadoop-1.2.1.tar.gz.md5 - file suffix is mds > > Ok, Whirr does try to download the md5 file, but if it fails to find it, > that's not a blocking error - it'll keep going anyway. What's after that in > the logs? > > A. > > On Mon, Aug 5, 2013 at 4:19 PM, a b <autohan...@yahoo.com> wrote: > > ok - i'm not sure what you are asking. > > *whirr launch-cluster --config ~/whirr/recipes/hadoop.properties* > > where these are the properties i think i changed or added from the > original recipe: > > *whirr.cluster-name=hadoop-ec2* > *whirr.instance-templates=1 hadoop-namenode+hadoop-jobtracker,1 > hadoop-datanode+hadoop-tasktracker* > *whirr.hardware-id=t1.micro* > ***whirr.image-id=us-east-1/ami-25d9a94c* > ***whirr.hadoop.version=1.2.1* > ***whirr.provider=aws-ec2*** > *whirr.identity=${env:AWS_ACCESS_KEY}* > *whirr.credential=${env:AWS_SECRET_KEY}* > *whirr.location-id=us-east-1* > *whirr.java.install-function=install_oracle_jdk7* > > ------------------------------ > *From:* Andrew Bayer <andrew.ba...@gmail.com> > *To:* user@whirr.apache.org; a b <autohan...@yahoo.com> > *Sent:* Monday, August 5, 2013 4:11 PM > *Subject:* Re: cannot find hadoop-1.2.1.tar.gz.md5 - file suffix is mds > > Ok, it looks like they've actually been doing .mds for a while. Where are > you seeing this error? > > A. > > On Mon, Aug 5, 2013 at 4:09 PM, Andrew Bayer <andrew.ba...@gmail.com>wrote: > > That looks like they broke the hadoop-1.2.1 release - the file should be > .md5. I'd bug the Hadoop project about that. > > A. > > > On Mon, Aug 5, 2013 at 3:42 PM, a b <autohan...@yahoo.com> wrote: > > i get a whirr error: > > Could not download > http://apache.osuosl.org/hadoop/common/hadoop-1.2.1/hadoop-1.2.1.tar.gz. > md5 > > when i browse over to: > http://apache.osuosl.org/hadoop/common/hadoop-1.2.1/ > i can see the file is named: hadoop-1.2.1.tar.gz.mds > > how do i tell whirr to use a different suffix? > > > > > > > > > > > >
