[
https://issues.apache.org/jira/browse/MESOS-4450?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15111025#comment-15111025
]
Andrew Tongen commented on MESOS-4450:
--------------------------------------
Also, just to note, this exact same thing works perfectly on a mesos slave
running version 0.25.0.
> Fetcher incorrectly prepends forward slash to google cloud storage uri
> ----------------------------------------------------------------------
>
> Key: MESOS-4450
> URL: https://issues.apache.org/jira/browse/MESOS-4450
> Project: Mesos
> Issue Type: Bug
> Components: fetcher
> Affects Versions: 0.26.0
> Reporter: Andrew Tongen
>
> Log of failure:
> {quote}
> I0121 17:08:50.458729 421 fetcher.cpp:185] Fetching URI
> 'gs://bucket/spark-1.5.2-bin-hadoop2.6.tgz'
> I0121 17:08:50.700636 421 fetcher.cpp:110] Downloading resource with Hadoop
> client from 'gs://bucket/spark-1.5.2-bin-hadoop2.6' to
> '/var/lib/mesos-slave/slaves/ca28da71-850d-4134-a390-2f007a032ade-S0/frameworks/d38cf30b-f037-487c-bd50-191819a570ea-0000/executors/3/runs/6190780d-e5d5-465a-b463-4e5002bf3de6/spark-1.5.2-bin-hadoop2.6.tgz'
> -copyToLocal: java.net.URISyntaxException: Expected scheme-specific part at
> index 3: gs:
> Usage: hadoop fs [generic options] -copyToLocal [-p] [-ignoreCrc] [-crc]
> <src> ... <localdst>
> E0121 17:08:51.544994 421 shell.hpp:90] Command '/opt/hadoop/bin/hadoop fs
> -copyToLocal '/gs://bucket/spark-1.5.2-bin-hadoop2.6.tgz'
> '/var/lib/mesos-slave/slaves/ca28da71-850d-4134-a390-2f007a032ade-S0/frameworks/d38cf30b-f037-487c-bd50-191819a570ea-0000/executors/3/runs/6190780d-e5d5-465a-b463-4e5002bf3de6/spark-1.5.2-bin-hadoop2.6.tgz''
> failed; this is the output:
> Failed to fetch 'gs://bucket/spark-1.5.2-bin-hadoop2.6.tgz': HDFS copyToLocal
> failed: Failed to execute '/opt/hadoop/bin/hadoop fs -copyToLocal
> '/gs://bucket/spark-1.5.2-bin-hadoop2.6.tgz'
> '/var/lib/mesos-slave/slaves/ca28da71-850d-4134-a390-2f007a032ade-S0/frameworks/d38cf30b-f037-487c-bd50-191819a570ea-0000/executors/3/runs/6190780d-e5d5-465a-b463-4e5002bf3de6/spark-1.5.2-bin-hadoop2.6.tgz'';
> the command was either not found or exited with a non-zero exit status: 255
> Failed to synchronize with slave (it's probably exited)
> {quote}
> You can see in
> https://github.com/apache/mesos/blob/master/src/launcher/fetcher.cpp#L100
> that sourceUri is logged correctly, but then has a forward slash prepended in
> the error message after
> https://github.com/apache/mesos/blob/master/src/launcher/fetcher.cpp#L112
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
