search

linkedin / dynamometer

A tool for scale and performance testing of HDFS with a specific focus on the NameNode.
BSD 2-Clause "Simplified" License
131 stars 34 forks source link

Executing test com.linkedin.dynamometer.TestDynamometerInfra stuck #100

Open pingsutw opened 5 years ago

pingsutw commented 5 years ago

I try to build latest dynamometer When it run TestDynamometerInfra, it seems it stuck at this point Dose anything need to be pre-installed for the build

kevin@kevin-pc:~/git/dynamometer(master)$ ./gradlew build Parallel execution is an incubating feature.

Configure project : Building version '0.1.7' (value loaded from 'version.properties' file).

Task :dynamometer-workload:compileJava Note: Some input files use unchecked or unsafe operations. Note: Recompile with -Xlint:unchecked for details.

Task :dynamometer-blockgen:javadoc /home/kevin/git/dynamometer/dynamometer-blockgen/src/main/java/com/linkedin/dynamometer/blockgenerator/XMLParserMapper.java:26: warning - Tag @link: reference not found: org.apache.hadoop.hdfs.tools.offlineImageViewer.OfflineImageViewer 1 warning

Task :dynamometer-workload:javadoc /home/kevin/git/dynamometer/dynamometer-workload/src/main/java/com/linkedin/dynamometer/workloadgenerator/TimedInputFormat.java:29: warning - WorkloadDriver#START_TIMESTAMP_MS (referenced by @value tag) is an unknown reference. 1 warning <============-> 94% EXECUTING [6m 27s] IDLE IDLE :dynamometer-infra:test > 2 tests completed :dynamometer-infra:test > Executing test com.linkedin.dynamometer.TestDynamometerInfra IDLE IDLE

xkrogen commented 5 years ago

Hi @pingsutw , besides some standard Unix utilities like tar, nothing needs to be preinstalled. Can you try running ./gradlew build --info to get more debug output?

Please note that the test can take some time (5+ minutes) because it takes some steps like downloading a Hadoop tarball, unzipping it, localizing that tarball as a resource to DataNodes it instantiates, then starting up an HDFS cluster.

jojochuang commented 5 years ago

I had a similar issue when i ran it on a very full host. Turns out YARN doesn't launch a job if a node is beyond 99% full. So try it on a different machine.

pingsutw commented 5 years ago

@xkrogen @jojochuang Thank you for your replying

Please note that the test can take some time (5+ minutes) because it takes some steps like downloading a Hadoop tarball, unzipping it, localizing that tarball as a resource to DataNodes it instantiates, then starting up an HDFS cluster.

It's weird that sometimes the test ran more than hour, and didn't finish yet sometimes it works

I had a similar issue when i ran it on a very full host. Turns out YARN doesn't launch a job if a node is beyond 99% full. So try it on a different machine.

I have tried on different machine before, it works for me but I just want to find this bug, and fix it