Simple spark-ec2 launch fails
piec opened this issue · 4 comments
Hi,
I'm trying to setup a simple spark cluster with 1 slave (using branch-2.0)
I use the following commands
export AWS_ACCESS_KEY_ID=...
export AWS_SECRET_ACCESS_KEY=...
./spark-ec2 --region=us-west-1 -k key_name -i ~/.ssh/my.key -s 1 launch es5
I tried several times and I always end of with this log:
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at ip-yyy.us-west-1.compute.internal
************************************************************/
Persistent HDFS installed, won't start by default...
[timing] persistent-hdfs setup: 00h 00m 05s
Setting up mapreduce
Pseudo-terminal will not be allocated because stdin is not a terminal.
RSYNC'ing /root/mapreduce/conf to slaves...
ec2-xxx.us-west-1.compute.amazonaws.com
[timing] mapreduce setup: 00h 00m 01s
Setting up spark-standalone
RSYNC'ing /root/spark/conf to slaves...
ec2-xxx.us-west-1.compute.amazonaws.com
RSYNC'ing /root/spark-ec2 to slaves...
ec2-xxx.us-west-1.compute.amazonaws.com
ec2-xxx.us-west-1.compute.amazonaws.com: no org.apache.spark.deploy.worker.Worker to stop
no org.apache.spark.deploy.master.Master to stop
starting org.apache.spark.deploy.master.Master, logging to /root/spark/logs/spark-root-org.apache.spark.deploy.master.Master-1-ip-yyy.us-west-1.compute.internal.out
ec2-xxx.us-west-1.compute.amazonaws.com: starting org.apache.spark.deploy.worker.Worker, logging to /root/spark/logs/spark-root-org.apache.spark.deploy.worker.Worker-1-ip-zzz.us-west-1.compute.internal.out
[timing] spark-standalone setup: 00h 00m 30s
Setting up rstudio
spark-ec2/setup.sh: line 110: ./rstudio/setup.sh: No such file or directory
[timing] rstudio setup: 00h 00m 01s
Setting up ganglia
RSYNC'ing /etc/ganglia to slaves...
ec2-xxx.us-west-1.compute.amazonaws.com
Shutting down GANGLIA gmond: [FAILED]
Starting GANGLIA gmond: [ OK ]
Shutting down GANGLIA gmond: [FAILED]
Starting GANGLIA gmond: [ OK ]
Connection to ec2-xxx.us-west-1.compute.amazonaws.com closed.
Shutting down GANGLIA gmetad: [FAILED]
Starting GANGLIA gmetad: [ OK ]
Stopping httpd: [FAILED]
Starting httpd: httpd: Syntax error on line 154 of /etc/httpd/conf/httpd.conf: Cannot load /etc/httpd/modules/mod_authz_core.so into server: /etc/httpd/modules/mod_authz_core.so: cannot open shared object file: No such file or directory
[FAILED]
[timing] ganglia setup: 00h 00m 01s
Connection to ec2-xxx-us-west-1.compute.amazonaws.com closed.
Spark standalone cluster started at http://ec2-xxx-us-west-1.compute.amazonaws.com:8080
Ganglia started at http://ec2-xxx-us-west-1.compute.amazonaws.com:5080/ganglia
Done!
I noticed these error parts:
spark-ec2/setup.sh: line 110: ./rstudio/setup.sh: No such file or directory
Shutting down GANGLIA gmond: [FAILED]
Starting GANGLIA gmond: [ OK ]
Shutting down GANGLIA gmond: [FAILED]
Starting GANGLIA gmond: [ OK ]
Starting httpd: httpd: Syntax error on line 154 of /etc/httpd/conf/httpd.conf: Cannot load /etc/httpd/modules/mod_authz_core.so into server: /etc/httpd/modules/mod_authz_core.so: cannot open shared object file: No such file or directory
So lots of errors, is there anything I'm doing wrong?
(also tried branch 1.6 or with eu-west with the same result)
Cheers & thanks for your work
No, those errors are "normal", unfortunately, and are mostly just noise. You should be able to login and Spark should be working fine.
We may need another fix to address the httpd error, like in #9. I don't use Ganglia (which requires httpd, I think), though, so I'm not sure.
Oh really? I terminated the instances so I'm going to re-setup in order to try
It works thanks! I was confused by the error messages and by the time it takes to startup
@piec I'm getting the same errors, but in my case I'm not able to connect to spark dashboard. Were you able to connect to it eventually? (as in does it take time to actually start up?)
-- Edit --
nvm. I may have found the source of the issue (the spark installation fails because it can't find the spark tarball)