@@ -91,14 +91,14 @@ memory, 4 Gb for disk).
91
91
The following environment variables may be set to influence the environment
92
92
of the job, and may need adjusting to run on the site:
93
93
<dl >
94
- <dt >PARROT_PATH </dt >
94
+ <dt >PARROT\_PATH </dt >
95
95
<dd >The path of <code >parrot_run</code >. Default is to look
96
96
<code >parrot_run</code > in the <code >PATH</code > environment variable
97
97
where lobster is started and send this version to the worker. If another
98
98
parrot binary is preferred, set this environment variable to the
99
99
corresponding directory.</dd >
100
100
101
- <dt >PARROT_DEBUG_FLAGS </dt >
101
+ <dt >PARROT\_DEBUG\_FLAGS </dt >
102
102
<dd >Which debug flags to use. Default are none. See the <code >parrot_run</code >
103
103
help for more details.</dd >
104
104
</dl >
@@ -129,6 +129,22 @@ above:
129
129
130
130
wget --no-check-certificate https://bitbucket.org/pypa/setuptools/raw/bootstrap/ez_setup.py -O - | python - --user
131
131
132
+ ## Using chirp with hadoop
133
+
134
+ On earth, do something akin to the following commands on earth:
135
+
136
+ cd /var/tmp/
137
+ cp -r /usr/lib/hadoop/ .
138
+ cp /usr/lib64/libhdfs* hadoop/lib/
139
+ env JAVA_HOME=/etc/alternatives/java_sdk/ HADOOP_HOME=$PWD/hadoop \
140
+ /afs/nd.edu/user37/ccl/software/cctools-autobuild/bin/chirp_server \
141
+ --root=hdfs://ndcms.crc.nd.edu:19000/
142
+
143
+ and note the port chirp is running on. Then add the follow line to your
144
+ lobster configuration and you should be all set:
145
+
146
+ stageout server: "earth.crc.nd.edu:<your_port>"
147
+
132
148
## Running opportunistically
133
149
134
150
The CRC login nodes ` opteron ` , ` newcell ` , and ` crcfe01 ` are connected to
0 commit comments