mattf

-- invoke with two arguments, the input file , and the output file. -input /bps/gen -output /bps/analytics

-- FYI...
-- If you run into errors, you can see them in
-- ./target/failsafe-reports/TEST-org.bigtop.bigpetstore.integration.BigPetStorePigIT.xml

-- First , we load data in from a file, as tuples.
-- in pig, relations like tables in a relational database
-- so each relation is just a bunch of tuples.
-- in this case csvdata will be a relation,

mattf

create node group
 - master
 - namenode, oozie, resourcemanager, historyserver

create node group
 - worker
 - datanode, nodemanager

create cluster
 - master: 1

mattf

#!/bin/sh

# mv /bin/hostname /bin/hostname.real
# install $0 as /bin/hostname (a+rx)

TARGET="kubelet"

IFACE=eth0

PARENT=$(ps -oucmd= $PPID)

mattf

#include <stdio.h>
#include <string.h>
#include <unistd.h>
#include <stdlib.h>

#include <deque>

#define MAX_LINE_LEN 1024

struct Poller

mattf

How do you install, setup and run Cumin on a head node -

0) yum install qpidd cumin sesame
1) echo cumin | sudo -u qpidd /usr/sbin/saslpasswd2 -f /var/lib/qpidd/qpidd.sasldb -u QPID cumin
2) sed -i 's,# brokers:.*,brokers: cumin/cumin@localhost:5672,' /etc/cumin/cumin.conf
3) (probably) sed -i 's/# host:.*/host: 0.0.0.0/' /etc/cumin/cumin.conf
4) (hopefully) sed -i 's/# persona:.*/persona: grid/' /etc/cumin/cumin.conf
5) service qpidd start
6) service sesame start (or systemctl start sesame.service)
7) service cumin start (follow instructions, return here when done)

mattf

#include <unistd.h>

int
main(int argc, char *argv[])
{
  setpgid(0, 0);

  execvp(argv[1], &(argv[1]));

  return 1;

mattf

FACTER = /usr/libexec/condor/facter.sh
STARTD_CRON_JOBLIST = $(STARTD_CRON_JOBLIST) FACTER
STARTD_CRON_FACTER_EXECUTABLE = $(FACTER)
STARTD_CRON_FACTER_PERIOD = 300

mattf

cmd = memcached.sh
args = -m $$(Memory)

log = memcached.log

kill_sig = SIGTERM

# Want chirp functionality
+WantIOProxy = TRUE

mattf

#!/bin/sh

# NOTE:
#  . Requires condor_q >= 7.5.2, old classads do not
#    have %
#  . When running, jobs show RUN_TIME of their current
#    run, not accumulated, which would require adding
#    in RemoteWallClockTime
#  . See condor_utils/condor_q.cpp:encode_status for
#    JobStatus map

mattf

#!/bin/sh

# Provide three arguments:
#  . instance id to link
#  . path to file with access key id
#  . path to file with secret access key

# TODO:
#  . Get EC2UserData (ec2-describe-instance-attribute --user-data)