I have the following config
0_logdna.config
commands:
01_install_logdna:
command: "/home/ec2-user/logdna.sh"
02_restart_logdna:
command: "service logdna-agent restart"
files:
"/home/ec2-user/logdna.sh" :
mode: "000777"
owner: root
group: root
content: |
#!/bin/sh
RACK_ENV=$(/opt/elasticbeanstalk/bin/get-config environment -k RACK_ENV)
echo "$RACK_ENV"
if [ $RACK_ENV == production ]
then
rpm --import https://repo.logdna.com/logdna.gpg
echo "[logdna]
name=LogDNA packages
baseurl=https://repo.logdna.com/el6/
enabled=1
gpgcheck=1
gpgkey=https://repo.logdna.com/logdna.gpg" | sudo tee /etc/yum.repos.d/logdna.repo
LOGDNA_INGESTION_KEY=$(/opt/elasticbeanstalk/bin/get-config environment -k LOGDNA_INGESTION_KEY)
yum -y install logdna-agent
logdna-agent -k $LOGDNA_INGESTION_KEY # this is your unique Ingestion Key
# /var/log is monitored/added by default (recursively), optionally add more dirs here
logdna-agent -d /var/app/current/log/logstasher.log
logdna-agent -d /var/app/containerfiles/logs/sidekiq.log
# logdna-agent --hostname allows you to pass your AWS env metadata to LogDNA (remove # to uncomment the line below)
# logdna-agent --hostname `{"Ref": "AWSEBEnvironmentName" }`
# logdna -t option allows you to tag the host with tags (remove # to uncomment the line below)
#logdna-agent -t `{"Ref": "AWSEBEnvironmentName" }`
chkconfig logdna-agent on
service logdna-agent start
fi
I want to be able to only run this config for my production environment but each time i run this code, i get an error that says
ERROR [Instance: i-091794aa00f84ab36,i-05b6d0824e7a0f5da] Command failed on instance. Return code: 1 Output: (TRUNCATED)...not found
/home/ec2-user/logdna.sh: line 17: logdna-agent: command not found
/home/ec2-user/logdna.sh: line 18: logdna-agent: command not found
error reading information on service logdna-agent: No such file or directory
logdna-agent: unrecognized service.
Not sure why this is not working. When i echo RACK_ENV i get production
as the value so i know that is correct but why is it failing my if statement and why is it not working properly?
Your use of echo
will lead to malformed /etc/yum.repos.d/logdna.repo
. To set it up properly, please use the following (indentations for EOL2
are important):
files:
"/home/ec2-user/logdna.sh" :
mode: "000777"
owner: root
group: root
content: |
#!/bin/sh
RACK_ENV=$(/opt/elasticbeanstalk/bin/get-config environment -k RACK_ENV)
echo "$RACK_ENV"
if [ $RACK_ENV == production ]
then
rpm --import https://repo.logdna.com/logdna.gpg
cat >/etc/yum.repos.d/logdna.repo << 'EOL2'
[logdna]
name=LogDNA packages
baseurl=https://repo.logdna.com/el6/
enabled=1
gpgcheck=1
gpgkey=https://repo.logdna.com/logdna.gpg
EOL2
LOGDNA_INGESTION_KEY=$(/opt/elasticbeanstalk/bin/get-config environment -k LOGDNA_INGESTION_KEY)
yum -y install logdna-agent
logdna-agent -k $LOGDNA_INGESTION_KEY # this is your unique Ingestion Key
# /var/log is monitored/added by default (recursively), optionally add more dirs here
logdna-agent -d /var/app/current/log/logstasher.log
logdna-agent -d /var/app/containerfiles/logs/sidekiq.log
# logdna-agent --hostname allows you to pass your AWS env metadata to LogDNA (remove # to uncomment the line below)
# logdna-agent --hostname `{"Ref": "AWSEBEnvironmentName" }`
# logdna -t option allows you to tag the host with tags (remove # to uncomment the line below)
#logdna-agent -t `{"Ref": "AWSEBEnvironmentName" }`
chkconfig logdna-agent on
service logdna-agent start
fi
For further troubleshooting please check /var/log/cfn-init-cmd.log
file.