aws elastic beanstalk "Request-URI Too Long"

Question

I have a setup running a python flask app on elastic beanstalk. My Issue is that I'm getting this 414 error code. I have added LimitRequestLine 200000 to httpd.conf, and restarting with sudo httpd service restart on the shell of the ec2 instance, but it seems to not do the trick..

This works perfectly for an apache server running on ec2 not on elastic beanstalk. Maybe the load balancer is to blame?

I'd really appreciate any help on this...

another weird thing - if I restart httpd service from the shell on the ec2 instance, the long uri can pass once, and only once - second time I get the 414 again..

Thanks

Answer 1

LimitRequestLine should reside within <VirtualHost> section. Its quite tricky to do it in Elastic Beanstalk since you need to add this line to /etc/httpd/conf.d/wsgi.conf which is autogenerated after both commands and container_commands are run. Following idea from this blog, adding the following to config file under .ebextensions worked:

commands:
  create_post_dir:
    command: "mkdir -p /opt/elasticbeanstalk/hooks/appdeploy/post"
    ignoreErrors: true
files:
  "/opt/elasticbeanstalk/hooks/appdeploy/post/99_adjust_request_limit.sh":
    mode: "000755"
    owner: root
    group: root
    content: |
      #!/usr/bin/env bash
      sed -i.back '/<VirtualHost/aLimitRequestLine 100000' /etc/httpd/conf.d/wsgi.conf
      supervisorctl -c /opt/python/etc/supervisord.conf restart httpd