Apache2 restarts too often about opsworks-cookbooks

[mrclmrvn]

(y)

[kostia]

👍 for a quick fix. You can forget about continuous deployment otherwise...

[arielpts]

+1 for this! VERY needed.

[heffergm]

There are a number of resources in recipes/default.rb that call 'logdir_existence_and_restart_apache2', none of which should cause a restart unless there are applicable updates (templates, etc). However, the following block exists in the recipe as well, for reasons I can't quite fathom, which ensures that the restart will be called on every run, regardless of whether a restart is otherwise called for or not. Can we dump this block please?

bash 'logdir_existence_and_restart_apache2' do
  action :run
end

[awsrequena]

@heffergm you're right. None of the 'logdir_existence_and_restart_apache2' calls should apply unless there are applicable updates.

Nonetheless, the last block executing the restart needs to be done, because we currently write the apache logs on a different EBS volume and this is mounted per AutoFS. We need to

bash 'logdir_existence_and_restart_apache2' do
  action :run
end

in order to ensure that there are no overlaying and that we're writing in the right file system.

Currently, we have to go this way.

Thanks everybody for your feedback.