-
Notifications
You must be signed in to change notification settings - Fork 104
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add CYGNUS_CONF_URL to fetch a custom agent config w/o rebuilding docker image #1418
base: master
Are you sure you want to change the base?
Conversation
I'll review this ASAP, thanks! |
@frbattid Another benefit I forgot to mention regarding this, consequence of the fact that rebuilding the image is no longer necessary, is that it makes much easier testing cygnus in a distributed cluster environment (a docker swarm cluster is my use case). |
Hi @AlvaroVega!, any comments on this? I'd like to stop using my custom cygnus images if possible. |
HI, was this PR tested? Does this PR ensure backward compatibility? |
Thanks Alvaro for such a fast reply. That being said, my question is... what kind of testing for this would you like to have?
|
Ping @AlvaroVega |
Regarding with this PR currently you can use CYGNUS_SKIP_CONF_GENERATION env var and others to provide your custom agent configuration. |
Using docker, currently you can provide your own cygnus agent conf in a file in a volumen. |
@taliaga is enough with this possiblity that @AlvaroVega explains? Or do you still that downloading a conf file (using wget) is still needed? In the second case, could you elaborate on the user case behind, please? Thanks! |
Hi there!
I'm new to cygnus, but I'm proposing this little change to make it easier to try different agent configurations without the need to rebuild the image. This way, I can put the agent config in, say, a github gist and try different things faster.
Note that I'm including some COPY in the Dockerfile because when building locally an image, a change in one of those script is actually being ignored. There's a non-trivial flow of variables definition and copies in the huge RUN command above, which ends up using the script in the configured cygnus repo. I couldn't see the need for such indirection... maybe from the time the Dockerfile was not inside the repo?
Anyway, let me know wdyt :)