Skip to content
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

transition to "primary/replica" instead of "master/slave" #1

Open
clonm opened this issue Jan 23, 2019 · 6 comments
Open

transition to "primary/replica" instead of "master/slave" #1

clonm opened this issue Jan 23, 2019 · 6 comments

Comments

@clonm
Copy link
Member

clonm commented Jan 23, 2019

https://en.wikipedia.org/wiki/Master/slave_(technology)#Terminology_concerns
As of December 2017, this alternative terminology is an accepted open standard for BIND. But, updating will require some care to make sure we don't break our nameservers.

I changed the naming things that I'm sure are trivial:
cloyne/docker-powerdns-secondary@265924e
cloyne/servers@739245e
https://cloud.docker.com/u/cloyne/repository/docker/cloyne/powerdns-secondary

But, I'm not sure if it's safe to change the actual config files. In addition, the version of PowerDNS we use is no longer supported. Since the most recent version still doesn't have an alternative for "slave" in the options and logging seems to be broken anyway, it might make sense to look into an alternative instead of upgrading to 4.x.

@mitar
Copy link
Member

mitar commented Jan 23, 2019

I do not think we use bind?

@clonm
Copy link
Member Author

clonm commented Jan 23, 2019

@mitar I thought you said we did?
Also:
pdns.cloyne

# Check interval.
bind-check-interval=3600

bindbackend.conf
etc/powerdns/bind/

I uncommented # launch=bind and add bind-config=/etc/powerdns/bindbackend.conf in /etc/powerdns/pdns.conf to get the service to restart on server2 (without that, it was saying Unable to launch, no backends configured for querying). It's still logging in the wrong place but it had (on its own) stopped logging at all and wasn't responsive to pdns_control status from within the container. If we aren't using BIND though, this was probably not the right fix...

@mitar
Copy link
Member

mitar commented Jan 23, 2019

No, we are using bind-backend which allows PowerDNS to use bind configuration/zone files. So bind zone files are so common that our DNS server supports using them. Because they are common, I am familiar with them so it was an easy solution for configuration to store in files.

@mitar
Copy link
Member

mitar commented Jan 23, 2019

For PowerDNS it seems they are still using master/slave terminology: https://doc.powerdns.com/authoritative/modes-of-operation.html

You can always try to submit a patch to them. Or, patch just our Docker image. Depending how much you care about this.

@clonm
Copy link
Member Author

clonm commented Jan 23, 2019

Thanks for explaining the difference. The version of PowerDNS we use is no longer supported anyway - are there alternatives to PowerDNS we might be able to use instead instead of upgrading? Or is that even more complicated than submitting an upstream patch (which I probably won't do since I understand it's still controversial within the community and I don't have that kind of clout) (yet).

@mitar
Copy link
Member

mitar commented Jan 23, 2019

So do you know if it is controversial in that particular community? Maybe just nobody brought it up.

I like PowerDNS otherwise. It is a pretty good tool. You might want to update to bind though. You already have all configuration. :-)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants