Freeze Break Request: Clean up iptables on all boxes.

[Stephen John Smoogen]

And we have a lot of small fixes whcih.

[Kevin Fenzi]

On Fri, 18 Mar 2016 16:59:14 -0600
Stephen John Smoogen  wrote:
> And we have a lot of small fixes whcih.

I'd say make the ipv6 version use a seperate variable like
blocked_ips_v6 (since they are going to be a different format, etc).

Otherwise seems fine. +1

kevin

[Peter Robinson]

On Fri, Mar 18, 2016 at 10:59 PM, Stephen John Smoogen  wrote:
> And we have a lot of small fixes whcih.

Why wouldn't a cleanup wait until post freeze? If they've been a mess
for this long why possibly impact in a freeze, especially on a Friday
night when people aren't around to test? Oh, and there's no details.

-1

[Stephen John Smoogen]

On 18 March 2016 at 17:09, Peter Robinson  wrote:
> On Fri, Mar 18, 2016 at 10:59 PM, Stephen John Smoogen  wrote:
> > And we have a lot of small fixes whcih.
>
> Why wouldn't a cleanup wait until post freeze? If they've been a mess
> for this long why possibly impact in a freeze, especially on a Friday
> night when people aren't around to test? Oh, and there's no details.

You are correct.

1) The change is to allow for hosts to be blocked via the ansible
playbook using a jinja template versus adding a version per system.
2) The change was tested and deployed on download servers to try and
get our rsync usable from some bad hosts.
3) Kevin asked me to put the change into all the templates before I forgot.
4) The change only comes into place if you have a variable defined
which is only in 1 playbook.
> -1
> _______________________________________________
> infrastructure mailing list
> [e-mail]
> http://lists.fedoraproject.org/admin/lists/[e-mail]...

[Peter Robinson]

On Fri, Mar 18, 2016 at 11:13 PM, Stephen John Smoogen  wrote:
> On 18 March 2016 at 17:09, Peter Robinson  wrote:
> > On Fri, Mar 18, 2016 at 10:59 PM, Stephen John Smoogen  wrote:
> > > And we have a lot of small fixes whcih.
> >
> > Why wouldn't a cleanup wait until post freeze? If they've been a mess
> > for this long why possibly impact in a freeze, especially on a Friday
> > night when people aren't around to test? Oh, and there's no details.
>
> You are correct.
> The change is to allow for hosts to be blocked via the ansible
>
> playbook using a jinja template versus adding a version per system.
> 2) The change was tested and deployed on download servers to try and
> get our rsync usable from some bad hosts.
> 3) Kevin asked me to put the change into all the templates before I forgot.
> 4) The change only comes into place if you have a variable defined
> which is only in 1 playbook.

So the above sounds reasonable but there was no details or patch to
ansible to be able to access impact.

[Stephen John Smoogen]

On 18 March 2016 at 17:29, Peter Robinson  wrote:
> On Fri, Mar 18, 2016 at 11:13 PM, Stephen John Smoogen  wrote:
> > On 18 March 2016 at 17:09, Peter Robinson  wrote:
> > > On Fri, Mar 18, 2016 at 10:59 PM, Stephen John Smoogen  wrote:
> > > > And we have a lot of small fixes whcih.
> > >
> > > Why wouldn't a cleanup wait until post freeze? If they've been a mess
> > > for this long why possibly impact in a freeze, especially on a Friday
> > > night when people aren't around to test? Oh, and there's no details.
> >
> > You are correct.
> > The change is to allow for hosts to be blocked via the ansible
> >
> > playbook using a jinja template versus adding a version per system.
> > 2) The change was tested and deployed on download servers to try and
> > get our rsync usable from some bad hosts.
> > 3) Kevin asked me to put the change into all the templates before I forgot.
> > 4) The change only comes into place if you have a variable defined
> > which is only in 1 playbook.
>
> So the above sounds reasonable but there was no details or patch to
> ansible to be able to access impact.

My email shows an attachment with the patch included. Did it get
stripped somewhere?
> infrastructure mailing list
> [e-mail]
> http://lists.fedoraproject.org/admin/lists/[e-mail]...