While we actually use SLAAC in aws, there's a dhcp6d sending out the
router advertisements, so without that the instance doesn't get an ipv6
ip and just doesn't work. With this it does.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
This is a quick, hacked up script that just runs once per minute and
updates the ip addresses for the osbuild koji plugin. The script calls
systemd's resolvectl without cache and puts the ips in a ipset. The
koji_builder firewall has a added rule to check that ipset for outgoing
connections that are allowed.
TODO: add some kind of error checking
TODO: probibly won't work on s390x builders as they can't reach the host
even with open firewalls, but should work for others.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
This is a hack to work around SPF screwing us for @fedoraproject.org
aliases. It only fixes email from @redhat.com, but due to bugzilla thats
a lot of email.
Without this:
bugzilla@redhat.com -> user@fedoraproject.org (expands) ->
user@gmail.com sent out directly to gmail and gets rejected because
we aren't in the redhat.com SPF record.
With this:
bugzilla@redhat.com -> user@fedoraproject.org (expands) ->
user@gmail.com but sent to mx2.redhat.com to deliver. Since
mx2.redhat.com definitely is in the redhat.com SPF record the email is
delivered fine and SPF checks pass.
This won't help for other domains with -all SPF records, but at least it
helps for all the redhat.com emails, of which there are a lot going to
fedoraproject.org aliases. :)
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
We changed this to DEFAULT:FEDORA32 a while back because the certs for
the old totpcgi sudo needed it to work. Now thats all gone and we are
100% on ipa and sssd, this should no longer be needed.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Something is broken with smtp_tls_connection_reuse = yes, so disable it
for now. Also, setup a tls_policy map file and tell it to not use tls
for mx2.redhat.com. The normal smtp connection reuse works just fine, so
this will keep mail flowing until we can one day figure out why tls
connection reuse is busted.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
We fixed the config in a PR from aheath1992 for most of the machines,
but we need to fix vpn (proxies in particular) and releng boxes now.
Also, while we are here, lets drop the phx2 file since it's not used
anymore.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Without this it wasn't caching tls connections and was going over the
small limit redhat.com mx had. Hopefully this gets mail flowing again.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
So, without this tlsproxy wasn't working and no connection reuse was
happening. With it, it seems to be processing away nicely and reusing
connections.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Recently, redhat.com changed internal MX servers. The new servers are
have rate limits on incoming emails from one ip and admins there don't
want to add a bunch of exceptions, so we need to adjust our end to not
flood connections to them. Currently, connections burst up to 100 (the
smtp postfix default) which goes over their limits and causes the
internal MX to reject emails from us for a while.
So, this change:
* Adds some domains to fast_flush. This allows us to use postqueue -s
domain to flush emails to a particular domain.
* Changes the smtp limit to 40. This is under the redhat.com limit.
* Has ansible actually install the master.cf.gateway on bastion servers.
Currently they were using the stock/default one.
* Enables the tlsproxy service, which is actually needed to get that tls
reuse working.
After these changes, we keep few connections to the redhat.com mx open,
but we reuse them and send more emails over existing connections. No
'too many connection emails' have happened since the changes.
The queue slowly seems to be processing down.
Since this was causing an outage of email, I have already applied these
things to bastion01, but I'd like to make sure we match up to whats in
ansible.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Right now we are hitting redhat.com mx server connection limits.
This might be because we are starting too many new connections at once.
Enabling this should reduce the new connections by reusing existing
ones.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
In addition to the cloud_aws group, we want to exclude ibiblio virthosts
as they use bonding and thats a more complex setup. Someday we should
get it working here.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
Everything should now be using linux-system-roles/network, so we drop
our hacky nmcli calls and everything that referred to them, including
exclude variables. Also, lets just let NM handle resolv.conf so it's not
wrong all the time on reboots.
Signed-off-by: Kevin Fenzi <kevin@scrye.com>
