IPv6 VIP addresses need to be /128

We currently hardcode /64 as our VIP addresses when using IPv6.
The problem with this is that some server code might bind to that
IP as a source address when doing inter-cluster communication
(rabbitmq/galera for example). So when the VIP moves there will
be effectively a network outage between the nodes, which should not
happen.

Likely this was hardcoded to /64 because the RA IPaddr2 needs a nic
parameter when /128 is specified. This is due to:
https://bugzilla.redhat.com/show_bug.cgi?id=1445628

We also make sure we use the ipv6_addrlabel option set to 99 so that
they will never be used as source ip addresses.

Depends-On: I7fcf15a00aedbdcfb21db501ad46c69fb97ec30c
Partial-Bug: #1686357

Change-Id: Ibefde870512ad1e03ff12f7aea91b3734f03f96f
Co-Authored-By: Sofer Athlan-Guyot <sathlang@redhat.com>
Co-Authored-By: Marios Andreou <mandreou@redhat.com>
Co-Authored-By: Damien Ciabrini <dciabrin@redhat.com>
(cherry picked from commit 6227484b60)
This commit is contained in:
Michele Baldessari 2017-04-26 11:40:51 +02:00
parent f9867d23c5
commit 7e1e6093db
1 changed files with 14 additions and 6 deletions

View File

@ -57,17 +57,25 @@ define tripleo::pacemaker::haproxy_with_vip(
$ensure = true)
{
if($ensure) {
# NB: Until the IPaddr2 RA has a fix for https://bugzilla.redhat.com/show_bug.cgi?id=1445628
# we need to specify the nic when creating the ipv6 vip.
if is_ipv6_address($ip_address) {
$netmask = '64'
$netmask = '128'
$nic = interface_for_ip($ip_address)
$ipv6_addrlabel = '99'
} else {
$netmask = '32'
$netmask = '32'
$nic = ''
$ipv6_addrlabel = ''
}
pacemaker::resource::ip { "${vip_name}_vip":
ip_address => $ip_address,
cidr_netmask => $netmask,
location_rule => $location_rule,
tries => $pcs_tries,
ip_address => $ip_address,
cidr_netmask => $netmask,
nic => $nic,
ipv6_addrlabel => $ipv6_addrlabel,
location_rule => $location_rule,
tries => $pcs_tries,
}
pacemaker::constraint::order { "${vip_name}_vip-then-haproxy":
first_resource => "ip-${ip_address}",