Age | Commit message (Collapse) | Author | Files | Lines |
|
This only takes effect is internal-tls is used, and forces haproxy to
do proper verifications of the SSL certificates provided by the
servers.
bp tls-via-certmonger
Change-Id: Ibd98ec46dd6570887db29f55fe183deb1c9dc642
|
|
|
|
|
|
It needs it so HAProxy will be able to set the X-Forwarded-Proto header.
Related-Bug: #1640126
Change-Id: I1726fa1742bc70518338b80fc6d27567bb020e7c
|
|
HAProxy won't pass X-Forwarded-Proto to these services in mode tcp, so we need
to switch it to http in order for it to work and for the services to properly
set the protocol in the links they serve.
Change-Id: Ib10282159fb9269eebe81af23171ec9fb1297cd0
Closes-Bug: #1640126
|
|
|
|
|
|
Previously the ctrl plane VIP would default to 'br-ex' which in non-vlan
deployments ends up being the wrong interface. The public VIP interface
was also defaulted to 'br-ex' which would be incorrect for vlan based
deployments. Since a user has already given the nic template (and in
most cases the subnet that corresponds to the nic) the installer should
be able to figure out which interface the public/control vip should be
on.
These changes enable that type of auto-detection, unless a user
explicitly overrides the heat parameters for ControlVirtualInterface and
PublicVirtualInterface. Also removes calling keepalived from haproxy
now that the services are composed separately on the Controller role.
Partial-Bug: 1606632
Change-Id: I05105fce85be8ace986db351cdca2916f405ed04
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
We have a variable in hiera which tells us if the keepalived
service is enabled, so use it here. Without this any deployment
disabling OS::TripleO::Services::Keepalived will fail.
Change-Id: I90faf51881bd05920067c1e1d82baf5d7586af23
Closes-Bug: #1642677
|
|
|
|
Change-Id: I35f283bdf8dd0ed979c65633724f0464695130a4
|
|
|
|
This optionally enables TLS for Barbican API in the internal network.
If internal TLS is enabled, each node that is serving the Barbican API
service will use certmonger to request its certificate.
bp tls-via-certmonger
Change-Id: I1c1d3dab9bba7bec6296a55747e9ade242c47bd9
|
|
|
|
this looks like a copy/paste error. Let barbican use its own
hiera data.
Change-Id: I84118c1a561c3db2f18504b55c5a8c4f042e7e3b
|
|
|
|
|
|
|
|
The lastest patchset of https://review.openstack.org/393361 was actually
not working.
The `if defined` idiom depends on *evaluation* order.
At the time it's red in the haproxy.pp class, the line that loads the
class 'haproxy' has still not yet been reached and thus the `defined`
result is false. The constraint is not added.
For this reason, the use of `defined` in module is not advised by
puppetlabs[1].
[1] https://docs.puppet.com/puppet/latest/reference/function.html#defined
Change-Id: Ibd352cb313f8863d62db8987419378bed5b87256
Relates-To: #1638029
|
|
|
|
|
|
aodh, ceilometer, gnocchi and neutron need the X-Forwarded-Proto in
order to return links with the correct protocol when SSL is enabled.
This enables it in HAProxy
Change-Id: Icceab92f86b1cc40d42195fa4ba0c75f302795b8
Closes-Bug: #1640126
|
|
In HA deployments, we now check mysql nodes every 1s and removed them
immediately if they are failed. Previously we would check every 2s and
allow them to fail 5 checks before being removed, producing errors from
other OpenStack services for 10s, which causes confusion and delay for
operators.
Additionally, these check options are now also a class parameter so can
be overridden by operators.
Closes-Bug: #1639189
Change-Id: I0b915f790ae5a4b018a212d3aa83cca507be05e9
|
|
It's been proposed this may help with the
('Connection aborted.', BadStatusLine("''",)) errors.
This patch increase queue, server and client timeouts to 2m (default is 1m)
Related-Bug: #1638908
Change-Id: Ie4f059f3fad2271bb472697e85ede296eee91f5d
|
|
This optionally enables TLS for Cinder API in the internal network.
If internal TLS is enabled, each node that is serving the Cinder API
service will use certmonger to request its certificate.
bp tls-via-certmonger
Change-Id: Ib4a9c8d3ca57f1b02e1bb0d150f333db501e9863
|
|
Change-Id: Id4dc2379b0c423012a0b3aaf49d1e1a7d633a03b
|
|
|
|
|
|
This optionally enables TLS for Nova API in the internal network.
If internal TLS is enabled, each node that is serving the Nova API
service will use certmonger to request its certificate.
Note that this doesn't enable internal TLS for the nova metadata
service since it doesn't run over httpd. This will be handled in
a later commit.
bp tls-via-certmonger
Change-Id: I88380a1ed8fd597a1a80488cbc6ce357f133bd70
|
|
|
|
|
|
ODL was missing transparent binding mode, which causes HA deployments to
fail since HA Proxy will try to come up on every node (even without
VIP).
Closes-Bug: 1637833
Change-Id: I0bb7839cdcfeacb4ca1a9fc6f878e8b51330be92
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
proxy for the UI"
|
|
|
|
|
|
|
|
The upload and extraction for the plan tarball to swift can take
longer than the default one minute in slower environments. Doubling
the timeout to two minutes has proven to help.
This is only a partial fix, because the error reporting for this
issue also needs to be improved.
Change-Id: I06592d38fdfefacc8bdf76289a0bfa20eb33a89b
Partial-Bug: 1635269
|
|
The docker tooling has a preference for interacting with encrypted
endpoints. Terminating the docker-registry endpoint with HAProxy
allows the SSL VIP to be used for this purpose.
Change-Id: Ifebfa7256e0887d6f26a478ff8dc82b0ef5f65f6
|
|
This optionally enables TLS for gnocchi in the internal network.
If internal TLS is enabled, each node that is serving the gnocchi
service will use certmonger to request its certificate.
bp tls-via-certmonger
Change-Id: Ie983933e062ac6a7f0af4d88b32634e6ce17838b
|
|
This optionally enables TLS for aodh in the internal network.
If internal TLS is enabled, each node that is serving the aodh
service will use certmonger to request its certificate.
This, in turn should also configure a command that should be ran when
the certificate is refreshed (which requires the service to be
restarted).
bp tls-via-certmonger
Change-Id: I50ef0c8fbecb19d6597a28290daa61a91f3b13fc
|
|
This optionally enables TLS for aodh in the internal network.
If internal TLS is enabled, each node that is serving the ceilometer
service will use certmonger to request its certificate.
This, in turn should also configure a command that should be ran when
the certificate is refreshed (which requires the service to be
restarted).
bp tls-via-certmonger
Change-Id: Ib5609f77a31b17ed12baea419ecfab5d5f676496
|
|
This optionally enables TLS for keystone in the internal network.
If internal TLS is enabled, each node that is serving the keystone
service will use certmonger to request its certificate.
This, in turn should also configure a command that should be ran when
the certificate is refreshed (which requires the service to be
restarted).
bp tls-via-certmonger
Change-Id: I303f6cf47859284785c0cdc65284a7eb89a4e039
|
|
Co-Authored-By: Juan Antonio Osorio Robles <jaosorior@redhat.com>
Change-Id: If2804b469eb3ee08f3f194c7dd3290d23a245a7a
|
|
proxy for the UI
Change-Id: I74eac4bbfc16720eeb6e2bf0ee251689dde3bafc
Implements: enable-communication-ui-undercloud
|
|
By enabling the statistics socket we allow the collection
of statistics over time for haproxy.
This socket is set to "user" level, so this socket is limited
to read-only. The "stats timeout" line is optional, but since the
default timeout of the stats socket is 10s, we set this higher.
Change-Id: I22d3ab771e981be0d2c74b60443d276973bc1639
|
|
The service profile in HAProxy has the capability of creating
certificates based on a map. The idea is to standardize this, as
some of those certificates should match certain networks the services
are listening on (with the exception of the external network which is
handled differently and the tenant network which doesn't need a
certificate). So, based on which network a certain service is
listening on, we fetch the appropriate certificate.
bp tls-via-certmonger
Change-Id: I89001ae32f46c9682aecc118753ef6cd647baa62
|
|
Right now we're hardcoding the server names for the services to be
the controllers. This is problematic if we start using custom roles
for services, which listen on nodes that are not controllers.
We already have the server names for each service, so using this
mapping instead fixes the issue.
Change-Id: Ic4b65edb3dc1b75abbc3421a87cab97425b058c4
Closes-Bug: #1629098
|
|
|
|
|
|
Note that there was a need to modify different timeouts due
to the nature of how websockets work. The source where the
reasoning and value came from is listed as a comment in the
code.
Related-Bug: #1625448
Co-Authored-By: Brad P. Crochet <brad@redhat.com>
Change-Id: I9de77d5f692c1c9d04e3c59c5de5312e63f81aed
|