Michal Arbet 09d0409ed4 Allow user to set sysctl_net_ipv4_tcp_retries2
This patch is adding configuration option to
manipulate with kernel option sysctl_net_ipv4_tcp_retries2.

More informations about kernel option in [1][2]
and RedHat suggestion [3] to set for DBs and HA.

[1]: https://pracucci.com/linux-tcp-rto-min-max-and-tcp-retries2.html
[2]: https://blog.cloudflare.com/when-tcp-sockets-refuse-to-die/
[3]: https://access.redhat.com/solutions/726753

Closes-Bug: #1917068
Change-Id: Ia0decbbfa4e33b1889b635f8bb1c9094567a2ce6
2021-06-23 10:34:12 +00:00

48 lines
1.4 KiB
ReStructuredText

.. _haproxy-guide:
=============
HAProxy Guide
=============
Kolla Ansible supports a Highly Available (HA) deployment of
Openstack and other services. High-availability in Kolla
is implented as via Keepalived and HAProxy. Keepalived manages virtual IP
addresses, while HAProxy load-balances traffic to service backends.
These two components must be installed on the same hosts
and they are deployed to hosts in the ``haproxy`` group.
Preparation and deployment
~~~~~~~~~~~~~~~~~~~~~~~~~~
HAProxy and Keepalived are enabled by default. They may be disabled by
setting the following in ``/etc/kolla/globals.yml``:
.. code-block:: yaml
enable_haproxy: "no"
enable_keepalived: "no"
Configuration
~~~~~~~~~~~~~
Failover tuning
---------------
When a VIP fails over from one host to another, hosts may take some
time to detect that the connection has been dropped. This can lead
to service downtime.
To reduce the time by the kernel to close dead connections to VIP
address, modify the ``net.ipv4.tcp_retries2`` kernel option by setting
the following in ``/etc/kolla/globals.yml``:
.. code-block:: yaml
haproxy_host_ipv4_tcp_retries2: 6
This is especially helpful for connections to MariaDB. See
`here <https://pracucci.com/linux-tcp-rto-min-max-and-tcp-retries2.html>`__,
`here <https://blog.cloudflare.com/when-tcp-sockets-refuse-to-die/>`__ and
`here <https://access.redhat.com/solutions/726753>`__ for
further information about this kernel option.