open-consul/website/pages/docs/architecture/gossip.mdx

62 lines
3.4 KiB
Plaintext
Raw Normal View History

2014-02-08 00:41:03 +00:00
---
2020-04-07 18:55:19 +00:00
layout: docs
page_title: Gossip Protocol
sidebar_title: Gossip Protocol
2020-04-07 18:55:19 +00:00
description: >-
Consul uses a gossip protocol to manage membership and broadcast messages to
the cluster. All of this is provided through the use of the Serf library. The
gossip protocol used by Serf is based on SWIM: Scalable Weakly-consistent
Infection-style Process Group Membership Protocol, with a few minor
adaptations.
2014-02-08 00:41:03 +00:00
---
# Gossip Protocol
2016-01-13 22:44:01 +00:00
Consul uses a [gossip protocol](https://en.wikipedia.org/wiki/Gossip_protocol)
2014-02-20 20:26:50 +00:00
to manage membership and broadcast messages to the cluster. All of this is provided
2016-08-08 16:44:27 +00:00
through the use of the [Serf library](https://www.serf.io/). The gossip protocol
2014-02-20 20:26:50 +00:00
used by Serf is based on
2018-01-22 01:41:58 +00:00
["SWIM: Scalable Weakly-consistent Infection-style Process Group Membership Protocol"](http://www.cs.cornell.edu/info/projects/spinglass/public_pdfs/swim.pdf),
2016-08-08 16:44:27 +00:00
with a few minor adaptations. There are more details about [Serf's protocol here](https://www.serf.io/docs/internals/gossip.html).
2014-02-08 00:41:03 +00:00
2014-02-20 20:26:50 +00:00
## Gossip in Consul
2014-02-08 00:41:03 +00:00
2014-02-20 20:26:50 +00:00
Consul makes use of two different gossip pools. We refer to each pool as the
LAN or WAN pool respectively. Each datacenter Consul operates in has a LAN gossip pool
containing all members of the datacenter, both clients and servers. The LAN pool is
used for a few purposes. Membership information allows clients to automatically discover
servers, reducing the amount of configuration needed. The distributed failure detection
allows the work of failure detection to be shared by the entire cluster instead of
concentrated on a few servers. Lastly, the gossip pool allows for reliable and fast
event broadcasts.
2014-02-08 00:41:03 +00:00
2014-02-20 20:26:50 +00:00
The WAN pool is globally unique, as all servers should participate in the WAN pool
regardless of datacenter. Membership information provided by the WAN pool allows
2014-04-09 18:06:27 +00:00
servers to perform cross datacenter requests. The integrated failure detection
2014-02-20 20:26:50 +00:00
allows Consul to gracefully handle an entire datacenter losing connectivity, or just
a single server in a remote datacenter.
2014-02-08 00:41:03 +00:00
2016-08-08 16:44:27 +00:00
All of these features are provided by leveraging [Serf](https://www.serf.io/). It
2014-02-20 20:26:50 +00:00
is used as an embedded library to provide these features. From a user perspective,
this is not important, since the abstraction should be masked by Consul. It can be useful
however as a developer to understand how this library is leveraged.
2016-09-14 17:08:34 +00:00
## Lifeguard Enhancements ((#lifeguard))
2016-09-14 17:08:34 +00:00
2016-09-14 17:17:23 +00:00
SWIM makes the assumption that the local node is healthy in the sense
that soft real-time processing of packets is possible. However, in cases
where the local node is experiencing CPU or network exhaustion this assumption
can be violated. The result is that the `serfHealth` check status can
2017-01-13 07:34:44 +00:00
occasionally flap, resulting in false monitoring alarms, adding noise to
2016-09-14 17:17:23 +00:00
telemetry, and simply causing the overall cluster to waste CPU and network
resources diagnosing a failure that may not truly exist.
Lifeguard completely resolves this issue with novel enhancements to SWIM.
2017-07-06 19:19:39 +00:00
For more details about Lifeguard, please see the
[Making Gossip More Robust with Lifeguard](https://www.hashicorp.com/blog/making-gossip-more-robust-with-lifeguard/)
blog post, which provides a high level overview of the HashiCorp Research paper
[Lifeguard : SWIM-ing with Situational Awareness](https://arxiv.org/abs/1707.00788). The
[Serf gossip protocol guide](https://www.serf.io/docs/internals/gossip.html#lifeguard)
also provides some lower-level details about the gossip protocol and Lifeguard.