Pcs is a Corosync and Pacemaker configuration tool. It permits users to easily view, modify and create Pacemaker based clusters. Pcs contains pcsd, a pcs daemon, which operates as a remote server for pcs.
There are three pcs branches:
- master
- This is where pcs-0.11 lives.
- Clusters running Pacemaker 2.1+ on top of Corosync 3.x are supported.
- The main development happens here.
- pcs-0.10
- Clusters running Pacemaker 2.0+ on top of Corosync 3.x are supported.
- This branch is in maintenance mode - bugs are being fixed but only a subset of new features lands here.
- pcs-0.9
- Clusters running Pacemaker 1.x on top of Corosync 2.x or Corosync 1.x with CMAN are supported.
- This branch is in no longer maintained.
These are the runtime dependencies of pcs and pcsd:
- python 3.6+
- python3-cryptography
- python3-dateutil 2.7.0+
- python3-lxml
- python3-pycurl
- python3-setuptools
- python3-setuptools_scm
- python3-pyparsing
- python3-tornado 6.1.0+
- python dataclasses (
pip install dataclasses
; required only for python 3.6, already included in 3.7+) - dacite
- ruby 2.2.0+
- killall (package psmisc)
- corosync 3.x
- pacemaker 2.1+
Apart from the dependencies listed above, these are also required for installation:
- python development files (packages python3-devel, python3-setuptools, python3-setuptools_scm, python3-wheel)
- ruby development files (package ruby-devel)
- rubygems
- rubygem bundler (package rubygem-bundler or ruby-bundler or bundler)
- autoconf, automake
- gcc
- gcc-c++
- FFI development files (package libffi-devel or libffi-dev)
- printf (package coreutils)
- redhat-rpm-config (if you are using Fedora)
- wget (to download bundled libraries)
During the installation, all required rubygems are automatically downloaded and compiled.
To install pcs and pcsd run the following in terminal:
./autogen.sh
./configure
# alternatively './configure --enable-local-build' can be used to also download
# missing dependecies
make
make install
If you are using GNU/Linux with systemd, it is now time to:
systemctl daemon-reload
Start pcsd and make it start on boot:
systemctl start pcsd
systemctl enable pcsd
Currently this is built into Fedora, RHEL and its clones and Debian and its derivates.
-
Authenticate cluster nodes
Set the same password for the
hacluster
user on all nodes.passwd hacluster
To authenticate the nodes, run the following command on one of the nodes (replacing node1, node2, node3 with a list of nodes in your future cluster). Specify all your cluster nodes in the command. Make sure pcsd is running on all nodes.
pcs host auth node1 node2 node3 -u hacluster
-
Create a cluster
To create a cluster run the following command on one node (replacing cluster_name with a name of your cluster and node1, node2, node3 with a list of nodes in the cluster).
--start
and--enable
will start your cluster and configure the nodes to start the cluster on boot respectively.pcs cluster setup cluster_name node1 node2 node3 --start --enable
-
Check the cluster status
After a few moments the cluster should startup and you can get the status of the cluster.
pcs status
-
Add cluster resources
After this you can add stonith agents and resources:
pcs stonith create --help
and
pcs resource create --help
ClusterLabs website is an excellent place to learn more about Pacemaker clusters.
If you have any bug reports or feature requests please feel free to open a github issue on the pcs project.
Alternatively you can use ClusterLabs users mailinglist which is also a great place to ask Pacemaker clusters related questions.