Bug 1839637

Summary: [RFE] Add a 'local' cluster setup command
Product: Red Hat Enterprise Linux 8 Reporter: Tomas Jelinek <tojeline>
Component: pcsAssignee: Tomas Jelinek <tojeline>
Status: CLOSED ERRATA QA Contact: cluster-qe <cluster-qe>
Severity: unspecified Docs Contact: Steven J. Levine <slevine>
Priority: high    
Version: 8.2CC: cfeist, cluster-maint, idevat, lmanasko, mlisik, mmazoure, mpospisi, nhostako, omular, pasik, slevine, tojeline
Target Milestone: rcKeywords: FutureFeature, TechPreview
Target Release: 8.3   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: pcs-0.10.6-1.el8 Doc Type: Enhancement
Doc Text:
.Local mode version of `pcs cluster setup` command is now fully supported By default, the `pcs cluster setup` command automatically synchronizes all configuration files to the cluster nodes. Since RHEL 8.3, the `pcs cluster setup` command has provided the `--corosync-conf` option as a Technology Preview. This feature is now fully supported in RHEL 8.5. Specifying this option switches the command to `local` mode. In this mode, the `pcs` command-line interface creates a `corosync.conf` file and saves it to a specified file on the local node only, without communicating with any other node. This allows you to create a `corosync.conf` file in a script and handle that file by means of the script.
 Story Points: ---
Clone Of:
: 1841019 (view as bug list) Environment:
Last Closed: 2020-11-04 02:28:18 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Attachments:
Description Flags
proposed fix + tests none

Description Tomas Jelinek 2020-05-25 07:06:10 UTC 
The current 'pcs cluster setup' command does all that is needed to set up a cluster: checking if nodes are suitable for the new cluster, creating corosync.conf and other cluster config files (authkeys, pcs configs, etc.) and distributing them to all the nodes. This is great for the cases when a user is creating the cluster manually. It is not so good when one wants to write a script to set up and configure a cluster.

To make it easier to set up a cluster from a script, a new command will be implemented in pcs. This command will have pretty much the same syntax as the current 'pcs cluster setup'. It will, however, work only on the local node, validating corosync.conf options and storing resulting corosync.conf in a specified location. It will be up to the script to distribute this file, along with other needed config files created by such script, to actually create a cluster. The local setup pcs command will not communicate with any nodes.
Comment 2 Tomas Jelinek 2020-05-25 07:17:27 UTC 
Created attachment 1691709 [details]
proposed fix + tests

The 'pcs cluster setup' command has a new option, --corosync_conf, which switches the command to 'local' mode. This is described in pcs man page and help:

Local only mode:
By default, pcs connects to all specified nodes to verify they can be used in the new cluster and to send cluster configuration files to them. If this is not what you want, specify --corosync_conf option followed by a file path. Pcs will save corosync.conf to the specified file and will not connect to cluster nodes. These are the task pcs skips in that case:
* make sure the nodes are not running or configured to run a cluster already
* make sure cluster packages are installed on all nodes and their versions are compatible
* make sure there are no cluster configuration files on any node (run 'pcs cluster destroy' and remove pcs_settings.conf file on all nodes)
* synchronize corosync and pacemaker authkeys, /etc/corosync/authkey and /etc/pacemaker/authkey respectively, and the corosync.conf file
* authenticate the cluster nodes against each other ('pcs cluster auth' or 'pcs host auth' command)
* synchronize pcsd certificates (so that pcs web UI can be used in an HA mode)
Comment 3 Tomas Jelinek 2020-05-25 07:22:03 UTC 
This feature has been also requested upstream:
https://lists.clusterlabs.org/pipermail/users/2020-April/027128.html
Comment 9 Miroslav Lisik 2020-06-11 14:25:48 UTC 
Test:

[root@r8-node-01 pcs]# rpm -q pcs
pcs-0.10.6-1.el8.x86_64

[root@r8-node-01 pcs]# pcs cluster setup LocalCluster r8-node-0{1,2} --corosync_conf  local_corosync.conf
No addresses specified for host 'r8-node-01', using 'r8-node-01'
No addresses specified for host 'r8-node-02', using 'r8-node-02'
[root@r8-node-01 pcs]# cat local_corosync.conf
totem {
    version: 2
    cluster_name: LocalCluster
    transport: knet
    crypto_cipher: aes256
    crypto_hash: sha256
}

nodelist {
    node {
        ring0_addr: r8-node-01
        name: r8-node-01
        nodeid: 1
    }

    node {
        ring0_addr: r8-node-02
        name: r8-node-02
        nodeid: 2
    }
}

quorum {
    provider: corosync_votequorum
    two_node: 1
}

logging {
    to_logfile: yes
    logfile: /var/log/cluster/corosync.log
    to_syslog: yes
    timestamp: on
}
Comment 22 errata-xmlrpc 2020-11-04 02:28:18 UTC 
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (pcs bug fix and enhancement update), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHEA-2020:4617