From 0fd01ac1a73b24e86163b8e7f158bccb9565e305 Mon Sep 17 00:00:00 2001 From: Ben Vincent Date: Sun, 9 Aug 2026 00:00:20 +1000 Subject: [PATCH] ceph: manage /etc/ceph/ceph.conf on osd and mon/mgr/mds hosts (#515) ## Why `/etc/ceph/ceph.conf` on the prodnxsr ceph cluster (fsid `de96a98f`) is currently unmanaged and hand-maintained via the runbook (`profiles::ceph::client::manage_ceph_conf` is `false` on the k8s and incus roles). This brings it under Puppet, drives it from a single source of truth in hiera, and normalizes `public_network` to every ceph host's `/32`. ## How hosts are identified (osd-only vs mon/mgr/mds) The authoritative, declarative signal is the role's `hiera_include` list (ENC `role` -> role hiera): - **osd-only hosts** = `roles::infra::k8s::*` (prodnxsr0001-0008 and prodnxsr0014-0019): include `profiles::ceph::osd` + `::client`, NOT `::mds` -> `[global]` only. - **mon/mgr/mds hosts** = `roles::infra::incus::node` (prodnxsr0009-0013): also include `profiles::ceph::mds` -> `[global]` + `[mds]` + `[mds.*]`. The template keys on `profiles::ceph::client::render_mds_config`, set `true` only in the incus node role hiera (the single role that includes `profiles::ceph::mds`). The `is_ceph_mds` fact is a runtime fallback, not used here. ## Changes - Add cluster topology as a single source of truth in `hieradata/common.yaml`: `cluster_public_ips` (all 19 ceph host `/32`s, 198.18.23.1-19), `mon_initial_members` (the 5 mons), and the `mds_instances` map (two mds daemons per mon/mgr/mds host). - Rework `profiles/ceph/client.conf.erb` to render `[global]` always and the `[mds]`/`[mds.*]` sections only when `render_mds_config` is true, preserving the live two-space indent, exact key order, and blank-line separators between mds sections. - Extend `profiles::ceph::client` with topology + tunable params (sample values as defaults) and gate `[mds]` on `render_mds_config`. - Drop the hard `Package[ceph-common]` dependency when the class does not manage the package (cephadm / `profiles::packages` deliver it on the k8s and incus hosts; k8s hosts do not declare `ceph-common`, so an unconditional require would fail to compile). - Enable `manage_ceph_conf: true` on `roles/infra/k8s.yaml`, `roles/infra/k8s/node.yaml`, and `roles/infra/incus/node.yaml`; add `render_mds_config: true` to the incus node role. prodnxsr0014-0019 run `roles::infra::k8s::compute` (confirmed live via PuppetDB; `is_ceph_osd` true, `is_ceph_mds` false), so they are already covered by `k8s.yaml` and get the `[global]`-only variant with no mds sections. ## Live-cluster safety review (please read before merge) This touches a running storage cluster's config. The rendered content is byte-equivalent to the live file **except** for the one deliberate `public_network` normalization below. - **Intended change:** `public_network` is normalized to all 19 ceph host `/32`s on every host (198.18.23.1-19). Adoption is now a **single `public_network`-line change on every cluster host**, no longer a no-op on the osd hosts: - osd hosts prodnxsr0001-0008 currently list `.1-.13` live -> first convergence adds `.14-.19`. - mon/mgr/mds hosts prodnxsr0009-0013 currently list only `.9-.13` live -> first convergence adds `.1-.8` and `.14-.19`. - prodnxsr0014-0019 (newly joined OSD hosts) get the file created with the full `[global]` variant. - **Verified byte diff against the live files** (`/etc/ceph/ceph.conf` pulled from prodnxsr0008 and prodnxsr0009): each differs by **exactly one line** (the expanded `public_network`); on prodnxsr0009 the `[mds]` + `[mds.*]` sections and their blank-line separators are byte-identical to live. Rendering prodnxsr0009 with its current live `.9-.13` value reproduces the live file byte-for-byte, confirming the format match. - RGW hosts (`roles::infra::ceph::rgw`, ausyd1nxvm2115-2119) include only `profiles::ceph::rgw` -> `profiles::ceph::conf`; they do not include `profiles::ceph::client`, so their variant is untouched and there is no duplicate `File[/etc/ceph/ceph.conf]`. Rendered samples: - osd-only (prodnxsr0008): `[global]` with all 19 `/32`s in `public_network`, nothing else. - mon/mgr/mds (prodnxsr0009): the same `[global]` plus `[mds]` and `[mds.-1|2]` for prodnxsr0009-0013, blank-line separated. Reviewed-on: https://git.unkin.net/unkin/puppet-prod/pulls/515 Co-authored-by: Ben Vincent Co-committed-by: Ben Vincent --- hieradata/common.yaml | 39 ++++++++++++++++++++ hieradata/roles/infra/incus/node.yaml | 4 +- hieradata/roles/infra/k8s.yaml | 2 +- hieradata/roles/infra/k8s/node.yaml | 2 +- site/profiles/manifests/ceph/client.pp | 34 +++++++++++++++-- site/profiles/templates/ceph/client.conf.erb | 29 ++++++++++++++- 6 files changed, 102 insertions(+), 8 deletions(-) diff --git a/hieradata/common.yaml b/hieradata/common.yaml index 0524318..a0939e5 100644 --- a/hieradata/common.yaml +++ b/hieradata/common.yaml @@ -409,6 +409,45 @@ profiles::ceph::client::mons: - 10.18.15.2 - 10.18.15.3 +# de96a98f cluster topology (prodnxsr0001-0019) - single source of truth for +# /etc/ceph/ceph.conf rendered by profiles::ceph::client on the k8s (osd) and +# incus (mon/mgr/mds) roles. fsid/mons are overridden per-role in the role hiera. +# public_network is the /32 of every ceph host's ceph-public loopback. +profiles::ceph::client::cluster_public_ips: + - 198.18.23.1 + - 198.18.23.2 + - 198.18.23.3 + - 198.18.23.4 + - 198.18.23.5 + - 198.18.23.6 + - 198.18.23.7 + - 198.18.23.8 + - 198.18.23.9 + - 198.18.23.10 + - 198.18.23.11 + - 198.18.23.12 + - 198.18.23.13 + - 198.18.23.14 + - 198.18.23.15 + - 198.18.23.16 + - 198.18.23.17 + - 198.18.23.18 + - 198.18.23.19 +profiles::ceph::client::mon_initial_members: + - prodnxsr0009 + - prodnxsr0010 + - prodnxsr0011 + - prodnxsr0012 + - prodnxsr0013 +# two mds daemon instances per mon/mgr/mds host (rendered only where +# render_mds_config is true, i.e. the incus node role). +profiles::ceph::client::mds_instances: + prodnxsr0009: 2 + prodnxsr0010: 2 + prodnxsr0011: 2 + prodnxsr0012: 2 + prodnxsr0013: 2 + profiles::ceph::conf::config: global: auth_client_required: 'cephx' diff --git a/hieradata/roles/infra/incus/node.yaml b/hieradata/roles/infra/incus/node.yaml index cc68287..1cb9fbb 100644 --- a/hieradata/roles/infra/incus/node.yaml +++ b/hieradata/roles/infra/incus/node.yaml @@ -203,7 +203,9 @@ profiles::accounts::sysadmin::extra_groups: - incus-admin # manage cephfs mounts -profiles::ceph::client::manage_ceph_conf: false +profiles::ceph::client::manage_ceph_conf: true +# mon/mgr/mds host: render the [mds] + [mds.*] sections +profiles::ceph::client::render_mds_config: true profiles::ceph::client::manage_ceph_package: false profiles::ceph::client::manage_ceph_paths: false profiles::ceph::client::fsid: 'de96a98f-3d23-465a-a899-86d3d67edab8' diff --git a/hieradata/roles/infra/k8s.yaml b/hieradata/roles/infra/k8s.yaml index b91b86a..fa9131a 100644 --- a/hieradata/roles/infra/k8s.yaml +++ b/hieradata/roles/infra/k8s.yaml @@ -92,7 +92,7 @@ profiles::packages::include: profiles::selinux::setenforce::mode: disabled -profiles::ceph::client::manage_ceph_conf: false +profiles::ceph::client::manage_ceph_conf: true profiles::ceph::client::manage_ceph_package: false profiles::ceph::client::manage_ceph_paths: false profiles::ceph::client::fsid: 'de96a98f-3d23-465a-a899-86d3d67edab8' diff --git a/hieradata/roles/infra/k8s/node.yaml b/hieradata/roles/infra/k8s/node.yaml index 6dfb9ae..9aca0a0 100644 --- a/hieradata/roles/infra/k8s/node.yaml +++ b/hieradata/roles/infra/k8s/node.yaml @@ -33,7 +33,7 @@ profiles::packages::include: profiles::selinux::setenforce::mode: disabled -profiles::ceph::client::manage_ceph_conf: false +profiles::ceph::client::manage_ceph_conf: true profiles::ceph::client::manage_ceph_package: false profiles::ceph::client::manage_ceph_paths: false profiles::ceph::client::fsid: 'de96a98f-3d23-465a-a899-86d3d67edab8' diff --git a/site/profiles/manifests/ceph/client.pp b/site/profiles/manifests/ceph/client.pp index db7187c..4561d6a 100644 --- a/site/profiles/manifests/ceph/client.pp +++ b/site/profiles/manifests/ceph/client.pp @@ -2,6 +2,9 @@ class profiles::ceph::client ( String $fsid, Array[Stdlib::Host] $mons, + # cluster topology (single source of truth: hieradata/common.yaml) + Array[Stdlib::Host] $cluster_public_ips, + Array[Stdlib::Host] $mon_initial_members, Stdlib::Absolutepath $config_file = '/etc/ceph/ceph.conf', Boolean $manage_ceph_conf = true, Boolean $manage_ceph_package = true, @@ -10,6 +13,24 @@ class profiles::ceph::client ( String $group = 'ceph', Stdlib::Filemode $mode = '0644', Hash $keyrings = {}, + # [global] tunables (defaults match the live hand-maintained ceph.conf) + String $auth_client_required = 'cephx', + String $auth_cluster_required = 'cephx', + String $auth_service_required = 'cephx', + Boolean $mon_allow_pool_delete = true, + Boolean $ms_bind_ipv4 = true, + Boolean $ms_bind_ipv6 = false, + Integer $osd_crush_chooseleaf_type = 1, + Integer $osd_pool_default_min_size = 2, + Integer $osd_pool_default_size = 3, + Integer $osd_pool_default_pg_num = 128, + # mds config sections; only rendered on mon/mgr/mds hosts (render_mds_config) + Boolean $render_mds_config = false, + Hash[String, Integer] $mds_instances = {}, + Hash $mds_common = { + 'keyring' => '/var/lib/ceph/mds/ceph-$id/keyring', + 'mds_standby_replay' => true, + }, ) { # dont run this on proxmox nodes @@ -22,6 +43,13 @@ class profiles::ceph::client ( } } + # only depend on the package when this class manages it; on the ceph + # hosts the package is delivered by cephadm / profiles::packages instead. + $config_require = $manage_ceph_package ? { + true => Package['ceph-common'], + default => undef, + } + # manage the ceph directory if $manage_ceph_paths { file { '/etc/ceph': @@ -29,11 +57,11 @@ class profiles::ceph::client ( owner => $owner, group => $group, mode => $mode, - require => Package['ceph-common'], + require => $config_require, } } - # create a basic client config + # render /etc/ceph/ceph.conf from cluster topology in hiera if $manage_ceph_conf { file { $config_file: ensure => file, @@ -41,7 +69,7 @@ class profiles::ceph::client ( group => $group, mode => $mode, content => template('profiles/ceph/client.conf.erb'), - require => Package['ceph-common'], + require => $config_require, } } diff --git a/site/profiles/templates/ceph/client.conf.erb b/site/profiles/templates/ceph/client.conf.erb index af4c678..470ec9e 100644 --- a/site/profiles/templates/ceph/client.conf.erb +++ b/site/profiles/templates/ceph/client.conf.erb @@ -1,3 +1,28 @@ [global] - fsid = <%= @fsid %> - mon_host = <%= @mons.join(' ') %> + auth_client_required = <%= @auth_client_required %> + auth_cluster_required = <%= @auth_cluster_required %> + auth_service_required = <%= @auth_service_required %> + fsid = <%= @fsid %> + mon_allow_pool_delete = <%= @mon_allow_pool_delete %> + mon_initial_members = <%= @mon_initial_members.join(',') %> + mon_host = <%= @mons.join(',') %> + ms_bind_ipv4 = <%= @ms_bind_ipv4 %> + ms_bind_ipv6 = <%= @ms_bind_ipv6 %> + osd_crush_chooseleaf_type = <%= @osd_crush_chooseleaf_type %> + osd_pool_default_min_size = <%= @osd_pool_default_min_size %> + osd_pool_default_size = <%= @osd_pool_default_size %> + osd_pool_default_pg_num = <%= @osd_pool_default_pg_num %> + public_network = <%= @cluster_public_ips.map { |ip| "#{ip}/32" }.join(',') %> +<% if @render_mds_config -%> + +[mds] + keyring = <%= @mds_common['keyring'] %> + mds_standby_replay = <%= @mds_common['mds_standby_replay'] %> +<% @mds_instances.sort.each do |host, count| -%> +<% (1..count).each do |instance| -%> + +[mds.<%= host %>-<%= instance %>] + host = <%= host %> +<% end -%> +<% end -%> +<% end -%>