vexxhost / ansible-collection-ceph Goto Github PK
View Code? Open in Web Editor NEWAnsible collection for deploying Ceph
Ansible collection for deploying Ceph
root@instance:~# time cephadm shell ceph -s
Inferring fsid 3ed20d13-abf7-55f5-a31b-720e50e123fb
Inferring config /var/lib/ceph/3ed20d13-abf7-55f5-a31b-720e50e123fb/mon.instance/config
Using ceph image with id '5be31c24972a' and tag 'v18.2.1' created on 2024-02-22 16:12:19 +0000 UTC
quay.io/ceph/ceph@sha256:9f35728f6070a596500c0804814a12ab6b98e05067316dc64876fb4b28d04af3
cluster:
id: 3ed20d13-abf7-55f5-a31b-720e50e123fb
health: HEALTH_WARN
1 stray daemon(s) not managed by cephadm
3 pool(s) have no replicas configured
services:
mon: 1 daemons, quorum instance (age 29m)
mgr: instance.zsmnlu(active, since 29m)
osd: 3 osds: 3 up (since 29m), 3 in (since 9h)
rgw: 1 daemon active (1 hosts, 1 zones)
data:
pools: 12 pools, 185 pgs
objects: 1.37k objects, 7.5 GiB
usage: 7.7 GiB used, 3.0 TiB / 3.0 TiB avail
pgs: 185 active+clean
real 0m14.520s
user 0m0.435s
sys 0m0.298s
vs
root@instance:~# time cephadm --image quay.io/ceph/ceph:v18.2.1 shell --fsid 3ed20d13-abf7-55f5-a31b-720e50e123fb --config /var/lib/ceph/3ed20d13-abf7-55f5-a31b-720e50e123fb/mon.instance/config ceph -s
cluster:
id: 3ed20d13-abf7-55f5-a31b-720e50e123fb
health: HEALTH_WARN
1 stray daemon(s) not managed by cephadm
3 pool(s) have no replicas configured
services:
mon: 1 daemons, quorum instance (age 28m)
mgr: instance.zsmnlu(active, since 28m)
osd: 3 osds: 3 up (since 28m), 3 in (since 9h)
rgw: 1 daemon active (1 hosts, 1 zones)
data:
pools: 12 pools, 185 pgs
objects: 1.37k objects, 7.5 GiB
usage: 7.7 GiB used, 3.0 TiB / 3.0 TiB avail
pgs: 185 active+clean
io:
client: 1.7 KiB/s rd, 426 B/s wr, 1 op/s rd, 1 op/s wr
real 0m1.741s
user 0m0.331s
sys 0m0.158s
almost 15x faster, we gotta do this, it'll speed up the CI tremendously.
By default, ceph mgr prometheus module is not enabled, and therefore, no metrics in prometheus for alerting
Currently CI job using v3 standard flavor that has been unsupported for a while. we can move most of job to use ubuntu-latest from github but not for adopt jobs.
I think we should also move ceph CI to use zuul so we can unblock without skip adopt jobs
There are scenarios in which we have a couple of write intensive disks to server a bunch of read intensive or spinning disks.
i.e https://docs.ceph.com/en/latest/rados/configuration/bluestore-config-ref/#block-and-block-db
there's a error block ansible-collection-ceph from publish https://github.com/vexxhost/ansible-collection-ceph/actions/runs/7425054349/job/20253763886
Error when finding available api versions from default (https://galaxy.ansible.com) (HTTP Code: 403, Message: Forbidden)
Because of run_once: true
in the Get ceph-volume lvm list status
, OSDs are not added to new nodes because the tasks run once only in a node which already has all disks added as OSD so everything is skipped on the new nodes.
Within Quincy ceph-volume has been moved to his own package. So at
https://github.com/vexxhost/ansible-collection-ceph/blob/main/roles/osd/tasks/main.yml#L17C11-L17C11
is missing the package ceph-volume.
It seems that the release-please
job is currently failing:
https://github.com/vexxhost/ansible-collection-ceph/actions/runs/7424904361/job/20205460119
I've seen this failure before but I believe perhaps @okozachenko1203 has done fixes for it in the past.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.