Back to bug 2149533

Who When What Removed Added
Red Hat One Jira (issues.redhat.com) 2022-11-30 04:21:57 UTC Link ID Red Hat Issue Tracker RHCEPH-5712
Pawan 2022-12-02 04:28:36 UTC Comment 0 updated
Red Hat Bugzilla 2023-01-01 05:37:26 UTC Assignee adking nobody
Red Hat Bugzilla 2023-01-01 08:28:38 UTC Docs Contact asriram
Red Hat Bugzilla 2023-01-01 08:37:24 UTC QA Contact msaini
Alasdair Kergon 2023-01-04 04:33:20 UTC Assignee nobody adking
Alasdair Kergon 2023-01-04 04:42:26 UTC QA Contact msaini
Adam King 2023-03-31 20:38:02 UTC Status NEW ASSIGNED
Adam King 2023-04-26 16:43:30 UTC Status ASSIGNED POST
Veera Raghava Reddy 2023-04-27 02:28:45 UTC Fixed In Version ceph-17.2.6-31.el9cp
CC tserlin
Status POST MODIFIED
CC msaini, vereddy
Flags needinfo?(msaini)
Flags needinfo?(vereddy)
Flags needinfo?(msaini) needinfo?(vereddy)
errata-xmlrpc 2023-04-27 04:00:26 UTC Status MODIFIED ON_QA
Akash Raj 2023-05-03 06:23:18 UTC Flags needinfo?(adking)
CC adking, akraj
Docs Contact akraj
Adam King 2023-05-03 18:21:03 UTC Doc Text Feature: Users can now set the crush location a mon deployed to a given host should be assigned in the mon spec file. For example a spec file like

service_type: mon
service_name: mon
placement:
hosts:
- host1
- host2
- host3
spec:
crush_locations:
host1:
- datacenter=a
host2:
- datacenter=b
- rack=2
host3:
- datacenter=a

would tell cephadm that the mon deployed on host1 should get crush location "datacenter=a", the mon deployed to host2 should get "datacenter=b,rack=2" etc. Cephadm will set the first crush location in the list for a host using the --set-crush-location argument that mon daemons accept, and set further options using the "ceph mon set_location" command in the background.


Reason: This was primarily added to make replacing tiebreaker mon daemons in stretch cluster deployed by cephadm more feasible. Without this change, uses would have to manually edit the files written by cephadm to deploy the tiebreaker mon, as the tiebreaker mon is not allowed to join without declaring its crush location.

Result: Users can now set the crush locations for mon daemons in the mon spec and should be able to use this to replace tiebreaker mon daemons without extra manual steps.
Flags needinfo?(adking)
Doc Type If docs needed, set a value Enhancement
Manisha Saini 2023-05-07 16:38:33 UTC QA Contact msaini vpapnoi
Akash Raj 2023-05-08 02:35:35 UTC Blocks 2192813
Vinayak Papnoi 2023-05-11 07:01:02 UTC Status ON_QA VERIFIED
Akash Raj 2023-05-18 05:44:01 UTC Doc Text Feature: Users can now set the crush location a mon deployed to a given host should be assigned in the mon spec file. For example a spec file like

service_type: mon
service_name: mon
placement:
hosts:
- host1
- host2
- host3
spec:
crush_locations:
host1:
- datacenter=a
host2:
- datacenter=b
- rack=2
host3:
- datacenter=a

would tell cephadm that the mon deployed on host1 should get crush location "datacenter=a", the mon deployed to host2 should get "datacenter=b,rack=2" etc. Cephadm will set the first crush location in the list for a host using the --set-crush-location argument that mon daemons accept, and set further options using the "ceph mon set_location" command in the background.


Reason: This was primarily added to make replacing tiebreaker mon daemons in stretch cluster deployed by cephadm more feasible. Without this change, uses would have to manually edit the files written by cephadm to deploy the tiebreaker mon, as the tiebreaker mon is not allowed to join without declaring its crush location.

Result: Users can now set the crush locations for mon daemons in the mon spec and should be able to use this to replace tiebreaker mon daemons without extra manual steps.
.Users can set the crush location for a Ceph Monitor to replace tiebreaker monitors

With this release, users can set the crush location for a monitor deployed on a host. It should be assigned in the mon specification file.

.Example
----
service_type: mon
service_name: mon
placement:
hosts:
- host1
- host2
- host3
spec:
crush_locations:
host1:
- datacenter=a
host2:
- datacenter=b
- rack=2
host3:
- datacenter=a
----

This is primarily added to make the replacing of a tiebreaker monitor daemon in stretch clusters deployed by Cephadm, more feasible. Without this change, users would have to manually edit the files written by Cephadm to deploy the tiebreaker monitor, as the tiebreaker monitor is not allowed to join without declaring its crush location.
errata-xmlrpc 2023-06-15 09:08:13 UTC Status VERIFIED RELEASE_PENDING
errata-xmlrpc 2023-06-15 09:16:25 UTC Status RELEASE_PENDING CLOSED
Resolution --- ERRATA
Last Closed 2023-06-15 09:16:25 UTC
errata-xmlrpc 2023-06-15 09:17:10 UTC Link ID Red Hat Product Errata RHSA-2023:3623

Back to bug 2149533