Back to bug 1959508

Who When What Removed Added
Sunil Kumar Nagaraju 2021-05-11 17:12:19 UTC Keywords Automation
Sunil Kumar Nagaraju 2021-05-12 06:12:44 UTC Target Release 5.1 5.0
Juan Miguel Olmo 2021-05-12 10:02:45 UTC Status NEW ASSIGNED
Veera Raghava Reddy 2021-05-13 13:44:34 UTC CC vereddy
Manasa 2021-05-19 05:46:31 UTC CC mgowri
Juan Miguel Olmo 2021-05-21 12:19:23 UTC Link ID Ceph Project Bug Tracker 50928
Juan Miguel Olmo 2021-05-21 12:31:30 UTC Status ASSIGNED MODIFIED
Link ID Github ceph/ceph/pull/41477
Flags needinfo?(mgowri) needinfo?(mgowri)
Sebastian Wagner 2021-05-25 19:05:10 UTC Target Release 5.0 5.1
CC sewagner
Sebastian Wagner 2021-05-26 11:34:05 UTC Severity high low
Manasa 2021-05-26 12:15:52 UTC Flags needinfo?(mgowri) needinfo?(mgowri)
Red Hat One Jira (issues.redhat.com) 2021-05-27 16:14:10 UTC Link ID Red Hat Issue Tracker RHCEPH-79
Sebastian Wagner 2021-05-31 13:15:55 UTC Priority unspecified medium
Preethi 2021-06-04 05:44:32 UTC CC pnataraj
Doc Type If docs needed, set a value Known Issue
Preethi 2021-06-04 05:45:25 UTC Blocks 1959686
Ranjini M N 2021-06-10 06:44:00 UTC CC jolmomar, rmandyam
Flags needinfo?(jolmomar)
Sunil Kumar Nagaraju 2021-06-14 05:22:06 UTC Flags needinfo?(sewagner) needinfo?(jolmomar)
Ranjini M N 2021-06-15 09:23:29 UTC Flags needinfo?(jolmomar)
Juan Miguel Olmo 2021-06-15 10:03:40 UTC Doc Text Cause: When listing OSD services ( for example with "ceph orch ls" ) the total number of total OSDs that can be created for each OSD service is bigger than the real number.

Consequence: It seems that OSD services are not starting the right number of OSD's , which can be understood as problems starting OSD's.

Ex:
# ceph orch ls
...
osd.all-available-devices 12/16 4m ago 4h *
...

This can drive is to think that there are 4 OSDs (16-12) not started.

Workaround (if any): In order to see if there is a real problem. Just check the ceph health ("ceph -s") to see if ALL the OSD's are up and running.

Result: The list of services will report the wrong total number of OSD's. The "ceph health" command will provide the correct information.
Flags needinfo?(jolmomar) needinfo?(sewagner) needinfo?(jolmomar) needinfo?(jolmomar)
Ranjini M N 2021-06-15 16:07:35 UTC Docs Contact knortema rmandyam
Doc Text Cause: When listing OSD services ( for example with "ceph orch ls" ) the total number of total OSDs that can be created for each OSD service is bigger than the real number.

Consequence: It seems that OSD services are not starting the right number of OSD's , which can be understood as problems starting OSD's.

Ex:
# ceph orch ls
...
osd.all-available-devices 12/16 4m ago 4h *
...

This can drive is to think that there are 4 OSDs (16-12) not started.

Workaround (if any): In order to see if there is a real problem. Just check the ceph health ("ceph -s") to see if ALL the OSD's are up and running.

Result: The list of services will report the wrong total number of OSD's. The "ceph health" command will provide the correct information.
.The `ceph orch ls` command does not list the correct number of OSDs that can be created in the {storage-product} cluster

The command `ceph orch ls` gives the following output:

.Example
-----
# ceph orch ls

osd.all-available-devices 12/16 4m ago 4h *
-----

As per the above output, four OSDs have not started which is not correct.

To workaround this issue, run the `ceph -s` command to see if all the OSDs are up and running in a {storage-product} cluster.
Flags needinfo?(jolmomar)
Juan Miguel Olmo 2021-06-21 07:22:16 UTC Flags needinfo?(jolmomar)
Sebastian Wagner 2021-09-28 14:01:04 UTC Status MODIFIED ASSIGNED
Assignee jolmomar gabrioux
Guillaume Abrioux 2021-10-13 05:46:20 UTC Link ID Github ceph/ceph/pull/43253
Status ASSIGNED POST
Guillaume Abrioux 2021-10-13 05:46:36 UTC Link ID Github ceph/ceph/pull/41477
Ken Dreyer (Red Hat) 2021-10-19 19:21:53 UTC CC kdreyer
errata-xmlrpc 2021-11-27 05:06:03 UTC CC tserlin
Status POST MODIFIED
Fixed In Version ceph-16.2.6-42.el8cp
Status MODIFIED ON_QA
Rahul Lepakshi 2021-12-13 06:14:15 UTC CC rlepaksh
Rahul Lepakshi 2021-12-17 05:01:22 UTC QA Contact vashastr rlepaksh
Sunil Kumar Nagaraju 2021-12-20 11:01:29 UTC Status ON_QA ASSIGNED
Sebastian Wagner 2021-12-20 15:09:57 UTC Assignee gabrioux sewagner
Link ID Github ceph/ceph/pull/44367
Sunil Kumar Nagaraju 2021-12-22 03:09:13 UTC QA Contact rlepaksh sunnagar
Aron Gunn 2022-01-18 21:32:14 UTC Blocks 2031073
CC agunn
Red Hat Bugzilla 2022-01-31 23:32:10 UTC CC sewagner
Assignee sewagner adking
Ranjini M N 2022-02-16 07:45:35 UTC CC adking
Flags needinfo?(adking)
Adam King 2022-02-23 21:57:04 UTC Doc Text .The `ceph orch ls` command does not list the correct number of OSDs that can be created in the {storage-product} cluster

The command `ceph orch ls` gives the following output:

.Example
-----
# ceph orch ls

osd.all-available-devices 12/16 4m ago 4h *
-----

As per the above output, four OSDs have not started which is not correct.

To workaround this issue, run the `ceph -s` command to see if all the OSDs are up and running in a {storage-product} cluster.
In RHCS 5.0, "ceph orch ls" could show an incorrect size for an osd service. For example, an osd service with 5 osds might show something like "5/8" even though there was never intended to be 8 osds attached to this service

In RHCS 5.1 the size for osds now simply displays the number of osds found for the service and no longer guesses how many there should be total. If cephadm sees, for example, 6 osd daemons for the service it will simply say there are 6 osd daemons.
Flags needinfo?(adking)
Doc Type Known Issue Bug Fix
Adam King 2022-02-23 21:57:37 UTC Status ASSIGNED POST
Ranjini M N 2022-02-24 05:50:39 UTC Fixed In Version ceph-16.2.6-42.el8cp ceph-16.2.7-71.el8cp
Status POST MODIFIED
Status MODIFIED ON_QA
Doc Text In RHCS 5.0, "ceph orch ls" could show an incorrect size for an osd service. For example, an osd service with 5 osds might show something like "5/8" even though there was never intended to be 8 osds attached to this service

In RHCS 5.1 the size for osds now simply displays the number of osds found for the service and no longer guesses how many there should be total. If cephadm sees, for example, 6 osd daemons for the service it will simply say there are 6 osd daemons.
.The `ceph orch ls` command now displays the correct size of the Ceph OSDs service

Previously, `ceph orch ls` command would show an incorrect size for Ceph OSD service. For example, an OSD service with five OSDs might show something like `5/8` although eight OSDs were not intended for this service.

With this release, the size displays the number of OSDs found for this service. For example, if there are six OSDs for a service, then `cephadm` displays six daemons.
Flags needinfo?(adking)
Adam King 2022-02-24 06:27:22 UTC Flags needinfo?(adking)
Sunil Kumar Nagaraju 2022-03-01 06:20:35 UTC Status ON_QA ASSIGNED
Flags needinfo?(adking)
Adam King 2022-03-01 13:49:17 UTC Flags needinfo?(adking)
Sunil Kumar Nagaraju 2022-03-02 07:20:38 UTC Status ASSIGNED ON_QA
Sunil Kumar Nagaraju 2022-03-02 07:21:16 UTC Status ON_QA VERIFIED
errata-xmlrpc 2022-04-04 08:01:05 UTC CC nravinas
Status VERIFIED RELEASE_PENDING
errata-xmlrpc 2022-04-04 10:20:39 UTC Resolution --- ERRATA
Status RELEASE_PENDING CLOSED
Last Closed 2022-04-04 10:20:39 UTC
errata-xmlrpc 2022-04-04 10:21:04 UTC Link ID Red Hat Product Errata RHSA-2022:1174

Back to bug 1959508