Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.

This project is now read‑only. Starting Monday, February 2, please use https://ibm-ceph.atlassian.net/ for all bug tracking management.

Bug 1477775

Summary:	ceph-disk: dm-crypt breaks after upgrade to RHCS 2.3
Product:	[Red Hat Storage] Red Hat Ceph Storage	Reporter:	Alexandre Marangone <amarango>
Component:	Ceph-Disk	Assignee:	Loic Dachary <ldachary>
Status:	CLOSED ERRATA	QA Contact:	Vasishta <vashastr>
Severity:	urgent	Docs Contact:	Bara Ancincova <bancinco>
Priority:	urgent
Version:	2.3	CC:	anharris, ceph-qe-bugs, hnallurv, icolle, kchai, kdreyer, ldachary, vashastr
Target Milestone:	rc	Flags:	icolle: needinfo+ icolle: needinfo+
Target Release:	2.4
Hardware:	Unspecified
OS:	Unspecified
Whiteboard:
Fixed In Version:	RHEL: ceph-10.2.7-32.el7cp Ubuntu: ceph_10.2.7-33redhat1	Doc Type:	Bug Fix
Doc Text:	.Encrypted OSDs no longer fail after upgrading to 2.3 Since version 2.3, a test has been added that checks if the `ceph_fsid` file exists inside the `lockbox` directory. If the file does not exist, an attempt to start encrypted OSDs fails. Because previous versions did not include this test, after upgrading to 2.3, the encrypted OSDs failed to start after rebooting. This bug has been fixed, and encrypted OSDs no longer fail after upgrading to version 2.3 or later.	Story Points:	---
Clone Of:		Environment:
Last Closed:	2017-09-05 20:49:52 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:
Bug Depends On:
Bug Blocks:	1439210, 1479701

Description Alexandre Marangone 2017-08-02 22:19:25 UTC

Description of problem:
2.3 added a piece a of code to check for a ceph_fsid file inside the lockbox directory of a dm-crypt OSD. If that ceph_fsid file doesn't exist, it exists and fails.

2.2 did not have that piece of code. After an upgrade to 2.3, rebooting a node will not remap the OSDs. Manually creation the ceph_fsid on each lockbox is the only way to get them up and running.

Comment 2 Ian Colle 2017-08-02 22:47:58 UTC

Need a quick patch to 2.y, so moving from 3.0.

Comment 3 Loic Dachary 2017-08-03 08:50:05 UTC

A quick fix would be to instruct the sysadmin to manually create the ceph_fsid file.

Comment 4 Loic Dachary 2017-08-03 09:00:51 UTC

https://github.com/ceph/ceph/pull/16776 is the proper fix, testing it now

Comment 5 Ian Colle 2017-08-07 13:55:14 UTC

https://github.com/ceph/ceph/pull/16859 Is Jewel backport

Comment 6 Ian Colle 2017-08-07 14:31:31 UTC

Correction to above: https://github.com/ceph/ceph/pull/16870 is correct PR.

Comment 12 Loic Dachary 2017-08-22 20:07:04 UTC

The description is perfect :-)

Comment 19 Loic Dachary 2017-08-29 08:49:35 UTC

This is great news. The text looks good to me !

Comment 22 Vasishta 2017-08-30 13:19:27 UTC

Hi,

We worked on upgrading cluster from 2.2 to 2.4 both on RHEL and Ubuntu, having OSDs with both collocated and dedicated journals and It worked fine.

We executed some smoke tests like rolling reboot of OSD nodes, service restarts after the upgrade  It worked as expected.

Moving the BZ to VERIFIED state.

Regards,
Vasishta

Comment 24 errata-xmlrpc 2017-09-05 20:49:52 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, 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/RHBA-2017:2640