Project

General

Profile

Actions

Feature #6828

closed

osd should not silently fail to start when journal partition has no UUID

Added by Loïc Dachary over 10 years ago. Updated over 5 years ago.

Status:
Rejected
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:

0%

Source:
other
Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:

Description

Providing a journal that is a partition that does not have the expected journal UUID should trigger an error.

root@bm0014:~/ceph# ceph-deploy --overwrite-conf osd create bm0101.the.re:/dev/sdb1:/dev/sdc1
[ceph_deploy.cli][INFO  ] Invoked (1.3.2): /usr/bin/ceph-deploy --overwrite-conf osd create bm0101.the.re:/dev/sdb1:/dev/sdc1
[ceph_deploy.osd][DEBUG ] Preparing cluster ceph disks bm0101.the.re:/dev/sdb1:/dev/sdc1
[bm0101.the.re][DEBUG ] connected to host: bm0101.the.re 
[bm0101.the.re][DEBUG ] detect platform information from remote host
[bm0101.the.re][DEBUG ] detect machine type
[ceph_deploy.osd][INFO  ] Distro info: Ubuntu 12.04 precise
[ceph_deploy.osd][DEBUG ] Deploying osd to bm0101.the.re
[bm0101.the.re][DEBUG ] write cluster configuration to /etc/ceph/{cluster}.conf
[bm0101.the.re][WARNIN] osd keyring does not exist yet, creating one
[bm0101.the.re][DEBUG ] create a keyring file
[bm0101.the.re][INFO  ] Running command: udevadm trigger --subsystem-match=block --action=add
[ceph_deploy.osd][DEBUG ] Preparing host bm0101.the.re disk /dev/sdb1 journal /dev/sdc1 activate True
[bm0101.the.re][INFO  ] Running command: ceph-disk-prepare --fs-type xfs --cluster ceph -- /dev/sdb1 /dev/sdc1
[bm0101.the.re][ERROR ] WARNING:ceph-disk:OSD will not be hot-swappable if journal is not the same device as the osd data
[bm0101.the.re][DEBUG ] meta-data=/dev/sdb1              isize=2048   agcount=4, agsize=244188660 blks
[bm0101.the.re][DEBUG ]          =                       sectsz=512   attr=2, projid32bit=0
[bm0101.the.re][DEBUG ] data     =                       bsize=4096   blocks=976754637, imaxpct=5
[bm0101.the.re][DEBUG ]          =                       sunit=0      swidth=0 blks
[bm0101.the.re][DEBUG ] naming   =version 2              bsize=4096   ascii-ci=0
[bm0101.the.re][DEBUG ] log      =internal log           bsize=4096   blocks=476930, version=2
[bm0101.the.re][DEBUG ]          =                       sectsz=512   sunit=0 blks, lazy-count=1
[bm0101.the.re][DEBUG ] realtime =none                   extsz=4096   blocks=0, rtextents=0
[bm0101.the.re][INFO  ] Running command: udevadm trigger --subsystem-match=block --action=add
[ceph_deploy.osd][DEBUG ] Host bm0101.the.re is now ready for osd use.

but it's not ;-)

Actions #1

Updated by Loïc Dachary over 10 years ago

I now understand this is not a bug but a feature. It probably deserves a warning of some kind ? http://dachary.org/?p=2548

Actions #2

Updated by Ian Colle over 10 years ago

  • Tracker changed from Bug to Feature
  • Subject changed from osd silently fails to start when journal partition has no UUID to osd should not silently fail to start when journal partition has no UUID
Actions #3

Updated by Ian Colle over 10 years ago

  • Project changed from devops to Ceph
  • Category deleted (ceph-deploy)
Actions #4

Updated by Patrick Donnelly over 5 years ago

  • Status changed from New to Rejected

Stale.

Actions

Also available in: Atom PDF