Project

General

Profile

Bug #27362

Wrong erasure pool MAX AVAIL size calculation with technique=reed_sol_r6_op

Added by Fyodor Ustinov over 5 years ago. Updated over 5 years ago.

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

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(RADOS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

ceph osd erasure-code-profile set hdderaprof k=4 m=2 crush-failure-domain=rack technique=reed_sol_r6_op crush-device-class=hdd
ceph osd crush rule create-erasure hdderarule hdderaprof

ceph osd erasure-code-profile set hdderaprof2 k=4 m=2 crush-failure-domain=rack crush-device-class=hdd
ceph osd crush rule create-erasure hdderarule2 hdderaprof2

ceph osd pool create rbdn 128 128 erasure hdderaprof hdderarule
ceph osd pool create rbd2 128 128 erasure hdderaprof2 hdderarule2

ceph df
GLOBAL:
    SIZE        AVAIL       RAW USED     %RAW USED
    117 TiB     115 TiB      2.8 TiB          2.36
POOLS:
    NAME        ID     USED     %USED     MAX AVAIL     OBJECTS
    rbdn        8       0 B         0           0 B           0
    rbd2        9       0 B         0        69 TiB           0

If such a pool is used for cephfs - the file system will have a size of 1Z

History

#1 Updated by John Spray over 5 years ago

  • Project changed from Ceph to RADOS
  • Description updated (diff)
  • Category deleted (ceph cli)

Also available in: Atom PDF