Project

General

Profile

Bug #44544

VDO, Wrong RAW space calculation after add 512gb of data in pool.

Added by Sergey Ponomarev 12 months ago. Updated 12 months ago.

Status:
New
Priority:
Normal
Assignee:
-
Target version:
Ceph - v14.2.7
% Done:

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
1 - critical
Reviewed:
Affected Versions:
Ceph - v14.2.7
ceph-qa-suite:
Pull request ID:
Crash signature:

Description

if in Ceph with VDO OSD add more than 512Gb of data, RAW space start wrong calculating.
I think in Ceph implemented recalculating pool used space after each 512Gb and after this value RAW space for VDO dont calculate right.

Fedora 31
Ceph version 14.2.7
For mote details see screenshots.
On screenshots 1 OSD with VDO, and Ceph cluster with replication factor 1.

Clean_cluster.JPG View (184 KB) Sergey Ponomarev, 03/10/2020 12:47 PM

Cluster_with_512Gb.JPG View (174 KB) Sergey Ponomarev, 03/10/2020 12:47 PM

Cluster_with_600Gb.JPG View (177 KB) Sergey Ponomarev, 03/10/2020 12:47 PM

History

#1 Updated by Sergey Ponomarev 12 months ago

Cluster filled 100% deduplicated data.
VDO version: 6.2.2.117

Also available in: Atom PDF