Note: This is a beta release of Red Hat Bugzilla 5.0. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Also email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback here.

Bug 1362431

Summary: [RFE] distribute OSDs and journal more predictably across the available disks
Product: Red Hat Storage Console Reporter: Daniel Horák <dahorak>
Component: CephAssignee: Shubhendu Tripathi <shtripat>
Status: CLOSED WONTFIX QA Contact: sds-qe-bugs
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 2CC: nthomas, sankarshan
Target Milestone: ---   
Target Release: 3   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-11-19 05:40:07 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:

Description Daniel Horák 2016-08-02 07:51:13 UTC
Description of problem:
  There is some algorithm for distributing OSDs and journals across the available disks described in Bug 1334344.
  
  This algorithm takes care only about the disk type (SSD/HDD) and disk size.
  Then for the same disks with the same size, the journal and OSDs distribution is quite random, which is quite obfuscated and makes any debugging more difficult.

  It would be great if the algorithm will count also with the disk name/order and use disks with the same type/size alphabetically. So for example OSDs will be sorted in ascending order and similarly journal for those OSDs.

Steps to Reproduce:
1. Create cluster with more similar disks peer OSD node (all disks with the same size/type).

Actual results:
  Journals and OSDs are randomly distributed across available disks with the same size.
  The distribution is different between multiple nodes with the same disk configuration or between reinstalled clusters on the same nodes.

Expected results:
  Journal and OSDs will be predictably distributed accross available disks.

Additional info:
  see also: Bug 1334344 - OSDs and journal "randomly" and absurdly distributed across available disks

Comment 2 Shubhendu Tripathi 2018-11-19 05:40:07 UTC
This product is EOL now