Actions
Feature #2548
closedceph-disk-activate: take mount options from ceph.conf
Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:
Updated by Sage Weil almost 12 years ago
- Translation missing: en.field_position set to 2
Updated by Sage Weil almost 12 years ago
- Translation missing: en.field_story_points set to 1
- Translation missing: en.field_position deleted (
10) - Translation missing: en.field_position set to 10
Updated by Sage Weil almost 12 years ago
- Translation missing: en.field_position deleted (
10) - Translation missing: en.field_position set to 8
Updated by Sage Weil almost 12 years ago
- Translation missing: en.field_position deleted (
13) - Translation missing: en.field_position set to 5
Updated by Sage Weil almost 12 years ago
- Translation missing: en.field_position deleted (
21) - Translation missing: en.field_position set to 6
Updated by Anonymous over 11 years ago
Defaults for XFS, from an email from Sage:
When I talked to Christoph at Linuxcon a couple weeks back he suggested
several things that may improve performance on XFS. These are things taht
helped with Sheepdog workloads, which are somewhat similar.
- mount -o inode64 will keep inodes closer to file data for larger disks.
- mount -o filestreams will enable an alternative allocator that works
better for fixed sized files, but it should probably used in conjuntion
with something that explicitly fallocates rbd objects to 4mb. I have a
preliminary branch wip-osd-alloc that puts some of the osd-side bits in
place, but it needs some work, including the rbd bit that adds the
PREALLOC op to the requests.
- mount -o noatime, of course.
Updated by Anonymous over 11 years ago
- Status changed from New to Resolved
In stable as commit 4aa13c6d228b9256870e02a57496f00a75e17349.
In master as commit ab84949da204da4a94473f353b4b0f3cc764e31e.
Actions