Project

General

Profile

Actions
Copy link

Subtask #4092

closed

Feature #4089: rbd: improve small I/O performance

rbd: re-read header when watch is re-established

Added by Josh Durgin about 11 years ago. Updated about 9 years ago.

Status:
Resolved
Priority:
Normal
Assignee:
Jason Dillaman
Target version:
-
% Done:

0%

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

Description

This avoids races that would result in a snapshot not being created correctly, like:

  A               B
  |               | I/O in progress
snap             network fault
notify            |
notify timeout   network restored
snap finished     | more I/O unaware of snap
Actions
Copy link
 #1

Updated by Alex Elder about 11 years ago

This same issue would apply to the kernel rbd client also,
right?

Actions
Copy link
 #2

Updated by Josh Durgin about 11 years ago

Yes, I'll create corresponding tasks for krbd once I get the rest of the general ones in.

Actions
Copy link
 #3

Updated by Jason Dillaman about 9 years ago

  • Status changed from New to In Progress
  • Assignee set to Jason Dillaman
Actions
Copy link
 #4

Updated by Jason Dillaman about 9 years ago

  • Status changed from In Progress to Fix Under Review
Actions
Copy link
 #5

Updated by Jason Dillaman about 9 years ago

  • Status changed from Fix Under Review to Resolved
  • Translation missing: en.field_remaining_hours set to 0.0
Actions
Copy link

Also available in: Atom PDF