Bug 155427 - kernel dm-multipath: How to handle IO barriers?
Summary: kernel dm-multipath: How to handle IO barriers?
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: device-mapper-multipath
Version: rawhide
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
Assignee: Alasdair Kergon
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2005-04-20 08:52 UTC by Lars Marowsky-Bree
Modified: 2007-11-30 22:11 UTC (History)
7 users (show)

Fixed In Version: 2.6.12-rc2-udm1
Clone Of:
Environment:
Last Closed: 2006-03-08 15:27:07 UTC
Type: ---
Embargoed:


Attachments (Terms of Use)
Make dm-multipath report barriers as "not supported" (1.99 KB, patch)
2005-04-20 12:05 UTC, Lars Marowsky-Bree
no flags Details | Diff

Description Lars Marowsky-Bree 2005-04-20 08:52:41 UTC
How should IO barriers be handled - should they be short-circuited and disabled
in the mapping function directly, do we want to support them - which would
probably involve sending them down all paths -, or is there another option...?

I'm tending towards short circuiting them.

Comment 1 Lars Marowsky-Bree 2005-04-20 12:05:52 UTC
Created attachment 113406 [details]
Make dm-multipath report barriers as "not supported"

Comment 2 Alasdair Kergon 2005-04-21 18:16:54 UTC
We need IO barriers which don't carry any data; then we have to send them down
all the paths that might have unsynced IO on them, leading to fun interactions
with priority groups, switch_pg when paths haven't failed etc.

Until then they don't work so not supported.

Comment 3 Alasdair Kergon 2005-04-21 18:29:43 UTC
Added to 2.6.12-rc2-udm1.


Note You need to log in before you can comment on or make changes to this bug.