Bug 187534 - Cannot change rr_min_io without rebooting
Summary: Cannot change rr_min_io without rebooting
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 4
Classification: Red Hat
Component: device-mapper-multipath
Version: 4.0
Hardware: All
OS: Linux
medium
medium
Target Milestone: ---
: ---
Assignee: Ben Marzinski
QA Contact:
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2006-03-31 18:35 UTC by Rod Nayfield
Modified: 2010-01-12 02:24 UTC (History)
10 users (show)

Fixed In Version: RHEA-2007-0256
Doc Type: Bug Fix
Doc Text:
Clone Of:
Environment:
Last Closed: 2007-05-01 17:35:44 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Product Errata RHEA-2007:0256 0 normal SHIPPED_LIVE device-mapper-multipath enhancement update 2007-05-01 17:35:42 UTC

Description Rod Nayfield 2006-03-31 18:35:58 UTC
Description of problem:

Cannot change rr_min_io on the fly

Version-Release number of selected component (if applicable):

U3

How reproducible:

Always

Steps to Reproduce:
1. change rr_min_io in multipath.conf and re-run multipath
2. OR reload table with changed weights:
  
[root@clu1 mnt]# dmsetup table mpath2
0 122880000 multipath 0 0 1 1 round-robin 0 2 1 8:32 1000 8:80 1000
[root@clu1 mnt]# echo "0 122880000 multipath 0 0 1 1 round-robin 0 2 1 8:32 100
8:80 100" | dmsetup reload mpath2
[root@clu1 mnt]# dmsetup table mpath2
0 122880000 multipath 0 0 1 1 round-robin 0 2 1 8:32 1000 8:80 1000

Note that the rr is still 1000 1000

Comment 1 Alasdair Kergon 2006-03-31 18:39:15 UTC
reload doesn't change the table until you issue a 'resume'

Comment 4 Rod Nayfield 2006-04-03 16:36:25 UTC
Workaround:

You need to unmount the filesystem, and stop LVM.  Once the map is unused you
can flush it and rebuild it...  

Sample commands:

umount /mnt/TEST/
vgchange -an
multipath -f mpath3
multipath 
vgchange -ay
mount /mnt/TEST/


note you can localize this by giving vgchange an argument to a vg instead of
affecting all vg's on the system...



Comment 5 Ben Marzinski 2006-04-20 19:24:00 UTC
It's possible to fix this to work like change from multibus to failover, but it's
not straightforward.  If people are o.k. with it, I'd rather just document this,
and fix it when we do the next pull from multipath upstream.

Comment 6 Rod Nayfield 2006-04-20 20:11:09 UTC
I think that is fine.

we should make sure the documentation uses vgchange with a named group as
argument so people don't run vgchange -an and lose other volumes.

The other piece we should document is using vgchange -aln with clvm so you can
change the min io's on, say, a gfs volume...



Comment 7 Ben Marzinski 2006-12-04 16:26:33 UTC
You are now able to change rr_min_io just like you can change multibus and failover.

Comment 8 RHEL Program Management 2006-12-04 17:05:14 UTC
This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux maintenance release.  Product Management has requested
further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed
products.  This request is not yet committed for inclusion in an Update
release.

Comment 9 Kiersten (Kerri) Anderson 2007-01-31 18:09:19 UTC
Devel ACK - need QE ACK

Comment 13 Red Hat Bugzilla 2007-05-01 17:35:44 UTC
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.

http://rhn.redhat.com/errata/RHEA-2007-0256.html



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