Bug 66643

Summary:	[2.4] evms2 soft raid5 sync block system with >=gentoo-sources-2.4.26-r5
Product:	Gentoo Linux	Reporter:	Jules Gagnon <eonwe>
Component:	[OLD] Core system	Assignee:	Gentoo Kernel Bug Wranglers and Kernel Maintainers <kernel>
Status:	RESOLVED NEEDINFO
Severity:	normal	CC:	kernel
Priority:	High
Version:	unspecified
Hardware:	x86
OS:	Linux
Whiteboard:
Package list:		Runtime testing required:	---
Attachments:	.config for 2.4.26-r[59]

Description Jules Gagnon 2004-10-07 06:37:09 UTC

I use evms2 with a 40G (3x20G) software raid5 array with ide disks. 
When I have to hit reset, my array is synced at boot.

With 2.4.25-r9, this syncing is done in background and I can use the systems almost normally.

But 2.4.26-r[59], the system boot stop at activating EVMS2 until the sync is completed (about half an hour).

I also noticed the same problem when replacing a faulty drive. I was unable to access any of the data stored in the array (/usr, /home) until the array reconstruction was completed. So it is not related to the init script.


Reproducible: Always
Steps to Reproduce:
1. Create a raid5 array (using evms)
2. Hit reset


Actual Results:  
the boot process stop at activating EVMS2 until the sync is completed (about 
half an hour) 

Expected Results:  
syncing should be done in background

Comment 1 Jules Gagnon 2004-10-07 06:38:50 UTC

Created attachment 41286 [details]
.config for 2.4.26-r[59]

Comment 2 Tim Yamin (RETIRED) gentoo-dev

2004-12-22 16:19:31 UTC

Hrm, not sure if this is due to a change in the EVMS patches. If you grab a vanilla kernel and apply the latest EVMS patches do you also get this behaviour?

Comment 3 Jules Gagnon 2004-12-31 08:05:14 UTC

I'm not sure if it's exactly what you wanted but I applied the lastest dm patch on a vanilla-2.4.26, I have reset the machine and I got no syncing delay. I used the same config as posted.

To me it seems to be a kind of conflict between gentoo patches and evms (or soft raid).

Anything else you want me to test?

Comment 4 Jules Gagnon 2005-02-19 07:24:08 UTC

I tested with gentoo-sources=2.4.28-r7 and the same problem exists

Comment 5 Jules Gagnon 2005-03-18 12:04:49 UTC

Same problem with gentoo-sources=2.4.28-r8 and I changed motherboard (different chipset and ide controller), so it's definitly not hardware related.

Comment 6 Tim Yamin (RETIRED) gentoo-dev

2005-07-20 08:14:23 UTC

Want to see if this is an issue with 2.4.31-r1? If so, the device-mapper patch
may need updating.

Comment 7 Daniel Drake (RETIRED) gentoo-dev

2005-08-06 05:05:48 UTC

Please reopen if this is still an issue on gentoo-sources-2.4.31