On Fri, 8 Mar 2002, Neil Brown wrote:
Hi !
I tried latest patch but still no cigar :((
> Thanks. However I still cannot reproduce the problem, or figure out
> any way that it could be happening...
>
I think we should concentrate on fact, that something stays locked. Now
I have raid1 setup and after reboot raid1sync proces is stucked and cat
/proc/mdstat blocks on console in D state.
> Here is another patch that you could try. It incorporates the
> previous patch and adds a bit more locking particularly around the
> list of md devices.
>
I'll try to see if I can find a place where we are locking ourselvs out,
but I have to admint that Your work is quite to advanced for me.
Since I found a host at our clients permises with 2.4.6 kernel and raid5
setup, where no OOPS were reported I started to look when things changed.
I found that in 2.4.9 was introduced change in md_error:
- if (!mark_rdev_faulty(rrdev))
+ if (rrdev->faulty)
return 0;
with mark_rdev_faulty being
-static inline int mark_rdev_faulty (mdk_rdev_t * rdev)
-{
- if (!rdev) {
- MD_BUG();
- return 0;
- }
- if (rdev->faulty)
- return 0;
- free_disk_sb(rdev);
- rdev->faulty = 1;
- return 1;
-}
Now, I know that's not the thing we are looking for now (as it still has
some window for race) it seems to me that something along this line would
be a quick fiks for the problem. I actualyy don't know, if I will be able
do make clean "backout" from this version but will try anyway.
lp
gody
__________________________________________________________________
| Matjaz Godec | Agenda d.o.o. | ISP for business |
| Tech. Manager | Gosposvetska 84 | WAN networks |
| gody@slon.net | si-2000 Maribor | Internet/Intranet |
| tel:+386.2.2340860 | Slovenija | Application servers |
|http://www.slon.net |http://www.agenda.si | Caldera OpenLinux |
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
