Re: Re: patch fr1-2.17 does not work with kernel 2.4.30

On Sun, 10 Apr 2005 15:15:43 +0200 (MET DST)
"Peter T. Breuer" <[email protected]> wrote:

> "Also sprach Stephan von Krawczynski:"
> > Ok, I see. This is only recognisable if you feed the speed_limit_min|max with
> > MAXINT. That's quite ok so far.
> 
> Well, it does whatever the original code would have done, given the
> real figures on i/o speed :-). It's another question if that is sensible
> or not. I've suggested that you try taking out the if
> (md_need_resched(current)) block before the speed calculation, to see
> empirically if it helps (or hinders).

I removed the resched, but that did not gain remarkably.


> All I can imagine is that the event count did not match, but I'd need
> to see more of the kernel messages to know.

Here it comes:

raidsetfaulty /dev/md0 /dev/ndb

Apr 10 15:17:08 mail-a01 kernel: raid1: Disk failure on ndb, disabling device.
Apr 10 15:17:08 mail-a01 kernel:        Operation continuing on 1 devices
Apr 10 15:17:08 mail-a01 kernel: md: recovery thread got woken up ...
Apr 10 15:17:08 mail-a01 kernel: md0: no spare disk to reconstruct array! -- continuing in degraded mode
Apr 10 15:17:08 mail-a01 kernel: md: recovery thread finished ...
Apr 10 15:17:08 mail-a01 kernel: md: updating md0 RAID superblock on device
Apr 10 15:17:08 mail-a01 kernel: md: (skipping faulty ndb )
Apr 10 15:17:08 mail-a01 kernel: md: nda [events: 00000008]<6>(write) nda's sb offset: 25679808

raidhotadd /dev/md0 /dev/ndb

Apr 10 15:17:17 mail-a01 kernel: md: trying to hot-add ndb to md0 ... 
Apr 10 15:17:17 mail-a01 kernel: md0: repair of faulty disk 2b10!
Apr 10 15:17:17 mail-a01 kernel: md: trying to remove ndb from md0 ...
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:2 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:1 dev:ndb
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:0, o:0, n:2 rd:2 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:1 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:0, o:0, n:2 rd:2 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: md0: unnotifying dev 2b10
Apr 10 15:17:17 mail-a01 kernel: ENBD enbd.c #5588[14]: enbd_ioctl received BLKMDUNTFY, now out of raid 900
Apr 10 15:17:17 mail-a01 kernel: md: unbind<ndb,1>
Apr 10 15:17:17 mail-a01 kernel: md: export_rdev(ndb)
Apr 10 15:17:17 mail-a01 kernel: md: updating md0 RAID superblock on device
Apr 10 15:17:17 mail-a01 kernel: md: nda [events: 00000009]<6>(write) nda's sb offset: 25679808
Apr 10 15:17:17 mail-a01 kernel: ENBD enbd.c #5798[18]: enbd_media_changed change ndb requested
Apr 10 15:17:17 mail-a01 kernel: ENBD enbd.c #5805[18]: enbd_media_changed REMOTE CHECK done locally, no remote ioctl!
Apr 10 15:17:17 mail-a01 kernel: ndb (read) [events: 00000007]
Apr 10 15:17:17 mail-a01 kernel: md0: new disk 2b10 too old for repair (disk 7 < bitmap 16)
Apr 10 15:17:17 mail-a01 kernel: md: bind<ndb,2>
Apr 10 15:17:17 mail-a01 kernel: md0: notifying dev 2b10
Apr 10 15:17:17 mail-a01 kernel: ENBD enbd.c #5568[16]: enbd_ioctl received BLKMDNTFY, am now in raid 900
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:1 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:0, o:0, n:2 rd:2 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:2 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:1, o:0, n:2 rd:2 us:1 dev:ndb
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: md0: set repair bit to 0 on superblock
Apr 10 15:17:17 mail-a01 kernel: md: updating md0 RAID superblock on device
Apr 10 15:17:17 mail-a01 kernel: md: ndb [events: 0000000a]<6>(write) ndb's sb offset: 25679808
Apr 10 15:17:17 mail-a01 kernel: md: nda [events: 0000000a]<6>(write) nda's sb offset: 25679808
Apr 10 15:17:17 mail-a01 kernel: md: recovery thread got woken up ...
Apr 10 15:17:17 mail-a01 kernel: md0: resyncing spare disk ndb to replace failed disk
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:2 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:1, o:0, n:2 rd:2 us:1 dev:ndb
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:17:17 mail-a01 kernel:  --- wd:1 rd:2 nd:2 
Apr 10 15:17:17 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:17:17 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel:  disk 2, s:1, o:1, n:2 rd:2 us:1 dev:ndb
Apr 10 15:17:17 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:17:17 mail-a01 kernel: md: syncing RAID array md0
Apr 10 15:17:17 mail-a01 kernel: md: minimum _guaranteed_ reconstruction speed: 2147483647 KB/sec/disc.
Apr 10 15:17:17 mail-a01 kernel: md: using maximum available idle IO bandwith (but not more than 2147483647 KB/sec) for reconstruction.
Apr 10 15:17:17 mail-a01 kernel: md: using 124k window, over a total of 25679808 blocks.
Apr 10 15:17:17 mail-a01 kernel: md0: removed bitmap e605bc80

And then:

Apr 10 15:39:11 mail-a01 kernel: raid1: synced dirty sectors 0-51359615
Apr 10 15:39:11 mail-a01 kernel: md: md0: sync done.
Apr 10 15:39:11 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:39:11 mail-a01 kernel:  --- wd:1 rd:2 nd:2 
Apr 10 15:39:11 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:39:11 mail-a01 kernel:  disk 1, s:0, o:0, n:1 rd:1 us:0 dev:[dev 00:00]
Apr 10 15:39:11 mail-a01 kernel:  disk 2, s:1, o:1, n:2 rd:2 us:1 dev:ndb
Apr 10 15:39:11 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:39:11 mail-a01 kernel: RAID1 conf printout:
Apr 10 15:39:11 mail-a01 kernel:  --- wd:2 rd:2 nd:2 
Apr 10 15:39:11 mail-a01 kernel:  disk 0, s:0, o:1, n:0 rd:0 us:1 dev:nda
Apr 10 15:39:11 mail-a01 kernel:  disk 1, s:0, o:1, n:1 rd:1 us:1 dev:ndb
Apr 10 15:39:11 mail-a01 kernel:  disk 2, s:0, o:0, n:2 rd:2 us:0 dev:[dev 00:00]
Apr 10 15:39:11 mail-a01 kernel:  disk 3, s:0, o:0, n:0 rd:0 us:0 dev:[dev 00:00]
Apr 10 15:39:11 mail-a01 kernel: md: updating md0 RAID superblock on device
Apr 10 15:39:11 mail-a01 kernel: md: ndb [events: 0000000b]<6>(write) ndb's sb offset: 25679808
Apr 10 15:39:11 mail-a01 kernel: md: nda [events: 0000000b]<6>(write) nda's sb offset: 25679808
Apr 10 15:39:11 mail-a01 kernel: md: recovery thread finished ...

As you see: complete resync. The array was brandnew before and just created with

mkraid --dangerous-no-resync --really-force /dev/md0

BTW this is reproducable. If I do the same thing again, it sync completely again, telling me it "removed bitmap XXXXX".
Remember, if I set up the same array _without_ persistent superblock everything is fine and hotadd works as expected.

Regards,
Stephan