SW-RAID 1 and kernel 2.4.18

SW-RAID 1 and kernel 2.4.18

[SCHEP. - Schepke, Arnt]

Hi, i have a running software raid 1. This consist of two ide devices hda
and hdc. The root filesystem is running on this raid array. The harddiscs
are installed in removable frames. 
The system seems to work correctly. But i get some errors. In the file
messages i found: 

linux kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error } 
linux kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }

Are the removables frames my problem? If i remove the frames (connect the
ide direct to the cable) the system runs better. I could not find the
message above. 

If i start a stress test (lot of file copies) the raid array starts
sometimes a resync process. I can't find more information at messages-file. 

Where does this come from? Where i have to look to get information to
prevent this? And at last the system warns me at restart that my
raid-superblocks are not correct. 

How do i resolve this? 
If i connect the two ide devices to the same ide port (hda and hdb used for
raid array) i also get a message about invalid raid superblocks. 

Someone knows something about this?

Regards

Arnt Schepke

Re: SW-RAID 1 and kernel 2.4.18

[Steven Dake]

Schep,

IDE was not designed to hotswap.  The trays are not "hot swappable" in 
that they can only be swapped while the system is off.

For hotswap support, look to SCSI or even better, FibreChannel.  These 
systems are designed to be electrically safe in hotswap operations and 
maintain the integrity of the data transfers.

Thanks
-steve

SCHEP. - Schepke, Arnt wrote:

>Hi, i have a running software raid 1. This consist of two ide devices hda
>and hdc. The root filesystem is running on this raid array. The harddiscs
>are installed in removable frames. 
>The system seems to work correctly. But i get some errors. In the file
>messages i found: 
>
>linux kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error } 
>linux kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
>
>Are the removables frames my problem? If i remove the frames (connect the
>ide direct to the cable) the system runs better. I could not find the
>message above. 
>
>If i start a stress test (lot of file copies) the raid array starts
>sometimes a resync process. I can't find more information at messages-file. 
>
>Where does this come from? Where i have to look to get information to
>prevent this? And at last the system warns me at restart that my
>raid-superblocks are not correct. 
>
>How do i resolve this? 
>If i connect the two ide devices to the same ide port (hda and hdb used for
>raid array) i also get a message about invalid raid superblocks. 
>
>Someone knows something about this?
>
>Regards
>
>Arnt Schepke
>
>
>-
>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
>
>
>
>  
>

Re: SW-RAID 1 and kernel 2.4.18

[H. Peter Anvin]

Followup to:  <3DE3CC8D.2080104@mvista.com>
By author:    Steven Dake <sdake@mvista.com>
In newsgroup: linux.dev.raid
> 
> IDE was not designed to hotswap.  The trays are not "hot swappable" in 
> that they can only be swapped while the system is off.
> 

There are IDE cards which can tristate -- effectively disconnect -- their
outputs.  Some of them can even control the power to the drive.  Such
drives are safe to hotswap, if they have been disconnected before
removing them.  This requires driver support, however, and I'm not
sure if Linux has that.

	-hpa
-- 
<hpa@transmeta.com> at work, <hpa@zytor.com> in private!
"Unix gives you enough rope to shoot yourself in the foot."
http://www.zytor.com/~hpa/puzzle.txt	<amsp@zytor.com>

Re: SW-RAID 1 and kernel 2.4.18

[Alvin Oga]

hi ya

On 26 Nov 2002, H. Peter Anvin wrote:

> Followup to:  <3DE3CC8D.2080104@mvista.com>
> By author:    Steven Dake <sdake@mvista.com>
> In newsgroup: linux.dev.raid
> > 
> > IDE was not designed to hotswap.  The trays are not "hot swappable" in 
> > that they can only be swapped while the system is off.
> > 
> 
> There are IDE cards which can tristate -- effectively disconnect -- their
> outputs.  Some of them can even control the power to the drive.  Such
> drives are safe to hotswap, if they have been disconnected before
> removing them.  This requires driver support, however, and I'm not
> sure if Linux has that.

electrical hotswap is easy to implement ?? ( many different ways )
since one has to make the hotswap tray too ..

i think the problem is after the new drive is inserted, how does
the IDE drivers know to start rewriting all the data to the fresh/virgin
disk .. that was aborted during the time it was previously attempting
to write a 4GB file when the disk was pulled out...
	- if the raid drivers can do all the right magic...
	it should be good to build a nice tri-stated hotwap ide tray

	( to me.. powering up a system with the system power switch or
	( inserting a disk into a properly isolated ide connectors
	( is the same issue... start the motors... than allow the cable
	( connections than the normal system access to the disks

c ya
alvin
http://www.Linux-1U.net
...  9" 1U w/ 2-4 laptop drives
... 26" 1U w/ 8x 180GB drives - 1.6TB - 1U-Raid5[tm]

some IDE trays are bad (was Re: SW-RAID 1 and kernel 2.4.18)

[Louis-David Mitterrand]

On Tue, Nov 26, 2002 at 02:56:41PM +0100, SCHEP. - Schepke, Arnt wrote:
> Hi, i have a running software raid 1. This consist of two ide devices hda
> and hdc. The root filesystem is running on this raid array. The harddiscs
> are installed in removable frames. 
                   ^^^^^^^^^^^^^^^^^

That is the issue.

> The system seems to work correctly. But i get some errors. In the file
> messages i found: 
> 
> linux kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error } 
> linux kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }

I've just had the same problem. My setup: 4 Maxtor 160G connected to a
Promise 133TX2 card, one of them was in an Icy Dock removable rack, the
three remaining were directly connected to the IDE cable. 

The tray-connected disk would always give us the BadCRC error upon
resynching with the raid5 array. That kind of error is typical of bad
cabling.

It seems the additionnal connectors and cable-length in these IDE trays
is too much to bear for picky/sensitive ATA133 drives. In any case the
BadCRC disappeared the moment the drive was directly connected the IDE
ribbon.

hdparm -t /dev/md1 gives me over 100MB/s on IDE raid5. Incredible! Over
twice what a top-of-the-line 15k scsi drive gives me.

-- 
    PHEDRE: Comme il ne respirait qu'une retraite prompte !
            Et combien sa rougeur a redoublé ma honte !
                                          (Phèdre, J-B Racine, acte 3, scène 1)
-
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

Re: some IDE trays are bad (was Re: SW-RAID 1 and kernel 2.4.18)

[Maurice Hilarius]

With regards to your message at 09:01 AM 12/3/02, Louis-David Mitterrand. 
Where you stated:
>On Tue, Nov 26, 2002 at 02:56:41PM +0100, SCHEP. - Schepke, Arnt wrote:
> > Hi, i have a running software raid 1. This consist of two ide devices hda
> > and hdc. The root filesystem is running on this raid array. The harddiscs
> > are installed in removable frames.
>                    ^^^^^^^^^^^^^^^^^
>
>That is the issue.
>
><<snip>>
>I've just had the same problem. My setup: 4 Maxtor 160G connected to a
>Promise 133TX2 card, one of them was in an Icy Dock removable rack, the
>three remaining were directly connected to the IDE cable.
>
>The tray-connected disk would always give us the BadCRC error upon
>resynching with the raid5 array. That kind of error is typical of bad
>cabling.
That type of error is also typical of Promise and other inferior IDE cards.
using IDE ports on motherboards, and using 3Ware cards we rarely see this 
issue.
Using Promise we frequently see it.
Also, cables are important, I agree. Yellow Teflon/TPE cables make a big 
difference.
The cheap grey ones are not suitable for this type of use.



With our best regards,

Maurice W. Hilarius       Telephone: 01-780-456-9771
Hard Data Ltd.               FAX:       01-780-456-9772
11060 - 166 Avenue        mailto:maurice@harddata.com
Edmonton, AB, Canada      http://www.harddata.com/
    T5X 1Y3

Ask me about NAS and near-line storage

Re: some IDE trays are bad (was Re: SW-RAID 1 and kernel 2.4.18)

[Gregory Leblanc]

On Tue, 2002-12-03 at 08:01, Louis-David Mitterrand wrote:
> On Tue, Nov 26, 2002 at 02:56:41PM +0100, SCHEP. - Schepke, Arnt wrote:
> > The system seems to work correctly. But i get some errors. In the file
> > messages i found: 
> > 
> > linux kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error } 
> > linux kernel: hdc: dma_intr: error=0x84 { DriveStatusError BadCRC }
> 
> I've just had the same problem. My setup: 4 Maxtor 160G connected to a
> Promise 133TX2 card, one of them was in an Icy Dock removable rack, the
> three remaining were directly connected to the IDE cable. 
> 
> The tray-connected disk would always give us the BadCRC error upon
> resynching with the raid5 array. That kind of error is typical of bad
> cabling.
> 
> It seems the additionnal connectors and cable-length in these IDE trays
> is too much to bear for picky/sensitive ATA133 drives. In any case the
> BadCRC disappeared the moment the drive was directly connected the IDE
> ribbon.

It's probably far more the quality of the connectors and such than the
length of the cables.  

> hdparm -t /dev/md1 gives me over 100MB/s on IDE raid5. Incredible! Over
> twice what a top-of-the-line 15k scsi drive gives me.

Are you saying that a RAID 5 array gives you better performance than a
single disk?  Make sense to me.  I'd be far more interested in a real
benchmark, rather than hdparm.  Across given runs on a single disk, I've
found it to be reasonably reliable, but from one disk to another, not
necessarily.  
	Greg

-- 
Gregory Leblanc <gleblanc@linuxweasel.com>

Re: some IDE trays are bad (was Re: SW-RAID 1 and kernel 2.4.18)

[Maurice Hilarius]

With regards to your message at 03:10 PM 12/3/02, Gregory Leblanc. Where 
you stated:
>It's probably far more the quality of the connectors and such than the
>length of the cables.
Actually in my experience it is both.
Also the fact that adding more connectors does attenuate a signal.
However we have used the "Icy Dock" trays on occasion,  and I found the 
order of importance for success and failure, in order of impact were:
1) RAID card. HPT were worst, Promise 2nd, and 3Ware best.
2) Cables. The inexpensive grey 80 wire cables are not adequate for this use.
3) Canisters. Icy dock are about typical of any of them. The circuit board 
used to manage the hot-swap does have some impact on the connection, but it 
simply seems to be mostly a matter of more connectors and/or longer cabling 
makes it more sensitive.
Timing issues also creep up.
I have seen a case where a Promise ATA133 card would not negotiate full 
speed with the drives when in a 32 bit, 5V slot, but would succeed when in 
a 64 bit 3.3V slot on the same motherboard.
Even though the cards are only 32 bit.

> > hdparm -t /dev/md1 gives me over 100MB/s on IDE raid5. Incredible! Over
> > twice what a top-of-the-line 15k scsi drive gives me.
I would not consider hdparm to be a reliable test for this type of situation.
bonnie++, using a file size at least 2X real system memory is more 
representative of reality.



With our best regards,

Maurice W. Hilarius       Telephone: 01-780-456-9771
Hard Data Ltd.               FAX:       01-780-456-9772
11060 - 166 Avenue        mailto:maurice@harddata.com
Edmonton, AB, Canada      http://www.harddata.com/
    T5X 1Y3

Ask me about NAS and near-line storage