5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Norman White]

We have a 15 drive addonics array with 3 5 port sata multiplexors, one 
of the sas cables was knocked out to one of the port multiplexors and now
mdadm sees 9 drives , a spare, and 5 failed, removed drives (after 
fixing the cabling problem).

A mdadm -E on each of the drives, see 5 drives (the ones that were 
uncabled) as seeing the original  configuration with 14 drives and a 
spare, while the other 10 drives report
9 drives, a spare and 5 failed , removed drives.

We are very confident that there was no io going on at the time, but are 
not sure how to proceed.

One obvious thing to do is to just do a:

mdadm --assemble --force --assume-clean /dev/md0 sd[b,c, ... , p]
but we are getting different advice about what force will do in this 
situation. The last thing we want to do is wipe the array.

Another option would be to fiddle with the super blocks with mddump, so 
that they all see the same 15 drives in the same configuration, and then 
assemble it.

Yet another suggestion was to recreate the array configuration and hope 
that the data wouldn't be touched.

And even another suggestion is to create the array with one drive 
missing (so it is degraded and won't rebuild)

Any pointers on how to proceed would be helpful. Restoring 30TB takes 
along time.

Best,
Norman White

Re: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Stan Hoeppner]

Norman White put forth on 9/8/2010 12:22 PM:
> We have a 15 drive addonics array with 3 5 port sata multiplexors, one
> of the sas cables was knocked out to one of the port multiplexors...

Exactly how/why was the SAS cable "knocked out".  Is this racked gear?
Fix that problem before worrying about the array state, as you may find
yourself here again in the future.

Bringing the array back up with guess work vs starting a restore now, if
you blow the array trying to fix it with mdadm hacks, you can always
still run the restore later.  Just make sure you do a lot of
verification after you bring the arrays back up and they _seem_ fine.
Don't trust surface appearances, esp when talking about 30TB of data.

Just my $0.02.

-- 
Stan

Re: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Mikael Abrahamsson]

On Wed, 8 Sep 2010, Norman White wrote:

> Any pointers on how to proceed would be helpful. Restoring 30TB takes 
> along time.

Look into the archives of this list, this has happened to numerous people 
and most have gotten out of it ok.

Make sure you have a recent kernel and the latest mdadm.

At this point I've several times successfully done --assemble --force in 
your situation with good result.

--create --assume-clean is the absolute last resort when everything else 
fails, is my recommendation.

-- 
Mikael Abrahamsson    email: swmike@swm.pp.se

Re: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Neil Brown]

On Wed, 08 Sep 2010 13:22:30 -0400
Norman White <nwhite@stern.nyu.edu> wrote:

> We have a 15 drive addonics array with 3 5 port sata multiplexors, one 
> of the sas cables was knocked out to one of the port multiplexors and now
> mdadm sees 9 drives , a spare, and 5 failed, removed drives (after 
> fixing the cabling problem).
> 
> A mdadm -E on each of the drives, see 5 drives (the ones that were 
> uncabled) as seeing the original  configuration with 14 drives and a 
> spare, while the other 10 drives report
> 9 drives, a spare and 5 failed , removed drives.
> 
> We are very confident that there was no io going on at the time, but are 
> not sure how to proceed.
> 
> One obvious thing to do is to just do a:
> 
> mdadm --assemble --force --assume-clean /dev/md0 sd[b,c, ... , p]
> but we are getting different advice about what force will do in this 
> situation. The last thing we want to do is wipe the array.

What sort of different advice?  From whom?

This should either do exactly what you want, or nothing at all.  I suspect
the former.  To be more confident I would need to see the output of
   mdadm -E /dev/sd[b-p]

NeilBrown


> 
> Another option would be to fiddle with the super blocks with mddump, so 
> that they all see the same 15 drives in the same configuration, and then 
> assemble it.
> 
> Yet another suggestion was to recreate the array configuration and hope 
> that the data wouldn't be touched.
> 
> And even another suggestion is to create the array with one drive 
> missing (so it is degraded and won't rebuild)
> 
> Any pointers on how to proceed would be helpful. Restoring 30TB takes 
> along time.
> 
> Best,
> Norman White
> --
> 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: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Norman White]

  On 9/8/2010 5:35 PM, Neil Brown wrote:
> On Wed, 08 Sep 2010 13:22:30 -0400
> Norman White<nwhite@stern.nyu.edu>  wrote:
>
>> We have a 15 drive addonics array with 3 5 port sata multiplexors, one
>> of the sas cables was knocked out to one of the port multiplexors and now
>> mdadm sees 9 drives , a spare, and 5 failed, removed drives (after
>> fixing the cabling problem).
>>
>> A mdadm -E on each of the drives, see 5 drives (the ones that were
>> uncabled) as seeing the original  configuration with 14 drives and a
>> spare, while the other 10 drives report
>> 9 drives, a spare and 5 failed , removed drives.
>>
>> We are very confident that there was no io going on at the time, but are
>> not sure how to proceed.
>>
>> One obvious thing to do is to just do a:
>>
>> mdadm --assemble --force --assume-clean /dev/md0 sd[b,c, ... , p]
>> but we are getting different advice about what force will do in this
>> situation. The last thing we want to do is wipe the array.
> What sort of different advice?  From whom?
>
> This should either do exactly what you want, or nothing at all.  I suspect
> the former.  To be more confident I would need to see the output of
>     mdadm -E /dev/sd[b-p]
>
> NeilBrown
>
>


Just to close this out, I sent Neil Brown the output of mdadm -E 
/dev/sd[b-p]
and he agreed it looked clean.

I then did an
mdadm --assemble --force /dev/md0 sd[b-p]

and got the message /dev/sdb was busy, no super block.

Rebooted the system, and reissued the mdadm --assemble --force.

Voila,
/dev/md0 was back.. Initial tests indicate no data loss.

We have, of course, (as suggested by some on this list), more securely 
attached the sas
cables to the back of the addonics array so this can't happen again. The 
siIicon image
port multiplers only seem to have  push in connections that don't lock 
at all. Just a pressure fit. We
have to be very careful working around the box.

On the other hand, we have a 30TB raid 6 array (about 21 TB formatted 
with a hot spare) that is extremely fast and inexpensive. (~ $4k ).
We are considering buying another and having a dedicated server  with 
several  arrays connected to it and put in a protected environment.

Thank you very much Neil.

We owe you.

Best,
Norman White
>> Another option would be to fiddle with the super blocks with mddump, so
>> that they all see the same 15 drives in the same configuration, and then
>> assemble it.
>>
>> Yet another suggestion was to recreate the array configuration and hope
>> that the data wouldn't be touched.
>>
>> And even another suggestion is to create the array with one drive
>> missing (so it is degraded and won't rebuild)
>>
>> Any pointers on how to proceed would be helpful. Restoring 30TB takes
>> along time.
>>
>> Best,
>> Norman White
>> --
>> 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: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Mr. James W. Laferriere]

 	Hello Norman ,

On Fri, 10 Sep 2010, Norman White wrote:
> On 9/8/2010 5:35 PM, Neil Brown wrote:
>> On Wed, 08 Sep 2010 13:22:30 -0400
>> Norman White<nwhite@stern.nyu.edu>  wrote:
>> 
...snip...
>
> On the other hand, we have a 30TB raid 6 array (about 21 TB formatted with a 
> hot spare) that is extremely fast and inexpensive. (~ $4k ).
> We are considering buying another and having a dedicated server  with several 
> arrays connected to it and put in a protected environment.
>
 	Would you please elaborate on the construction of this array ?  Or if 
purchased it's manufacturer & product number ?
 	I find such a device very interesting .  Maybe many others will as well 
.
 		Tia ,  JimL
-- 
+------------------------------------------------------------------+
| James   W.   Laferriere | System    Techniques | Give me VMS     |
| Network&System Engineer | 3237     Holden Road |  Give me Linux  |
| babydr@baby-dragons.com | Fairbanks, AK. 99709 |   only  on  AXP |
+------------------------------------------------------------------+

Re: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Norman White]

  On 9/10/2010 1:47 PM, Mr. James W. Laferriere wrote:
>     Hello Norman ,
>
> On Fri, 10 Sep 2010, Norman White wrote:
>> On 9/8/2010 5:35 PM, Neil Brown wrote:
>>> On Wed, 08 Sep 2010 13:22:30 -0400
>>> Norman White<nwhite@stern.nyu.edu>  wrote:
>>>
> ...snip...
>>
>> On the other hand, we have a 30TB raid 6 array (about 21 TB formatted 
>> with a hot spare) that is extremely fast and inexpensive. (~ $4k ).
>> We are considering buying another and having a dedicated server  with 
>> several arrays connected to it and put in a protected environment.
>>
>     Would you please elaborate on the construction of this array ?  Or 
> if purchased it's manufacturer & product number ?
>     I find such a device very interesting .  Maybe many others will as 
> well .
>         Tia ,  JimL
You can find more detail on the construction of the array, costs, 
components, performance, issues etc on my blog at

http://researchcomputing.blogspot.com/

Best,
Norman White

Re: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[CoolCold]

Neil can you share that decision making algorithm ?
We have servers with "lucky" aic9410 & LSI 1068E controllers which
hang system sometimes, then drive(s) are dropped.
In simple cases, when one drive has different Events count it's enough
to force assemble, but in other cases when, say 8 drives are dropped
like in Kyler's case (
http://marc.info/?l=linux-raid&m=127534131202696&w=2 ) and examine
shows different info for drives from the same array, ie
http://lairds.us/temp/ucmeng_md/20100526/examine_sdj1 ,
http://lairds.us/temp/ucmeng_md/20100526/examine_sda1 . Keeping in
mind that drives can be exported in different order on every boot,
it's not so straightforward to detect "right" options.

I promise to put that knowledge on wiki.

On Thu, Sep 9, 2010 at 1:35 AM, Neil Brown <neilb@suse.de> wrote:
> On Wed, 08 Sep 2010 13:22:30 -0400
> Norman White <nwhite@stern.nyu.edu> wrote:
>
>> We have a 15 drive addonics array with 3 5 port sata multiplexors, one
>> of the sas cables was knocked out to one of the port multiplexors and now
>> mdadm sees 9 drives , a spare, and 5 failed, removed drives (after
>> fixing the cabling problem).
>>
>> A mdadm -E on each of the drives, see 5 drives (the ones that were
>> uncabled) as seeing the original  configuration with 14 drives and a
>> spare, while the other 10 drives report
>> 9 drives, a spare and 5 failed , removed drives.
>>
>> We are very confident that there was no io going on at the time, but are
>> not sure how to proceed.
>>
>> One obvious thing to do is to just do a:
>>
>> mdadm --assemble --force --assume-clean /dev/md0 sd[b,c, ... , p]
>> but we are getting different advice about what force will do in this
>> situation. The last thing we want to do is wipe the array.
>
> What sort of different advice?  From whom?
>
> This should either do exactly what you want, or nothing at all.  I suspect
> the former.  To be more confident I would need to see the output of
>   mdadm -E /dev/sd[b-p]
>
> NeilBrown
>
>
>>
>> Another option would be to fiddle with the super blocks with mddump, so
>> that they all see the same 15 drives in the same configuration, and then
>> assemble it.
>>
>> Yet another suggestion was to recreate the array configuration and hope
>> that the data wouldn't be touched.
>>
>> And even another suggestion is to create the array with one drive
>> missing (so it is degraded and won't rebuild)
>>
>> Any pointers on how to proceed would be helpful. Restoring 30TB takes
>> along time.
>>
>> Best,
>> Norman White
>> --
>> 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
>
> --
> 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
>



-- 
Best regards,
[COOLCOLD-RIPN]
--
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: 5 drives lost in an inactive 15 drive raid 6 system due to cable problem - how to recover?

[Neil Brown]

On Fri, 10 Sep 2010 23:39:08 +0400
CoolCold <coolthecold@gmail.com> wrote:

> Neil can you share that decision making algorithm ?

Nothing special really.  I just confirmed that the metadata contents agreed
with what Norman said had happened.  Sometime people do other things to
arrays first and don't realise the consequences and don't report them.  So if
I based my "just use --force, that should work" on what they tell me I am
sometimes wrong - because what they tell me is flawed.  So I am not more
cautious and try to only base it on what mdadm tells me.

And I didn't really need to see the mdadm output.  As I said before
mdadm -Af should either do the right thing or nothing.  But I know people can
be very protective of there multi-gigabyte data sets so if I can be a bit
more definite, it helps them.

Also, I like to see want mdadm -E output from odd failures as it might show
me something that needs fixing.  Largely as a consequence of watching how
people interact with their failed arrays, the next release of mdadm will be a
lot more cautious about letting you add a device to an array - particularly a
failed array.  People often seem to do this thinking it means 'add this back
in', but really it means 'make this a spare and attach it to the array'.

In this particular case, I checked that what each device thought of its own
role in the array was compatible with with what the newest device thought,
where 'compatible' means either they agreed, or the newer reported a clean
failure.  If any device thought it was a spare and shouldn't have done, or
two devices both thought they filled the same roles, that would have been a
concern.

NeilBrown



> We have servers with "lucky" aic9410 & LSI 1068E controllers which
> hang system sometimes, then drive(s) are dropped.
> In simple cases, when one drive has different Events count it's enough
> to force assemble, but in other cases when, say 8 drives are dropped
> like in Kyler's case (
> http://marc.info/?l=linux-raid&m=127534131202696&w=2 ) and examine
> shows different info for drives from the same array, ie
> http://lairds.us/temp/ucmeng_md/20100526/examine_sdj1 ,
> http://lairds.us/temp/ucmeng_md/20100526/examine_sda1 . Keeping in
> mind that drives can be exported in different order on every boot,
> it's not so straightforward to detect "right" options.
> 
> I promise to put that knowledge on wiki.
> 
> On Thu, Sep 9, 2010 at 1:35 AM, Neil Brown <neilb@suse.de> wrote:
> > On Wed, 08 Sep 2010 13:22:30 -0400
> > Norman White <nwhite@stern.nyu.edu> wrote:
> >
> >> We have a 15 drive addonics array with 3 5 port sata multiplexors, one
> >> of the sas cables was knocked out to one of the port multiplexors and now
> >> mdadm sees 9 drives , a spare, and 5 failed, removed drives (after
> >> fixing the cabling problem).
> >>
> >> A mdadm -E on each of the drives, see 5 drives (the ones that were
> >> uncabled) as seeing the original  configuration with 14 drives and a
> >> spare, while the other 10 drives report
> >> 9 drives, a spare and 5 failed , removed drives.
> >>
> >> We are very confident that there was no io going on at the time, but are
> >> not sure how to proceed.
> >>
> >> One obvious thing to do is to just do a:
> >>
> >> mdadm --assemble --force --assume-clean /dev/md0 sd[b,c, ... , p]
> >> but we are getting different advice about what force will do in this
> >> situation. The last thing we want to do is wipe the array.
> >
> > What sort of different advice?  From whom?
> >
> > This should either do exactly what you want, or nothing at all.  I suspect
> > the former.  To be more confident I would need to see the output of
> >   mdadm -E /dev/sd[b-p]
> >
> > NeilBrown
> >
> >
> >>
> >> Another option would be to fiddle with the super blocks with mddump, so
> >> that they all see the same 15 drives in the same configuration, and then
> >> assemble it.
> >>
> >> Yet another suggestion was to recreate the array configuration and hope
> >> that the data wouldn't be touched.
> >>
> >> And even another suggestion is to create the array with one drive
> >> missing (so it is degraded and won't rebuild)
> >>
> >> Any pointers on how to proceed would be helpful. Restoring 30TB takes
> >> along time.
> >>
> >> Best,
> >> Norman White
> >> --
> >> 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
> >
> > --
> > 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
> >
> 
> 
> 

--
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