RAID5 questions

RAID5 questions

[Don Jessup]

  Is the linux RAID5 optimized for SMP? 
   
   I have RH9.0 kernel 2.4.20-19.9 errate kernel with Large Block Device Patch.
  This works for RAID0 and Linear.  Is there a LBD patch for RAID5?  


   

=====
Don Jessup
Asaca/Shibasoku Corp. of America
400 Corporate Circle, Unit G
Golden, CO  80401
303-278-1111 X232
donj@asaca.com
http://www.asaca.com

Weird Interrupt Problem..

[Gordon Henderson]

Hi,

I doubt this is RAID releated, but maybe some of you fine fellows might
have an insight to a small problem I seem to have...

I've built many servers using Linux RAID and all are working fine, but I
have one wich uses 2 x Promise PCI IDE cards which is giving me some
intersting problems. Kernels from 2.4.20+ac1 to 2.4.22 support the cards:

    Unknown mass storage controller: Promise Technology, Inc. 20269 (rev 2).

and the motherboards boot off it OK. I have 2 cards with 4 IDE drives (one
per master on each controller). Only using the on-board controller for
CD-ROM.

In general, it works, but ... When its heavilly loaded I see processes
hanging in a "D" state - presumably waiting on the hardware somehow.
Sometimes for up to 10 seconds before the appear to resume again. Under
extreme loads up to a dozen processes get stuck - from nfsd to cvs to
mysql to just about anything thats doing a disk access (including the
kernel processes to keep the journal going - it's ext3, but prior to ext3,
it was xfs) The hanging processes seem to come alive again, but it makes
the machine appear very clunky and slow.

I've tried different motherboard, and kernel combinations (from a dual
Xeon PII/500 to dual Athlon XP2200+) and nothing seems to have made any
difference.

Is anyone else using these PCI cards, or can recommend another to use
instead?

Thanks for any advice anyone has - I'm clutching at straws now I think!

Gordon

Re: Weird interrupt problem

[Michel Dänzer]

On Thu, 2002-08-08 at 14:07, Benjamin Herrenschmidt wrote:
>
> >Today this happened to me the second time: All of a sudden, the USB
> >mouse stopped working. At the same time this appeared in the syslog:
> >
> >Aug  7 17:19:57 tibook kernel: Unhandled interrupt 1d, disabled
> >
> >Plugging the mouse to the other socket worked, but it wouldn't work
> >again in the original one. As you can see in the attached
> >/proc/interrupts, interrupt 29 (=1d) is unknown. Interestingly,
> >interrupt 28 seems to be for the USB socket which stopped working (I
> >think it was the same one when the first problem first occured, might be
> >coincidence though). Was it somehow rewarded for its 100'000th
> >occurence? ;) Seriously, is it possible that the number changed somehow?
> >
> >
> >This wouldn't be all that bad, if the whole system wasn't very sluggish
> >afterwards. When it occured the first time, I had CONFIG_TAU enabled in
> >the kernel, which Ben said might cause the sluggishness at least. But
> >now it's disabled.
> >
> >
> >This is on a TiBook III/667 running 2.4.18-ben0-lotsanicestuff if it
> >matters. The first occurence was with 2.4.19-presomething-ben0.
> >
> >
> >Has anyone experienced anything similar?
>
> Really weird. It looks like the openpic is going south. Seriously, looks
> like a HW problem, that is the interrupt controller starts giving us
> bogus interrupt numbers. Doesn't sound good at all :(

Well, now the machine is off to Apple to repair PMU problems. Maybe this
was a kind of an early warning sign.


--
Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member   /  CS student, Free Software enthusiast

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Weird interrupt problem

[Anton Blanchard]

> It may be some kind of race within the openpic driver when we keep
> enabling/disabling the source in ack() and end(). I'm pretty sure
> we don't really need them, oh well...

Yeah I pulled that out of the ppc64 tree and I havent found a problem
yet. Maybe there was some buggy ppc32 hardware that required this?

Anton

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Weird interrupt problem

[Benjamin Herrenschmidt]

>Today this happened to me the second time: All of a sudden, the USB
>mouse stopped working. At the same time this appeared in the syslog:
>
>Aug  7 17:19:57 tibook kernel: Unhandled interrupt 1d, disabled
>
>Plugging the mouse to the other socket worked, but it wouldn't work
>again in the original one. As you can see in the attached
>/proc/interrupts, interrupt 29 (=1d) is unknown. Interestingly,
>interrupt 28 seems to be for the USB socket which stopped working (I
>think it was the same one when the first problem first occured, might be
>coincidence though). Was it somehow rewarded for its 100'000th
>occurence? ;) Seriously, is it possible that the number changed somehow?
>
>
>This wouldn't be all that bad, if the whole system wasn't very sluggish
>afterwards. When it occured the first time, I had CONFIG_TAU enabled in
>the kernel, which Ben said might cause the sluggishness at least. But
>now it's disabled.
>
>
>This is on a TiBook III/667 running 2.4.18-ben0-lotsanicestuff if it
>matters. The first occurence was with 2.4.19-presomething-ben0.
>
>
>Has anyone experienced anything similar?

Really weird. It looks like the openpic is going south. Seriously, looks
like a HW problem, that is the interrupt controller starts giving us
bogus interrupt numbers. Doesn't sound good at all :(

Ideally, when that happens, you could drop into macsbug and peek at
the vector/priority & destination registers for this interrupt source
and tell me what they say. (Hook into the code for the error message
which is in arch/ppc/kernel/irq.c, and call a function you'll write
in open_pic.c that will dump these (to access them, look at what
openpic_enable/disable IRQ does).

It may be some kind of race within the openpic driver when we keep
enabling/disabling the source in ack() and end(). I'm pretty sure
we don't really need them, oh well...

Ben.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Weird interrupt problem

[Michel Dänzer]

On Thu, 2002-08-08 at 02:28, Kevin B. Hendricks wrote:
>
> Twice now I have seen something that may be related.  The whole system
> becomes very sluggish with either one or both cpus consuming 99% of
> capacity while receiving funny interrupt messages.
>
> I literally had to shutdown the system which took forever (I had to keep
> hitting <return> to allow the second cpu to have a chance at things).
>
> This is with an almost stock ben 2.4.19-pre10 kernel compiled for SMP on a
> Dual G4 1-gig machine.  I have never seen anything in the message logs
> that provides a good hint at why this is happening.
>
> Again it has only happended twice now but it is very annoying (the system
> becomes so sluggish you almost think it has hung completely).
>
> I have no idea if this is related but it may be.

I'm not sure, as this really seems to be related to USB, the CPU isn't
hogged and while the system is very jerky, it still runs at about half
the normal speed. No problem shutting down.


--
Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member   /  CS student, Free Software enthusiast


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Weird interrupt problem

[Kevin B. Hendricks]

Hi,

Twice now I have seen something that may be related.  The whole system
becomes very sluggish with either one or both cpus consuming 99% of
capacity while receiving funny interrupt messages.

I literally had to shutdown the system which took forever (I had to keep
hitting <return> to allow the second cpu to have a chance at things).

This is with an almost stock ben 2.4.19-pre10 kernel compiled for SMP on a
Dual G4 1-gig machine.  I have never seen anything in the message logs
that provides a good hint at why this is happening.

Again it has only happended twice now but it is very annoying (the system
becomes so sluggish you almost think it has hung completely).

I have no idea if this is related but it may be.

Kevin

On August 7, 2002 07:05, Michel Dänzer wrote:
> Today this happened to me the second time: All of a sudden, the USB
> mouse stopped working. At the same time this appeared in the syslog:
>
> Aug  7 17:19:57 tibook kernel: Unhandled interrupt 1d, disabled
>
> Plugging the mouse to the other socket worked, but it wouldn't work
> again in the original one. As you can see in the attached
> /proc/interrupts, interrupt 29 (=1d) is unknown. Interestingly,
> interrupt 28 seems to be for the USB socket which stopped working (I
> think it was the same one when the first problem first occured, might be
> coincidence though). Was it somehow rewarded for its 100'000th
> occurence? ;) Seriously, is it possible that the number changed somehow?
>
>
> This wouldn't be all that bad, if the whole system wasn't very sluggish
> afterwards. When it occured the first time, I had CONFIG_TAU enabled in
> the kernel, which Ben said might cause the sluggishness at least. But
> now it's disabled.
>
>
> This is on a TiBook III/667 running 2.4.18-ben0-lotsanicestuff if it
> matters. The first occurence was with 2.4.19-presomething-ben0.
>
>
> Has anyone experienced anything similar?


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: Weird interrupt problem

[Michel Dänzer]

D'oh, sorry for following up to myself...

On Thu, 2002-08-08 at 01:05, Michel Dänzer wrote:
>
> This is on a TiBook III/667 running 2.4.18-ben0-lotsanicestuff if it
> matters.

Make that 2.4.19-ben0-lotsanicestuff.


--
Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member   /  CS student, Free Software enthusiast


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Weird interrupt problem

[Michel Dänzer]

[-- Attachment #1: Type: text/plain, Size: 1269 bytes --]


Today this happened to me the second time: All of a sudden, the USB
mouse stopped working. At the same time this appeared in the syslog:

Aug  7 17:19:57 tibook kernel: Unhandled interrupt 1d, disabled 

Plugging the mouse to the other socket worked, but it wouldn't work
again in the original one. As you can see in the attached
/proc/interrupts, interrupt 29 (=1d) is unknown. Interestingly,
interrupt 28 seems to be for the USB socket which stopped working (I
think it was the same one when the first problem first occured, might be
coincidence though). Was it somehow rewarded for its 100'000th
occurence? ;) Seriously, is it possible that the number changed somehow?


This wouldn't be all that bad, if the whole system wasn't very sluggish
afterwards. When it occured the first time, I had CONFIG_TAU enabled in
the kernel, which Ben said might cause the sluggishness at least. But
now it's disabled.


This is on a TiBook III/667 running 2.4.18-ben0-lotsanicestuff if it
matters. The first occurence was with 2.4.19-presomething-ben0.


Has anyone experienced anything similar?


-- 
Earthling Michel Dänzer (MrCooper)/ Debian GNU/Linux (powerpc) developer
XFree86 and DRI project member   /  CS student, Free Software enthusiast

[-- Attachment #2: interrupts.BAD --]
[-- Type: text/plain, Size: 735 bytes --]

           CPU0       
  1:       1908   OpenPIC   Edge      Built-in Sound out
  2:          0   OpenPIC   Edge      Built-in Sound in
 19:     132189   OpenPIC   Level     ide0
 20:         51   OpenPIC   Level     ide1
 25:    1557896   OpenPIC   Level     VIA-PMU
 26:        226   OpenPIC   Level     keywest i2c
 27:      33611   OpenPIC   Level     usb-ohci
 28:     108865   OpenPIC   Level     usb-ohci
 41:    1800835   OpenPIC   Level     eth0
 42:          0   OpenPIC   Level     keywest i2c
 47:     398650   OpenPIC   Level     GPIO1/ADB
 55:          0   OpenPIC   Edge      NMI - XMON
 57:         19   OpenPIC   Level     Airport
 61:          1   OpenPIC   Edge      Headphone detect
BAD:          1