Re: How to solve mysterious system lockups?

On Tue, 5 Jun 2007, N. Harrington wrote:


--- Garrett Cooper <youshi10@xxxxxxxxxxxxxxxx> wrote:

N. Harrington wrote:
--- Garrett Cooper <youshi10@xxxxxxxxxxxxxxxx>
wrote:


N. Harrington wrote:

Hello
I have several systems that are used as squid
caching servers. I have some systems that use
SCSI
disks and some that use SATA disks. They are
identical in everyway except for the sata vs
SCSI
drives.

At random times, the sata based systems seem to

be

freezing. You can ping them and they respond,
but

you

cannot log in. Nor are any logs processed during

that

time.

I figure it mist be something to do with the

disks,

but I am not sure how to solve it. There seems
to

be

little rhyme or reason. It does not happen

necessarily

during busy times. It can happen in the middle
of

the

night.

Any pointers in how to track down the cause
would

be

much appreciated.

Tyan S2881 Motherboard - 4gigs mem
Using 4 SATA (or scsi) drives
FreeBSD amd64 6.2-STABLE.

Thanks!

Nicole


Nicole,
What's the driver in use for the SATA and the
SCSI drives?
-Garrett


Hi Garret
Here is the driver info.

-- SATA

atapci0: <SiI 3114 SATA150 controller> port


0xbc00-0xbc07,0xb400-0xb403,0xb000-0xb007,0xac00-0xac03,0xa800-0xa80f

mem
0xfeafec00-0xfeafefff irq 17 at device 5.0 on pci3
ata2: <ATA channel 0> on atapci0
ata3: <ATA channel 1> on atapci0
ata4: <ATA channel 2> on atapci0
ata5: <ATA channel 3> on atapci0
pci3: <display, VGA> at device 6.0 (no driver
attached)
isab0: <PCI-ISA bridge> at device 7.0 on pci0
isa0: <ISA bus> on isab0
atapci1: <AMD 8111 UDMA133 controller> port
0x1f0-0x1f7,0x3f6,0x170-0x177,0x376,0xffa0-0xffaf
at
device 7.1 on pci0
ata0: <ATA channel 0> on atapci1
ata1: <ATA channel 1> on atapci1
pci0: <serial bus, SMBus> at device 7.2 (no driver
attached)
pci0: <bridge> at device 7.3 (no driver attached)
pcib2: <ACPI PCI-PCI bridge> at device 10.0 on
pci0
pci2: <ACPI PCI bus> on pcib2

-- SCSI

ahd0: <Adaptec AIC7902 Ultra320 SCSI adapter> port

0x8000-0x80ff,0x7800-0x78ff
mem 0xfc89c000-0xfc89dfff irq 24 at device 10.0 on
pci2
ahd0: [GIANT-LOCKED]
aic7902: Ultra320 Wide Channel A, SCSI Id=7, PCI-X
67-100Mhz, 512 SCBs
ahd1: <Adaptec AIC7902 Ultra320 SCSI adapter> port

0x8800-0x88ff,0x8400-0x84ff
mem 0xfc89e000-0xfc89ffff irq 25 at device 10.1 on
pci2
ahd1: [GIANT-LOCKED]
aic7902: Ultra320 Wide Channel B, SCSI Id=7, PCI-X
67-100Mhz, 512 SCBs
pci0: <base peripheral, interrupt controller> at
device 10.1 (no driver attached)
pcib3: <ACPI PCI-PCI bridge> at device 11.0 on
pci0
pci1: <ACPI PCI bus> on pcib3
pci0: <base peripheral, interrupt controller> at
device 11.1 (no driver attached)



Thanks!

Nicole
Ok, so it's an AMD 8111 northbridge versus an
Adaptec onboard SCSI
controller.

1. What release / version of FreeBSD are you using?
You should upgrade
to 6.2 STABLE because there have been a variety of
issues worked out in previous releases.

I have a range of Versions from 6.1-Pre to 6.2-STABLE
as of a few months ago.

2. Do you have any logs for activity during the
hours when it locks up
(in particular anything interesting / fishy popping
up)?

Nope. That would make it too easy :)
They commit suicide without a note.

3. What scheduler are you using? 4BSD, ULE?

4BSD

4. Does your machine (using the SATA controllers)
lock up under heavy
load? If so, you may have a northbridge cooling
issue that you need to
put a fan on. For instance, the motherboard that I
was using for a while
(ASUS P5N-E SLI) was really close to my CPU
heatsink, and there was a
lot of heat transfer between my northbridge and CPU
heatsink, which was
raising the onboard temperatures 5~10 degrees C. The
new motherboard
(ASUS P5B DLX) doesn't do that though.

The lockups seem rather random. I have healthd
running and they never seem to show very warm. The
room is cold and the servers have great fans. Altho
healthd can seem wonky as the cpu temp has actually
gone below the minimum. Also the -2Volt line seems
very low. But some servers runs forever that way.

At least with SCSI, since it seems to manage itself
as another layer away from the system, you get some
error messages. Sort of like windows 3.1 dropping to
dos. Verses sata issues where it's just blue screen of
death but without even some debugging code.

I am going to try the patch chuck Swiger sent me and
see how that effects things. Also try a few
replacement sata cards. Altho that is always fun
especially in 1U servers. As well as seeing if using
SAS drives may help if I can find some cheap enough.
Do you think that using the ULE scheduler could
really help?

Don't try it in 6.x. It's not stable by any means.

7-CURRENT's getting a lot closer though, especially as of late (past week)..

-Garrett

_______________________________________________
freebsd-questions@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • Re: How to solve mysterious system lockups?
    ... the sata based systems seem to ... SCSI drives? ... pci0: ... room is cold and the servers have great fans. ...
    (freebsd-questions)
  • Re: How to solve mysterious system lockups?
    ... I have some systems that use SCSI ... identical in everyway except for the sata vs SCSI ... FreeBSD-6.x servers running Squid with SATA disks. ...
    (freebsd-questions)
  • Re: How to solve mysterious system lockups?
    ... I have some systems that use SCSI ... disks and some that use SATA disks. ... identical in everyway except for the sata vs SCSI ... I had once experienced the same problem of system lockups in my FreeBSD-6.x servers running Squid. ...
    (freebsd-questions)
  • RE: Are hardware vendors starting to bail on FreeBSD ... ?
    ... All the servers I'm installing this year will have SATA ... Raptor drives are every bit as good as the SCSI drives I used to get. ... It seems SATA is the way to go on small to medium sized servers. ...
    (freebsd-questions)
  • Re: What still uses the block layer?
    ... often _only_ be used by going through the SCSI midlayer. ... That's because modern USB, ATAPI, SATA ... As far as I can tell the hard drives do not have serial numbers easily ...
    (Linux-Kernel)