Re: bus-master DMA error (was: stability of Cobalt Qube 2)

Below are the results:

cobalt# uname -a
NetBSD cobalt 5.1.2 NetBSD 5.1.2 (GENERIC) #0: Thu Feb  2 13:33:26 UTC 2012=
  [email protected]:/home/builds/ab/netbsd-5-1-2-RELEASE/cobalt/20120202=
1012Z-obj/home/builds/ab/netbsd-5-1-2-RELEASE/src/sys/arch/cobalt/compile/G=
ENERIC cobalt

Hmmm, seems he's indeed using the GENERIC kernel. Strange, since I can only=
 find -INSTALL in /ext2/boot/ (but I guess this has nothing to do with it)

cobalt# ls /ext2/boot
boot.gz             netbsd-INSTALL.gz   vmlinux.gz          vmlinux_raq-280=
0.gz
vmlinux-nfsroot.gz  vmlinux_RAQ.gz
cobalt# 


----- Original Message -----
From: "Erik Berls" <[email protected]>
To: "kenneth westelinck" <[email protected]>
Cc: [email protected], [email protected]
Sent: Tuesday, April 24, 2012 11:35:38 AM
Subject: Re: bus-master DMA error (was: stability of Cobalt Qube 2)

On Tue, Apr 24, 2012 at 00:55,  <[email protected]> wrote:
> Hi,
>
> I don't care if it's slow, as long as it works :)
> I installed the Qube using the 5.1.2 restore cd (which is very easy) and =
this installs the netbsd-INSTALL kernel. Could it be that I should be using=
 the netbsd-GENERIC kernel? What kernel (config) are you using, since my se=
tup seems less exotic than yours and yours is working just fine? How can I =
enable the netbsd-GENERIC kernel?

Really?  That seems surprising.  Can you run "uname -a" on the system
and post the results?

-=3Derik.

>
> Sorry for my stupid questions, I am a longtime Linux user taking my first=
 steps in BSD land :(
>
>
> regards,
>
> Kenneth
>
>
> ----- Original Message -----
> From: "Claudio Leiva" <[email protected]>
> To: "kenneth westelinck" <[email protected]>, port-cobalt@net=
bsd.org
> Sent: Tuesday, April 24, 2012 3:34:31 AM
> Subject: RE: stability of Cobalt Qube 2
>
> Hi
>
> I have a Qube 2 which I use as web/ftp/media/samba server with 3 drives (=
10x2.5", 160x2.5" and 750x3.5") and 256 mb of ram running 5.1 generic with =
an aftermarket psu =C2=A0and I can also say that it is stable but very slow=
 on network transfers, I fixed previous stability issues using cron to rebo=
ot the system every month, before that stayed up with maximum uptime of 170=
+ days before it crashed.
>
> Network as I said before is slow, I use the 750 =C2=A0gb as a backup for =
6 computers/laptops and it takes some time but at the end it does his job.
>
> For what I need the qube does a very good job, hope you can fix your issu=
es.
>
> Regards.
>
>
> Claudio Leiva S
> Las Vegas, Nevada
> USA
> http://cleiva.no-ip.com
> (Powered by NetBSD for Cobalt)
>
>
> -----Original Message-----
> From: [email protected] [mailto:[email protected]] =
On Behalf Of [email protected]
> Sent: Monday, April 23, 2012 11:33 AM
> To: [email protected]
> Subject: Re: stability of Cobalt Qube 2
>
> All,
>
> Thank you for your answers. My setup is as follows (you can also see this=
 in images on http://kennywest.blogspot.com/2010/02/story-of-cobalt-qube.ht=
ml):
> - one CF to IDE adapter (containing NetBSD base install), master
> - one 320 GB hard drive containing some Linux partitions, slave
>
> Base install went like a breeze and it boots fine (320 GB drive disconnec=
ted). But, whenever I connect the 320 GB drive, I get this:
> VIA Technologies VT83C572 USB Controller (USB serial bus, revision 0x02) =
at pci0 dev 9 function 2 not configured
> tlp1 at pci0 dev 12 function 0: DECchip 21143 Ethernet, pass 4.1
> tlp1: interrupting at level 2
> tlp1: Ethernet address 00:10:e0:00:3c:5d
> lxtphy1 at tlp1 phy 1: LXT970 10/100 media interface, rev. 3
> lxtphy1: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto
> wd0 at atabus0 drive 0: <SanDisk SDCFB-1024>
> wd0: 977 MB, 1986 cyl, 16 head, 63 sec, 512 bytes/sect x 2001888 sectors
> wd1 at atabus0 drive 1: <WDC WD3200AAJB-56R1A0>
> wd1: 298 GB, 620181 cyl, 16 head, 63 sec, 512 bytes/sect x 625142448 sect=
ors Kernelized RAIDframe activated
> viaide0:0:1: lost interrupt
> =C2=A0 =C2=A0 =C2=A0 =C2=A0type: ata tc_bcount: 512 tc_skip: 0
> viaide0:0:1: bus-master DMA error: missing interrupt, status=3D0x61
> viaide0:0:1: device timeout, c_bcount=3D512, c_skip0
> wd1d: device timeout reading fsbn 0 (wd1 bn 0; cn 0 tn 0 sn 0), retrying
> viaide0 channel 0: reset failed for drive 0 drive 1
> viaide0:0:1: lost interrupt
> =C2=A0 =C2=A0 =C2=A0 =C2=A0type: ata tc_bcount: 512 tc_skip: 0
> viaide0:0:1: bus-master DMA error: missing interrupt, status=3D0x61
> viaide0:0:1: device timeout, c_bcount=3D512, c_skip0
> wd1d: device timeout reading fsbn 0 (wd1 bn 0; cn 0 tn 0 sn 0), retrying
>
> this goes on and after some time it finally boots to a login prompt.
>
> ...
> viaide0:0:1: lost interrupt
> =C2=A0 =C2=A0 =C2=A0 =C2=A0type: ata tc_bcount: 512 tc_skip: 0
> viaide0:0:1: bus-master DMA error: missing interrupt, status=3D0x61
> viaide0:0:1: device timeout, c_bcount=3D512, c_skip0
> wd1d: device timeout reading fsbn 1 (wd1 bn 1; cn 0 tn 0 sn 1), retrying
> viaide0:0:1: lost interrupt
> =C2=A0 =C2=A0 =C2=A0 =C2=A0type: ata tc_bcount: 512 tc_skip: 0
> viaide0:0:1: bus-master DMA error: missing interrupt, status=3D0x61
> viaide0:0:1: device timeout, c_bcount=3D512, c_skip0
> wd1d: device timeout reading fsbn 1 (wd1 bn 1; cn 0 tn 0 sn 1), retrying
> wd1: soft error (corrected)
> wd1: no disk label
> boot device: wd0
>
> Does someone have an idea what this means?
> (please note, above setup boots fine and without any errors in Debian Squ=
eeze)
>
>
> regards,
>
> Kenneth
>
>
>
> ----- Original Message -----
> From: "Andy Ruhl" <[email protected]>
> To: "kenneth westelinck" <[email protected]>
> Cc: [email protected]
> Sent: Monday, April 23, 2012 4:55:13 PM
> Subject: Re: stability of Cobalt Qube 2
>
> On Mon, Apr 23, 2012 at 12:39 AM, =C2=A0<[email protected]> w=
rote:
>> Dear list,
>>
>> I am a proud owner of a Cobalt Qube 2 (http://kennywest.blogspot.com/201=
0/02/story-of-cobalt-qube.html) running Debian Squeeze. I want to use this =
as a NAS device for backing up various Linux machines on the network (I wan=
t to do this using mondo over NFS). Apparently this proves more difficult t=
han I thought (see also: http://lists.debian.org/debian-mips/2012/03/msg000=
15.html). Seems the Qube 2 dies under heavy network load. So whenever mondo=
 tries to upload the restore images to the NFS server, it starts to hang. A=
lso tried using a newer network card (since there is a free PCI slot anyway=
), still the same issue.
>> So, I was wondering, what if I tried NetBSD, would this be more stable? =
Are there people on this list using a Qube 2 reliably?
>
> My experience with the Qube2 is that it's very stable with NetBSD, but no=
t fast for I/O. There was some issue with the PCI bus which impacted perfor=
mance. If you search this list far enough back, you might find the details.=

>
> I used mine for small jobs like DHCP server, internet gateway/firewall, a=
nd some other menial stuff until I got a Seagate Dockstar, which I use now.=
 The Dockstar has it's issues with NetBSD as well.
>
> Andy
>