troubleshooting page fault/kernel panic

2004-05-16 Thread Streiner, Justin
This is a freebsd 5.2.1 system.  Sometimes, when copying large amounts of
data from the network, the machine will page-fault, spit out some errors
and reboot.

Any insight as to what could be causing this and what I can do to fix
it would be greatly appreciated.  I can also provide more details on the
hardware if needed.

/var/log/messages shows the following from the event that caused the
reboot:

May 16 18:32:20 whammy kernel: Fatal trap 12: page fault while in kernel mode
May 16 18:32:20 whammy kernel: cpuid = 0; apic id = 00
May 16 18:32:20 whammy kernel: fault virtual address= 0xf0db264b
May 16 18:32:20 whammy kernel: fault code   = supervisor read, page not 
present
May 16 18:32:20 whammy kernel: instruction pointer  = 0x8:0xc064f2a7
May 16 18:32:20 whammy kernel: stack pointer= 0x10:0xd77ccc48
May 16 18:32:20 whammy kernel: frame pointer= 0x10:0xd77ccc6c
May 16 18:32:20 whammy kernel: code segment = base 0x0, limit 0xf, 
type 0x1b
May 16 18:32:20 whammy kernel: = DPL 0, pres 1, def32 1, gran 1
May 16 18:32:20 whammy kernel: processor eflags = interrupt enabled, resume, IOPL = 0
May 16 18:32:20 whammy kernel: current process  = 45 (swi7: task queue)
May 16 18:32:20 whammy kernel: trap number  = 12
May 16 18:32:20 whammy kernel: panic: page fault
May 16 18:32:20 whammy kernel: cpuid = 0;
May 16 18:32:20 whammy kernel: boot() called on cpu#0
May 16 18:32:20 whammy kernel:
May 16 18:32:20 whammy kernel: syncing disks, buffers remaining... 3233 3233 3233 3233 
3233 3233 3233 3233 3233 3233 3233 3233 3233 3233 ath0: device timeout
May 16 18:32:20 whammy kernel: 3233 ad1: WARNING - WRITE_DMA interrupt was seen but 
timeout fired LBA=6728498
May 16 18:32:20 whammy kernel: 3233 3233 3233 3233 3233 ad1: WARNING - WRITE_DMA 
interrupt was seen but timeout fired LBA=6728498
May 16 18:32:20 whammy kernel:
May 16 18:32:20 whammy kernel: giving up on 2914 buffers
May 16 18:32:20 whammy kernel: Uptime: 2h24m54s
May 16 18:32:20 whammy kernel: ad1: WARNING - WRITE_DMA interrupt was seen but 
taskqueue stalled LBA=6728498
May 16 18:32:20 whammy kernel: ad0: WARNING - FLUSHCACHE interrupt was seen but 
taskqueue stalled
May 16 18:32:20 whammy kernel: ad1: WARNING - FLUSHCACHE interrupt was seen but 
taskqueue stalled
May 16 18:32:20 whammy kernel: Shutting down ACPI
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


getting Mozilla Firefox port to run / libmap.conf

2004-05-14 Thread Streiner, Justin
I grabbed the current version of Firefox from the ports site and ran the
appropriate portupgrades and installed the linuxflashplugin package to get
what I think are the appropriate version of libpthread.so.

I've also created what I think to be a correct version of /etc/libmap.conf
to allow ld-elf.so to find libpthread.so.1, but it continues to complain
that libpthread.so.1 can't be found.

Below is my /etc/libmap.conf:

libc_r.so.5 libpthread.so.1 # Everything that uses 'libc_r'
libc_r.so   libpthread.so   # now uses 'libpthread'

The library exists in /usr/compat/linux/lib.

-rwxr-xr-x  1 root  0  559279 Nov  5  2003
/usr/compat/linux/lib/libpthread-0.9.so
lrwxr-xr-x  1 root  0  15 May 14 16:15
/usr/compat/linux/lib/libpthread.so - libpthread.so.1
lrwxr-xr-x  1 root  0  17 May  2 13:01
/usr/compat/linux/lib/libpthread.so.0 - libpthread-0.9.so
lrwxr-xr-x  1 root  0  17 May  2 13:04
/usr/compat/linux/lib/libpthread.so.1 - libpthread-0.9.so

When I try to launch firefox, the exact error is:
/libexec/ld-elf.so.1: Shared object libpthread.so.1 not found

What am I doing wrong?

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


READ_BIG errors continue, atapi-cd driver updates?

2004-05-02 Thread Streiner, Justin
When copying data from a CD burned on my Windows machine, I see lots of
errors like the ones at the end of this message continue to pop up in
dmesg.

I saw these on my old machine as well, but was surprised when I continued
to see them on my new machine.

I saw some references to similar errors in other freebsd mailing lists and
there was mention of a patch to atapi-cd.c that worked well for some
people, but the patch for atapi-cd.c doesn't seem to match up with the
atapi-cd.c on my machine.  Has this patch been integrated into the
stock 5.2.1 distribution?  More about the patch in question may be found
at the URL below:

www.atm.tut.fi/list-archive/freebsd-stable/msg02702.html

This machine is running freebsd 5.2.1 - several parts have been
portupgrade'd since the initial OS install.

I have a lot of data to move onto this machine, most of it is currently
stored on burned CDs, so I really need to find a solution to this problem.

jms

vm_fault: pager read error, pid 80512 (cp)
vm_fault: pager read error, pid 80512 (cp)
vm_fault: pager read error, pid 80512 (cp)
acd1: FAILURE - READ_BIG status=51READY,DSC,ERROR sensekey=MEDIUM ERROR
error=4ABORTED
vm_fault: pager read error, pid 80527 (cp)
acd1: FAILURE - READ_BIG status=51READY,DSC,ERROR sensekey=MEDIUM ERROR
error=4ABORTED
vm_fault: pager read error, pid 80527 (cp)
vm_fault: pager read error, pid 80527 (cp)
vm_fault: pager read error, pid 80527 (cp)
acd1: FAILURE - READ_BIG status=51READY,DSC,ERROR sensekey=MEDIUM ERROR
error=4ABORTED
vm_fault: pager read error, pid 80527 (cp)
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Matrox mga_hal driver in 5.2.1?

2004-04-22 Thread Streiner, Justin
Has anyone gotten the mga_hal code to compile cleanly under freebsd
5.2.1/i386?  I downloaded the 'nix distribution for that, as well as the
source for their powerdesk utility from the Matrox website, but haven't
had much luck getting the drivers to compile.

The driver distribution doesn't include an autoconf utility, so I suspect
it may come down to extensive surgery on the Makefile to get it to look
for required headers and libraries in the right places.  I don't know if
the powerdesk distribution is more polished or not - haven't seen a reason
to mess with it until I get the drivers built...

If anyone has specific familiarity with this driver set, please let me
know.  The card I'm using is a Matrox G450 PCI and I'm trying to get the
HAL functionality working so I can do dualhead and/or TV output.

Thanks
jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: Detecting a USB mouse without a restart

2004-04-12 Thread Streiner, Justin
On Mon, 12 Apr 2004, Mazen S. Alzogbi wrote:

 I want to be able to issue commands so my FreeBSD 4.9 can detect my
 just-attached-it USB mouse without restarting the system. The mouse is
 detected and operational if I've attached it before switching on the
 machine.

That may depend on the kind of USB mouse you use.  I have a Logitech
optical USB mouse and a generic 5.2.1 installation picked it up with mo
major issues with I first plugged it in.

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


building new kernel fails during build of 'ath' driver

2004-04-08 Thread Streiner, Justin
version: 5.2.1

I added this driver in my kernel config to natively support my
Atheros-based wireless NIC (Netgear WCG311 I think).  The build fails
immediately after starting to link the necessary objects together.  I
didn't notice any unusual errors during the compile phase.

linking kernel
if_ath.o: In function `ath_attach':
if_ath.o(.text+0x84): undefined reference to `ath_hal_attach'
if_ath.o: In function `ath_tx_start':
if_ath.o(.text+0x3468): undefined reference to `ath_hal_computetxtime'
if_ath.o(.text+0x34e6): undefined reference to `ath_hal_computetxtime'
if_ath.o(.text+0x3514): undefined reference to `ath_hal_computetxtime'
if_ath.o(.text+0x3548): undefined reference to `ath_hal_computetxtime'
if_ath.o: In function `ath_getchannels':
if_ath.o(.text+0x46b3): undefined reference to `ath_hal_init_channels'
if_ath.o(.text+0x4715): undefined reference to `ath_hal_mhz2ieee'
if_ath.o(.data+0x20): undefined reference to `sysctl__hw_ath_children'
if_ath.o(.data+0x60): undefined reference to `sysctl__hw_ath_children'
if_ath.o(.data+0xa0): undefined reference to `sysctl__hw_ath_children'
if_ath.o(.data+0xe0): undefined reference to `sysctl__hw_ath_children'
if_ath.o(.data+0x120): undefined reference to `sysctl__hw_ath_children'
if_ath.o(.data+0x160): more undefined references to
`sysctl__hw_ath_children' follow
if_ath_pci.o: In function `ath_pci_probe':
if_ath_pci.o(.text+0xd4): undefined reference to `ath_hal_probe'
*** Error code 1

Stop in /usr/obj/usr/src/sys/WHAMMY.
*** Error code 1


It looks like the system found a header file that defined the function,
since it compiled if_ath.c in the first place.  Any ideas why it wouldn't
be linking properly?

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


READ_BIG errors on reading data from CD drive

2004-04-06 Thread Streiner, Justin
All:

I've tried mount and mount_cd9660 in my testing of this issue and I run
into the same problems.  I've also tried trying my internal CD drive
(Samsung 140B I borrowed out of another machine) onto my other controller
(ultra ATA133), with no improvements.  In fact when I did that I could
usually get the kernel to panic while copying data from a CD.

I'm really getting to the point where I'm at a loss to explain and fix
this issue.

Could it be a DMA issue of some sort?  The motherboard is an older
PPRO-MMX I salvaged out of another machine.  Maybe some sort of IDE bus
contention issue?  I doubt it since there's not much else on that bus - a
small hard drive and a 3.5 floppy drive.

Any insight anyone could offer would be greatly appreciated.

jms

-- Forwarded message --
Date: Sat, 27 Mar 2004 16:29:39 -0500 (EST)
From: Streiner, Justin [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: data overrun from CD / USB cd mount errors - freebsd 5.2.1

I recently installed freebsd 5.2.1 from CD images that I burned on another
machine.  The CD drive in my new machine worked fine to install the OS and
things are for the most part up and running now.  The issue I'm having now
is in copying more data I have burned on CDs onto the new system.  Sometimes
when I copy a CD, I start getting the following errors:

Mar 22 20:25:33 whammy kernel: acd0: FAILURE - READ_BIG status=51
READY,DSC,ERROR sensekey=MEDIUM ERROR error=4ABORTED

Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
40962048
Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
20480

The files copies to the hard drive are usually blantantly incomplete or
corrupted (file sizes don't match), and in many cases when the file sizes
do match, an MD5 checksum comparison between the original file on the CD
and the file that was copied to the hard drive does not.

This is a pretty normal ATAPI CDROM.  dmesg reports:
acd0: CDROM SAMSUNG SC-140B at ata1-master PIO4

Other posts I've read over on similar subjects suggest that perhaps the DMA
mode for this CDROM needs to be tweaked.  Could this be the case here?

I also have an external USB CD burner that appears to be at least partially
supported by my system, as reported by dmesg:
Mar 22 20:09:05 whammy kernel: GEOM: create disk cd0 dp=0xc1cec600
Mar 22 20:09:05 whammy kernel: cd0 at umass-sim0 bus 0 target 0 lun 0
Mar 22 20:09:05 whammy kernel: cd0: ATAPI CD-R/RW 40X12 3.CB Removable
CD-ROM SCSI-0 device
Mar 22 20:09:05 whammy kernel: cd0: 1.000MB/s transfers
Mar 22 20:09:05 whammy kernel: cd0: cd present [358497 x 2048 byte records]

When I try to mount this device, I get the following errors:
umass0: Unsupported ATAPI command 0x43

I don't think the block device for /dev/cd0 is set up properly, but I can't
seem to build a new one with mknod.

crw-r-  1 root  operator4,  34 Mar 22 20:08 /dev/cd0

Any ideas what may be going on here / how I can fix these issues?

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


data overrun from CD / USB cd mount errors - freebsd 5.2.1 (fwd)

2004-03-31 Thread Streiner, Justin
Is it possible that the way the CD drive gets mounted, or the type of CD
(burned on a winXP machine) could have an impact on the ability to
reliably read from the disc?  If I just do:

mount /cdrom
cp /cdrom/* /data/

I can reproduce the read/overrun errors pretty reliably.

If I do something like pull lots of distributions from the freebsd cd set
I burned from the ISO images, sysinstall never complains about any problems
and the packages are copied and installed cleanly.

Any ideas on what could be causing this?

jms

-- Forwarded message --
Date: Sat, 27 Mar 2004 16:29:39 -0500 (EST)
From: Streiner, Justin [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Subject: data overrun from CD / USB cd mount errors - freebsd 5.2.1

I recently installed freebsd 5.2.1 from CD images that I burned on another
machine.  The CD drive in my new machine worked fine to install the OS and
things are for the most part up and running now.  The issue I'm having now
is in copying more data I have burned on CDs onto the new system.  Sometimes
when I copy a CD, I start getting the following errors:

Mar 22 20:25:33 whammy kernel: acd0: FAILURE - READ_BIG status=51
READY,DSC,ERROR sensekey=MEDIUM ERROR error=4ABORTED

Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
40962048
Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
20480

The files copies to the hard drive are usually blantantly incomplete or
corrupted (file sizes don't match), and in many cases when the file sizes
do match, an MD5 checksum comparison between the original file on the CD
and the file that was copied to the hard drive does not.

This is a pretty normal ATAPI CDROM.  dmesg reports:
acd0: CDROM SAMSUNG SC-140B at ata1-master PIO4

Other posts I've read over on similar subjects suggest that perhaps the DMA
mode for this CDROM needs to be tweaked.  Could this be the case here?

I also have an external USB CD burner that appears to be at least partially
supported by my system, as reported by dmesg:
Mar 22 20:09:05 whammy kernel: GEOM: create disk cd0 dp=0xc1cec600
Mar 22 20:09:05 whammy kernel: cd0 at umass-sim0 bus 0 target 0 lun 0
Mar 22 20:09:05 whammy kernel: cd0: ATAPI CD-R/RW 40X12 3.CB Removable
CD-ROM SCSI-0 device
Mar 22 20:09:05 whammy kernel: cd0: 1.000MB/s transfers
Mar 22 20:09:05 whammy kernel: cd0: cd present [358497 x 2048 byte records]

When I try to mount this device, I get the following errors:
umass0: Unsupported ATAPI command 0x43

I don't think the block device for /dev/cd0 is set up properly, but I can't
seem to build a new one with mknod.

crw-r-  1 root  operator4,  34 Mar 22 20:08 /dev/cd0

Any ideas what may be going on here / how I can fix these issues?

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: I need help badly with freebsd

2004-03-30 Thread Streiner, Justin
On Mon, 29 Mar 2004, Steven Soria wrote:

 I really need help getting FreeBSD. I cant seem to figure out how to
 get FreeBSD. I really want and need this but I need to know how to get
 it. My platform is I386 and please email me witha solution.

There are several ways to get it, depending on what hardware you have.

If you have a machine equipped with a CD burner, you can download the ISO
images for the current FreeBSD i386 distribution, burn them onto CDs
(total of 4 for the most recent 5.2 release), and install from that. Make
sure your CD writer software recognizes that the files you downloaded are
ISO images and not regular data files.

If you don't have a CD burner, your best bet may be to either order the
CDs from freebsdmall.com, or make an appropriate boot/install set on
floppies, then complete the actual installation by downloading the
packages you need from the net.

Instructions for getting and installing FreeBSD are on the website at
www.freebsd.org.

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


Re: good network troubleshooting tool

2004-03-30 Thread Streiner, Justin
On Tue, 30 Mar 2004, Roger 'Rocky' Vetterberg wrote:

 Im looking for suggestions on a good tool to track down packetlosses.
 MTR (/usr/ports/net/mtr) is exactly what I want with one exception:
 MTR uses ICMP, I would like something TCP based.
 I have tried a lot of the utilities in ports but has so far not
 found anything that suits my needs. Does anyone have any suggestions
 on utilities or pointers where to look?

You can try something like lft (layer four traceroute).  This may do what
you're looking for.

http://www.mainnerve.com/lft/

Hope this helps.

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]


data overrun from CD / USB cd mount errors - freebsd 5.2.1

2004-03-27 Thread Streiner, Justin
I recently installed freebsd 5.2.1 from CD images that I burned on another
machine.  The CD drive in my new machine worked fine to install the OS and
things are for the most part up and running now.  The issue I'm having now
is in copying more data I have burned on CDs onto the new system.  Sometimes
when I copy a CD, I start getting the following errors:

Mar 22 20:25:33 whammy kernel: acd0: FAILURE - READ_BIG status=51
READY,DSC,ERROR sensekey=MEDIUM ERROR error=4ABORTED

Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
40962048
Mar 23 15:47:27 whammy kernel: acd0: WARNING - READ_BIG read data overrun
20480

The files copies to the hard drive are usually blantantly incomplete or
corrupted (file sizes don't match), and in many cases when the file sizes
do match, an MD5 checksum comparison between the original file on the CD
and the file that was copied to the hard drive does not.

This is a pretty normal ATAPI CDROM.  dmesg reports:
acd0: CDROM SAMSUNG SC-140B at ata1-master PIO4

Other posts I've read over on similar subjects suggest that perhaps the DMA
mode for this CDROM needs to be tweaked.  Could this be the case here?

I also have an external USB CD burner that appears to be at least partially
supported by my system, as reported by dmesg:
Mar 22 20:09:05 whammy kernel: GEOM: create disk cd0 dp=0xc1cec600
Mar 22 20:09:05 whammy kernel: cd0 at umass-sim0 bus 0 target 0 lun 0
Mar 22 20:09:05 whammy kernel: cd0: ATAPI CD-R/RW 40X12 3.CB Removable
CD-ROM SCSI-0 device
Mar 22 20:09:05 whammy kernel: cd0: 1.000MB/s transfers
Mar 22 20:09:05 whammy kernel: cd0: cd present [358497 x 2048 byte records]

When I try to mount this device, I get the following errors:
umass0: Unsupported ATAPI command 0x43

I don't think the block device for /dev/cd0 is set up properly, but I can't
seem to build a new one with mknod.

crw-r-  1 root  operator4,  34 Mar 22 20:08 /dev/cd0

Any ideas what may be going on here / how I can fix these issues?

jms
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to [EMAIL PROTECTED]