Kernel Panic

LostInIgnorance

CRAP!!! It faulted out again (Soekris):

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address   = 0x0
fault code              = supervisor read, page not present
instruction pointer     = 0x20:0x0
stack pointer           = 0x28:0xd553ebc0
frame pointer           = 0x28:0xd553ebcc
code segment            = base 0x0, limit 0xfffff, type 0x1b
                        = DPL 0, pres 1, def32 1, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 12 (irq5: vr1)
[thread]
Stopped at      0:      *** error reading from address 0 ***
db> bt
Tracing pid 12 tid 64026 td 0xc3aa4780
m_tag_delete(c4dde600,c4de0b3d) at m_tag_delete+0x32
m_tag_delete_chain(c4dde600,0,d553ec2c,c0c94f99,c4dde600,...) at m_tag_delete_chain+0x82
mb_dtor_mbuf(c4dde600,100,0,c0a9041f,c3a92000,...) at mb_dtor_mbuf+0x25
uma_zfree_arg(c1d7b000,c4dde600,0,c3a28000,d553ec70,...) at uma_zfree_arg+0x29
m_freem(c4dde600,c3debe00,c3a92000,7f,7f,...) at m_freem+0x43
vr_txeof(0,c12c69c0,d5530008,0,0,...) at vr_txeof+0x340
vr_intr(c3a28000,0,109,dbd32e12,b15,...) at vr_intr+0x1c7
intr_event_execute_handlers(c39757f8,c3972180,c0e6ba48,52d,c39721f0,...) at intr_event_execute_handlers+0x14b
ithread_loop(c3a8e450,d553ed38,6000,f600001b,0,...) at ithread_loop+0x6b
fork_exit(c09ae910,c3a8e450,d553ed38) at fork_exit+0x91
fork_trampoline() at fork_trampoline+0x8
--- trap 0, eip = 0, esp = 0xd553ed70, ebp = 0 ---
db>

At the time I was logged on with 2 clients.  One (my netbook) was pushing pandora radio, and the other (my lappy) was working on configuration on the local network via remote desktop.  The desktop handshake for security (kerbros) is when it crashed.[/thread]

xbipin

i have started facing this on my alix now using nanobsd, problem is when it crashes, webgui wont work nor does the serial console show anything so i have to simply power off and on again.

Slaygon

I can confirm the freezing on the latest build.
No indication on the console, nothing in logs, no crash dumps, nada. It's just stone dead.

Only thing you can do is to power-cycle the machine.

xbipin

so i guess im not the only one, im using the 26th jan snap and one way i could replicate is as follows:

alix booted and running and pppoe conencted, now i went to status -> interfaces -> wan interface and clicked disconnect and after some time it auto reconnects and now after say about less than 30mins and the system is dead.

eri--

xibpin and others please upgrade to the 27-JAN snapshot when its out.

Please do not mess this thread because it is because of another specific issue not related to the hangs which should be fixed on a 27-JAN snap.

Slaygon

@ermal:

xibpin and others please upgrade to the 27-JAN snapshot when its out.

Please do not mess this thread because it is because of another specific issue not related to the hangs which should be fixed on a 27-JAN snap.

The one that froze on me most recently was the 27-JAN snapshot ("2.0-BETA5 (amd64) built on Thu Jan 27 01:29:01 EST 2011").

LostInIgnorance

Should I just wait till the snap comes out? I never did get an answer concerning the test/crash last night loading the stuff ermal provided. Worked great, as you can see, for quite some time.

vito

ermal,
My openvpn connection is still up from last night.
Copied a good amount of data, no panics

So i know going forward,
Do i need to apply the new kernel with the newer snaps?

At what point will the new snaps contain the new working kernel?

Thanks again for all your help. (and of course jimp also) :)

ftcsistemi

Hello, I tried the latest snap and the latest kernel but the problem persists

jimp

@ftcsistemi:

Hi ! i jast try with latest snap and latest kernel but he problem persist.

(to everyone) Please don't just say "latest kernel" - give a timestamp, be as precise as possible.

ftcsistemi

@ermal:

Uploaded a new kernel http://files.pfsense.org/kernel_new.gz
Beaware that you need to be updated to the latest snapshot before using this kernel otherwise you will get hangs.

the kernel is in the quote, and the snap is the latest at this time in the update manager

jimp

Just because that is true now doesn't mean that will be true in a couple hours. Timestamps are important.

LostInIgnorance

As I had asked before, since things keep getting astray, does the panic I received on my Soekris board after doing this process relevant to anything or should I be ignoring it?

Updated to pfSense-Full-Update-2.0-BETA5-i386-20110126-0422.tgz 26-Jan-2011 09:47 83M

/etc/rc.conf_mount_rw
fetch http://files.pfsense.org/kernel_new.gz
cp kernel_new.gz /boot/kernel/kernel.gz

Rebooted
Pushed traffic over OpenVPN

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address   = 0x0
fault code              = supervisor read, page not present
instruction pointer     = 0x20:0x0
stack pointer           = 0x28:0xd553ebc0
frame pointer           = 0x28:0xd553ebcc
code segment            = base 0x0, limit 0xfffff, type 0x1b
                        = DPL 0, pres 1, def32 1, gran 1
processor eflags        = interrupt enabled, resume, IOPL = 0
current process         = 12 (irq5: vr1)
[thread]
Stopped at      0:      *** error reading from address 0 ***
db> bt
Tracing pid 12 tid 64026 td 0xc3aa4780
m_tag_delete(c4dde600,c4de0b3d) at m_tag_delete+0x32
m_tag_delete_chain(c4dde600,0,d553ec2c,c0c94f99,c4dde600,...) at m_tag_delete_chain+0x82
mb_dtor_mbuf(c4dde600,100,0,c0a9041f,c3a92000,...) at mb_dtor_mbuf+0x25
uma_zfree_arg(c1d7b000,c4dde600,0,c3a28000,d553ec70,...) at uma_zfree_arg+0x29
m_freem(c4dde600,c3debe00,c3a92000,7f,7f,...) at m_freem+0x43
vr_txeof(0,c12c69c0,d5530008,0,0,...) at vr_txeof+0x340
vr_intr(c3a28000,0,109,dbd32e12,b15,...) at vr_intr+0x1c7
intr_event_execute_handlers(c39757f8,c3972180,c0e6ba48,52d,c39721f0,...) at intr_event_execute_handlers+0x14b
ithread_loop(c3a8e450,d553ed38,6000,f600001b,0,...) at ithread_loop+0x6b
fork_exit(c09ae910,c3a8e450,d553ed38) at fork_exit+0x91
fork_trampoline() at fork_trampoline+0x8
--- trap 0, eip = 0, esp = 0xd553ed70, ebp = 0 ---
db>[/thread]

eri--

On the same link posted another kernel.
Thank you very much to every one helping on this.

LostIgnorance its very valuable just that this is not chat and i just do not want to spam with useless comments.

vito

No problem for testing!

so should i download just the kernel from the same link with out a snap update?
I am on snap
2.0-BETA5 (i386)
built on Wed Jan 26 09:44:03 EST 2011

Just want to make sure.

eri--

Yeah just download the file after this post i updated it again.

eri--

FYI i just committed some patches to pfSense repo which should fix the hangs and the panics.
All those on non full installs of i386 grab the next snapshot and test when it comes out.

jimp

Since the snapshot builder won't be done with its current run for a while yet, I ran off an i386 full update on my builder and uploaded it here:
http://cvs.pfsense.org/~jimp/pfSense-Full-Update-2.0-BETA4-i386-20110127-1627.tgz

If someone wants to give that a shot, have at it. The amd64 builder doesn't take quite as long, it should be up in a couple hours, the nanobsd builds probably won't be up until after midnight.

vito

Jimp,
does this have the kernel that had to be downloaded included or do we need to still do the update then download the kernel from ermal's post?
Thanks!

jimp

Ermal checked the patches into the repo so this should be equivalent to that, or better.