Kernel Panic
-
Can you type "bt" at that db prompt and post the backtrace output?
-
2.0-BETA5 (amd64)
built on Tue Jan 25 07:56:16 EST 2011In the ultimate of ironies, I installed the dev kernel about 5 hours ago and rebooted. I checked 5 minutes ago and pfsense was in panic. Unfortunately when I tried to save the console output my serial terminal program froze, probably because my nfs share had disappeared.
Fortunately, within 2 minutes of having pfsense back up again, and while composing this post, it panicked again. I've attached two files for your reading pleasure, but I'm back to the SMP kernel for now.
-
huh…just tested again but with a about 500mb
Did two transfers of this size during the same openvpn session and everything was fine.
Will keep an eye on it.
update: just happened again. I will try to get to the console.
-
Yeah this one is definitely new. When ermal comes online he should hopefully be able to spot the problem.
-
I understand the em is mainly a driver issue, but what about the soekris vr? Is that just a driver issue too?
-
Not sure. It may be a more general bug that certain driver code is more likely to hit. Ermal had some ideas based on the testing done here with that custom kernel yesterday, but he was pretty overloaded with other stuff so I'm not sure if that specific fix made it in last night. Plenty of other changes went in though.
-
After 2 hours 40 minutes, one of the firewalls panicked again. This time it didn't freeze though, so I managed to get a backtrace. Here's what it says:
Stopped at pf_state_tree_id_RB_REMOVE_COLOR+0x90: cmpl $0x1,0xa (rest is not visible on the photo)
db> bt
Tracing pid 8 tid 64074 td 0xffffff00025a3000
pf_state_tree_id_RB_REMOVE_COLOR() at pf_state_tree_id_RB_REMOVE_COLOR+0x (something)
pf_state_tree_id_RB_REMOVE() at pf_state_tree_id_RB_REMOVE+0x126
pf_unlink_state() at pf_unlink_state+0x5e
pf_purge_expired_states() at pf_purge_expired_states+0xed
pf_purge_thread() at pf_purge_thread+0xb4
fork_exit() at fork_exit+0x118
fork_trampoline() at fork_trampoline+0xe
–- trap 0, rip = 0, rsp = 0xffffff801f73ed30, rbp = 0 ---[might be an error or two, I typed that off of a photo]
Cheers.
-
You have any ftp/pptp traffic going in?
-
@ermal:
You have any ftp/pptp traffic going in?
Only very little ftp, but that doesn't seem to trigger it.
I do, however, have quite a bit of http traffic (~300-600mbit) from a million or so clients. -
Nope, still a problem. I updated the carp backup to the latest snapshot (Jan 26, with the master still on Jan 24), added a new VIP addr on the master, and the backup immediately panic'd.
-
Hi! this is my first post on this forum.
I have been using pfsense for one year, I have installed it on a alix 2d3 an on a jetway nc9c-550lf. since 1 month pfsense rebooting itself on both installation.
The snapshot is 2.0-BETA5 (i386) built on Wed Jan 26 09:44:03 EST 2011
On screen I see this messages:
Fatal trap 12: page fault while in kernel mode
cpuid= 1; apic id= 01
fault virtual address = 0x0
fault code = supervisor read, page not present
instruction pointer = 0x20:0x0
stack pointer = 0x28:8xc5610b80
frame pointer = 0x28:8xc5610b8c
code segment = base 0x0, limit 0xfffff, type0x1b
= DPL 0, pres 1, def32 1, gran1
processor eflags= interrupt enable, resume,IOPL = 0
current process= 12(swi5: + ) -
Without the output of "bt" at that debug prompt, it's hard to say. Yours doesn't look like the other panics in this thread.
-
the next time I will try the bt command and i will post the screen.
thx
-
the next time I will try the bt command and i will post the screen.
…after you install the dev kernel. Instructions are on the wiki (for full installs) or about 2 screens back for embedded.
-
I'm working on getting text dumps working so that even people on a normal install will have the crash dump analysis left in /var/crash after an auto reboot. Hitting a couple roadblocks, though.
-
I think I am running into this issue. I do not have the dev kernel installed. Thought it was a bad hard drive, Had a lot of drive errors after rebooting a few times until it no longer booted. Replaced the drive. Machine ran fine for a couple hours off network. As soon as traffic started hitting it it quit within a few minutes. After a reboot it passed traffic a few minutes and quit again.
Had been running the Dec31 Full i386(it ran fine). Updated to Jan 21 over the weekend. Issue started yesterday. Loaded Jan 26 amd64 today and still see the same behavior.
Does this sound similar to what everyone else is seeing? Where can I find the Dec 31 snap, i must have erased it?
I have another box at the office and some spare public IP's. I can set it up for testing if any of the devs want access (when its up).
Let me know.
This is on a 5015A-EHF-D525 w/4G RAM
http://www.supermicro.com/products/system/1U/5015/SYS-5015A-EHF-D525.cfm -
Text dumps will be a nice feature. Thanks for working on that Jimp.
-
On the setup we have where secondary panics used to be replicable it's fixed now.
jnorell: can you email me a backup of your config? cmb at pfsense dot org
-
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. -
Is this just for the intel em chip fix? or will it also work on my Soekris vr chip?
