Fatal trap 12 mostly cvsup

[Johan Hendriks]

Hello all 

I use a test box with current for quite some time know.
It is mostly used as a cvsup-mirror for my other servers.

It has 6  250 GB disks ad4, ad5, ad6, ad7, ad8, and ad10.

Ad5 to ad10 are labeled   as disk01 to disk05 and i have created a raidz2 pool from those disk.

It all worked well, but as of today it gives me panics when i enable cvsupd or do a /usr/local/etc/cvsup/update.sh

My cvsup dir ncvs is on the raidz2 pool as is my /usr/ports dir.

 

This is the error i get

 

Fatal trap 12: page fault while in kernel mode
cpuid = 0; apic id = 00
fault virtual address                                      = 0x0
fault code                                                          = supervisor write data, page not present
instruction pointer                                         = 0x20:0xffffffff80834046
stack pointer                                                    = 0x28:0xffffff802e090690
frame pointer                                                  = 0x28:0xffffff802e0906f0
code segment                                                 = base 0x0, limit 0xfffff, type 0x1b
                                                                              = DPL 0, pres 1, long 1, def32 0, gran 1
proccessor eflags                                           = interupt enabled, resume, IOPL = 0
current process                                               = 58199 (cvsup)
trap number                                                     = 12
panic: page fault
cpuid = 0
Uptime: 1h4514s
Physical memory: 1009M

Dumping 1700 MB: And here the cursor stays.



I can switch to other terminals by pressing ALT Fx, it takes about 5 to 10 seconds and then it goes to that terminal.
If a client does a cvsup –g –L2  then it says it is connnected to the hanging machine but nothing happens.

I can not do anything on the consoles.
If i do not use cvsup-mirror adn cvsupd then the machine runs fine.
But a //usr/local/etc/cvsup/update.sh or running the cvsupd crashes the machine.

 

No special things in /etc/make.conf.



CPUTYPE?=nocona

KERNCONF=KRNL

BATCH_DELETE_OLD_FILES= yes

 

I even installed the base system again with the latest snapshot and imported the raidz2 pool 
did a buildworld with sources from yesterday.
And the system behaves the same.
 
In loader.conf i have 
vm.kmem_size=”512M”

vm.kmem_size_max=”512”

 

Mostly it is the cvsup process, but i did see some other processes as well 
I do not remember them well, but one started with g_f… 



regards,

Johan Hendriks