fusefs-ntfs panic on writte on 8.1

Seeker

Seeker

As soon as I transfered from 8.0 -> 8.1, ntfs-3g mounted slices, after writting to it, every here and then, kernel panic occurs!

I tried another 8.1 install on clean USB stick..., same thing!
 
I did upward recursive reinstall of fusefs-ntfs (all ports on which it depends) and nada.
I even did a fresh USB install from zero and same thing occurs.
Sometimes 'cp', causes a panic and sometimes it is a 'vi', the other time it is a 'mv', etc... :P
 
Once again... in 'vi' opened file residing at ntfs.
As soon as I've hit:
Code: 
:w
panic!

info.*
Code: 
  Panic String: page fault
  Dump Parity: 1080399392
  Bounds: 14
  Dump Status: good

core.txt.*
Code: 
panic: page fault

...
Unread portion of the kernel message buffer:


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	        = 0x28:0xea651b84
frame pointer	        = 0x28:0xea651c44
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		= 1358 (vi)
trap number		= 12
panic: page fault
cpuid = 0
Uptime: 10m46s
Physical memory: 3557 MB
Dumping 124 MB: 109 93 77 61 45 29 13

...
#0  doadump () at pcpu.h:246
246	pcpu.h: No such file or directory.
	in pcpu.h
(kgdb) #0  doadump () at pcpu.h:246
#1  0xc089e9b7 in boot (howto=260) at /usr/src/sys/kern/kern_shutdown.c:416
#2  0xc089ec19 in panic (fmt=Variable "fmt" is not available.
) at /usr/src/sys/kern/kern_shutdown.c:590
#3  0xc0bd3adc in trap_fatal (frame=0xea651b44, eva=0)
    at /usr/src/sys/i386/i386/trap.c:938
#4  0xc0bd3d60 in trap_pfault (frame=0xea651b44, usermode=0, eva=0)
    at /usr/src/sys/i386/i386/trap.c:851
#5  0xc0bd46a5 in trap (frame=0xea651b44) at /usr/src/sys/i386/i386/trap.c:533
#6  0xc0bb67bb in calltrap () at /usr/src/sys/i386/i386/exception.s:165
#7  0x00000000 in ?? ()
Previous frame inner to this frame (corrupt stack?)
(kgdb) 

...
iostat

iostat: kvm_read(_tk_nin): invalid address (0x0)
iostat: disabling TTY statistics
iostat: kvm_getcptime: invalid address (0x0)
iostat: disabling CPU time statistics
            ada0              da0              cd0 
  KB/t tps  MB/s   KB/t tps  MB/s   KB/t tps  MB/s 
  8.48 186  1.54   1.65   3  0.01   2.34   0  0.00 

...
fstat

USER     CMD          PID   FD MOUNT      INUM MODE         SZ|DV R/W
root     ntfs-3g     1135 root /             2 drwxr-xr-x    1024  r
root     ntfs-3g     1135   wd /             2 drwxr-xr-x    1024  r
root     ntfs-3g     1135 text /usr     519130 -r-xr-xr-x   33676  r
root     ntfs-3g     1135    0 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1135    1 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1135    2 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1135    3 /         49754 -rw-r--r--     567  r
root     ntfs-3g     1135    4 /dev        103 crw-r-----  ada0s5 rw
root     ntfs-3g     1135    5 /dev        183 crw-rw----   fuse1 rw
root     ntfs-3g     1135    6* local dgram c8957968 <-> c8956764
root     ntfs-3g     1132 root /             2 drwxr-xr-x    1024  r
root     ntfs-3g     1132   wd /             2 drwxr-xr-x    1024  r
root     ntfs-3g     1132 text /usr     519130 -r-xr-xr-x   33676  r
root     ntfs-3g     1132    0 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1132    1 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1132    2 /dev         31 crw-rw-rw-    null rw
root     ntfs-3g     1132    3 /         49754 -rw-r--r--     567  r
root     ntfs-3g     1132    4 /dev         99 crw-r-----  ada0s1 rw
root     ntfs-3g     1132    5 /dev        182 crw-rw----   fuse0 rw
root     ntfs-3g     1132    6* local dgram c8956408 <-> c8956764

...
 
have you tried the patch attached to PR ports/149674 ?
 
I assume that some times cp work just fine but when i edit anything with any text editor this happens. Anyway i will follow the PR.
 
Seeker said:
Well, I've filled a PR, at 12.08.2010, but still nada!

Patch is BROKEN for 8.1 RELEASE.
Can't be applied cleanly.
Click to expand...

it applied fine here. did you download the patch.txt file or copy and paste the patch from the web page ? the latter surely won't work
 
acheron said:
it applied fine here. did you download the patch.txt file or copy and paste the patch from the web page ? the latter surely won't work
Click to expand...
Copy-paste. Then, I've manually applied patch and build went fine.
But now, I can't use ntfs-3g anymore:
Code: 
fuse: failed to open fuse device: No such file or directory
 
acheron said:
did you kldload fuse ?
Click to expand...
kldunload-ed then:
Code: 
kldload /usr/local/modules/fuse.ko
Now I got rid of patch and after rebuilding sysutils/fusefs-kmod and sysutils/fusefs-ntfs, I am getting:
Code: 
fuse: failed to open fuse device: No such file or directory
I can't revert back?!
This is a most possible cause.
So, it might be, that patch is actually perfectly working, but I ma unaware of it because of preceding problem.
 
acheron said:
do you have a /dev/fuse device ? what perms ?
Click to expand...
Doesn't exists.
Look at a link above, as I thing I have other problem and this is just a secondary manifestation.
I need to solve it, before fixing fuse-ntfs
 
The patch from PR/149674 have a problem.


Code: 
===>  Applying FreeBSD patches for fusefs-kmod-0.3.9.p1.20080208_6
1 out of 1 hunks failed--saving rejects to fuse_module/fuse_main.c.rej
=> Patch patch-fuse_module__fuse_main.c failed to apply cleanly.
=> Patch(es) patch-fuse_module__Makefile patch-fuse_module__fuse.h patch-fuse_module__fuse_dev.c patch-fuse_module__fuse_io.c applied cleanly.
*** Error code 1

Stop in /usr/ports/sysutils/fusefs-kmod.
*** Error code 1


For me it won't work. I checked it double but its is just like the PR says.

uname -a

Code: 
8.1-RELEASE FreeBSD 8.1-RELEASE #1: Thu Aug 26 23:09:57 EEST 2010 amd64

Any ideas please.
 
You must log in or register to reply here.
Back
Top