there is a bug in twe driver or disk subsystem for sure
matrix at itlegion.ru
Sun Apr 4 23:26:14 PDT 2004
Hello Vinod and everybody
I am not pretty good at reproducing the problem. I THINK
(not sure) that it might be related to 3dm software because
when i do very fast (i click reload page as fast as i can)
MONITOR page reloads in 3dm i will get kernel panic
within 1-2 minutes. However, panic happened by itself a couple
of time during havy disk load.
I did 'tr' while in debugger and wrote down the output on a paper.
Belows - is the reproduction of that outout. However, it might
contain some typos, because it is very hard to reproduce such
info by hand.
here is 'tr' output:
__panic(c07d482d, 8a, c07d4847,c07b9714,c07d47fe) at __panic+0x13d
mtrash_ctor(c7ac9e00,200,0) at mtrash_ctor+0x59
uma_zalloc_arg(c1038e00,0,1) at uma_zalloc_arg+0x169
malloc(200,c0815c20,1,c74da4c0,e5614c60) at malloc+0xb7
twe_get_param(c74be000,200,2,4,0) at twe_get_param+0x3f
twe_ioctl(c74be00,800c5469,e5614c60,e5614b70,c05b0a9e) at twe_ioctl+0x25a
spec_ioctl (e5614b88,e5614c34,c063682f,e5614b88,c085dbc0) at
spec_vnoperate(e5614b88) at spec_vnoperate+0x13
vn_ioctl(c7900198,800c5469,e5614c60,c3477600,c78e83f0) at vn_ioctl+0x17f
ioctl(c78e83f0,e5614d14,3,9d4,297) at ioctl+0x37c
syscall(2f,2f,2f,800c5469,200) at syscall+0x217
Xint0x80_syscall() at Xint0x80_syscall+0x1d
---syscall (54,FreeBSD ELF32, ioctl),
I hope this will help figure out what is causing the panic.
> The 3ware (twe) driver is obviously not causing this panic.
> It's something else (at line 128 in file /usr/src/sys/udm_dbg.c).
> Every time you try deleting a unit with mounted filesystems,
> or shutdown the system with mounted filesystems on 3ware units,
> you will see the message you mention, since the driver tries
> to delete the unit, but finds it to be busy. Don't let the system
> panic, and simply shut it down. You should still see the message.
> -----Original Message-----
> From: Artem Koutchine [mailto:matrix at itlegion.ru]
> Sent: Friday, April 02, 2004 9:05 AM
> To: freebsd-current at freebsd.org
> Subject: there is a bug in twe driver or disk subsystem for sure
> I am running 5.2-CURRENT. The box has 3w 8506-4LP
> SATA raid controller and freebsd is install onto raid5 array.
> Almost everytime i pull a hotswap disk (one out of four)
> or when very heavy disk activity (like background fsck after
> crashing on a 310GB label) i get the following after 10 or
> so minutes of uptime:
> Memory modified after free 0x788f400(508) val=20202020 @ 0xe788f400
> panic: Most recently used by devbuf
> at line 128 in file /usr/src/sys/udm_dbg.c
> Debugger ("panic")
> Stopped at Debugger +0.46: xchgl %ebx, in_Debugger.0
> and i typed 'c' in debugger:
> the system started to shutdown and here is what i saw:
> twe0: failed to delete unit 0
> stray irq9
> The box is:
> permicro X5DPE-G2 motherboard
> DUAL XEON 2.66Ghz (HT enabled) 533Mhz bus
> 2GB RAM
> 4 SATA SEAGTE 120GB DRIVES
> 3WARE 8506-4LP SATA RAID5 CONTROLLER
More information about the freebsd-bugs