Hi,
I'm facing recurrent issue with MySQL inside a domU.
Look's like it is not really only MySQL, but the environment in which
mysql runs. My VM dies each time I'm trying to have intensive i/o to the
db (from another domU using php-mysql).
When the VM freezes, I log onto the console, using "xm console domuID"
are here's what I get:
########################################################################
Starting inetd.
Starting cron.
Fri May 12 00:37:17 CEST 2006
NetBSD/i386 (akela.tumfatig.net) (console)
login: panic: kernel debugging assertion "(vp->v_flag & VONWORKLST)" failed:
file "../../../../miscfs/genfs/genfs_vnops.c", line 1322
Stopped in pid 490.3 (mysqld) at netbsd:cpu_Debugger+0x4: popl
%
ebp
cpu_Debugger(0,c0697480,0,cb1dfcc8,c032a019) at netbsd:cpu_Debugger+0x4
panic(c037d643,c035c270,c036ec19,c036ea98,52a) at netbsd:panic+0xa9
__assert(c035c270,c036ea98,52a,c036ec19,c0819100) at netbsd:__assert+0x19
genfs_putpages(cb1dfe3c,c0342360,cb228748,0,0) at netbsd:genfs_putpages+0x961
VOP_PUTPAGES(cb228748,0,0,0,0) at netbsd:VOP_PUTPAGES+0x40
nfs_flush(cb228748,ca0c00fc,1,cb1424d0,0) at netbsd:nfs_flush+0x1d
nfs_close(cb1dfeb4,c03418a0,cb228748,3,ca0c00fc) at netbsd:nfs_close+0x54
VOP_CLOSE(cb228748,3,ca0c00fc,cb1424d0,cb228748) at netbsd:VOP_CLOSE+0x34
vn_close(cb228748,3,ca0c00fc,cb1424d0,cb1dff50) at netbsd:vn_close+0x2b
vn_closefile(cb2c0894,cb1424d0,0,0,0) at netbsd:vn_closefile+0x17
closef(cb2c0894,cb1424d0,cb1dff68,c0383908,0) at netbsd:closef+0x15a
syscall_plain() at netbsd:syscall_plain+0xc2
--- syscall (number 6) ---
0xbd91c1b7:
ds 0x11
es 0x11
fs 0x31
gs 0x11
edi 0xc037d643 copyright+0x311a3
esi 0x100
ebp 0xcb1dfc98
ebx 0xcb1dfcb8
edx 0xc0467000 emul_freebsd_object+0x92f6c
ecx 0xfffffffe
eax 0xb3e
eip 0xc02ed1cb cpu_Debugger+0x4
cs 0x9
eflags 0x202
esp 0xcb1dfc98
ss 0x11
netbsd:cpu_Debugger+0x4: popl %ebp
Stopped in pid 490.3 (mysqld) at netbsd:cpu_Debugger+0x4: popl
%
ebp
db>
########################################################################
Does anyone see what's not correct in my VM ?
Memory seems to be enough (according to my rrdtool graph):
the VM is using ~124Mo over the 192Mo is has, and no data in swap file.
If it matters, dom0 has 192Mo available, fully (190Mo average) used but
none of the swap file is used.
any idea ?
TIA,
Jo
Attachment:
pgpiw_f1JBuZ4.pgp
Description: PGP signature