cores in new 0.35 (Re: arla 0.35)

[Karl Kleinpaste]

Linux RH6.2, kernel 2.2.16, uniproc.

0.35's arlad is dropping core on me occasionally, thus:

(gdb) bt
#0  0x40049d21 in __kill () from /lib/libc.so.6
#1  0x40049996 in raise (sig=6) at ../sysdeps/posix/raise.c:27
#2  0x4004b0b8 in abort () at ../sysdeps/generic/abort.c:88
#3  0x40043bae in __assert_fail () at assert.c:59
#4  0x806c391 in xfs_message_wakeup (fd=6, h=0x40289008, size=296) at xfs.c:247
#5  0x806c25b in xfs_message_receive (fd=6, h=0x40289008, size=296)
    at xfs.c:203
#6  0x805a80b in process_message (msg_length=296, msg=0x40289008 "(\001")
    at kernel.c:86
#7  0x805a8ca in sub_thread (v_myself=0x40289008) at kernel.c:120
#8  0x808cb0c in Create_Process_Part2 () at lwp_asm.c:630
#9  0xefeeedec in ?? ()
#10 0x81c3954 in ?? ()
#11 0x808c627 in LWP_INTERNALSIGNAL (event=0x40289008, yield=1)
    at lwp_asm.c:501
#12 0x805b46f in kernel_interface (args=0xbffff75c) at kernel.c:395
#13 0x808cb0c in Create_Process_Part2 () at lwp_asm.c:630
#14 0xefeeedec in ?? ()
#15 0x808cdc8 in Dispatcher () at lwp_asm.c:693
#16 0xb807ebc0 in ?? ()
Cannot access memory at address 0x310cc483

At the time, I was doing a "cp -p <bunch of files> <afs directory>",
and because I'm on a slow link, I had cd'd into the same directory in
another shell to do an "ls -l" on it, to observe progress.  All my
terminals were simultaneously informed of:
        Message from syslogd at cinnamon at Sat Jan 27 22:14:55 2001 ...
        cinnamon kernel: xfs_readdir: data not valid: -19 
at the time of the core.

--karl