Re: Communicator crash; kill -9 won't remove (pre-10!)

Jeremey Barrett (jeremey@bluemoney.com)
Tue, 7 Oct 1997 16:50:46 -0700


-----BEGIN PGP SIGNED MESSAGE-----

Rob Hagopian writes:
> OK, I had another communicator crash that kill -9 won't remove. No locked
> buffers, and strace acts funny.
>

This is very similar to a problem I had recently with SSLeay. Running
the ssleay executable, I was able to consistently reproduce a state
of "disk sleep" (i.e. the D in your ps output) when performing one
particular operation. Once in a state of disk sleep, the process could
not be killed, and the only solution was a hard reboot.

I'm running 2.0.30. Reproducing what I did, here's the very end of
the strace output:

read(4, "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 6460) = 55
close(4) = 0
time([876266313]) = 876266313
open("/usr/lib/zoneinfo/GMT", O_RDWR) = 4
ioctl(4, 0x101, 0xbfffeea4) = 0
ioctl(4, 0x102, 0xbfffeec4) = 0
close(4

And here it sits. The uptime stats continue to gradually climb over
time to about 2.1, then level off. Hope this is useful. In general I
get the exact same behavior Rob described.

Jeremey.
- --
Jeremey Barrett BlueMoney Software Corp.
Crypto, Ecash, Commerce Systems http://www.bluemoney.com/
PGP key fingerprint = 3B 42 1E D4 4B 17 0D 80 DC 59 6F 59 04 C3 83 64

-----BEGIN PGP SIGNATURE-----
Version: 2.6.2
Comment: Processed by Mailcrypt 3.4, an Emacs/PGP interface

iQCVAwUBNDrKly/fy+vkqMxNAQEiiQQAio82pcC1lYYQ8+hsjGLgvd01tbnaa8Ri
39ZyjtP4iLYjq+FzEDMGuEDuaIQnKVEzBcl/zUwGNVSa60JDi1ze1GcRBboMq9zD
HjWxcQBW5hzC9u6F3NPBiM7Lxo5rumFJEzC7j9teK8ZUYlVXXxqdePz+2qU51bdQ
SX2D9a79gQs=
=ymcX
-----END PGP SIGNATURE-----