7 Aug
2019
7 Aug
'19
10:18 p.m.
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Adam
7 Aug
7 Aug
11:30 p.m.
Thanks for sharing! =)
---8<---Corte aqui---8<---
http://www.tabajara-labs.blogspot.com
http://www.tabalabs.com.br
---8<---Corte aqui---8<---
Em qui, 8 de ago de 2019 ?s 02:18, Adam Thornton via cctalk <
cctalk at classiccmp.org> escreveu:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr
have no passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not
guaranteed. All systems are hosted on Raspberry Pis (the 36-bit ones on a
Pi 3B+ and the 16-bit and 32-bit ones on a Pi 2B+) on Debian Buster.
Absolutely no guarantee of availability or usability is made.
Adam
8 Aug
8 Aug
12:44 a.m.
?for example how many can be? on the? tops 10 system at a time?
this is neat! #
In a message dated 8/7/2019 11:30:35 PM US Mountain Standard Time, cctalk at
classiccmp.org writes:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr
have no passwords.
6:59 a.m.
On 8 Aug 2019, at 15:18, Adam Thornton via cctalk
<cctalk at classiccmp.org> wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Just logged into TOPS-10 for the first time in many years! Far too much brain bit rot but
it will encourage me to build up my own RPi emulation setup!
Thanks.
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
11:20 a.m.
On 8/8/2019 7:59 AM, Huw Davies via cctalk wrote:
Does the PI in use have good media storage device?
I suspect TOPS and UNIX swap pages like mad. A little SD card might wear
out in few weeks.
Ben.
On 8 Aug 2019, at 15:18, Adam Thornton via cctalk
<cctalk at classiccmp.org> wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Just logged into TOPS-10 for the first time in many years! Far too much brain bit rot but
it will encourage me to build up my own RPi emulation setup!
Thanks. 
11:26 a.m.
On Thu, Aug 8, 2019 at 12:20 PM ben via cctalk <cctalk at classiccmp.org>
wrote:
On 8/8/2019 7:59 AM, Huw Davies via cctalk wrote:
Even the crappiest of crap SD cards these days aren't that fragile. You'd
need to swap on the order of GB/s to wear it out that fast. Most of the SD
cards can handle hundreds of full drive writes. At 128GB, you're looking at
needing to generate about ~25TB of effective writes before you'd wear them
out. Even with a crazy 10x write amp (typical is 2-3), there's no way you'd
get that through an interface that's measured in the tens of MB/s.
Warner
Warner
> On 8 Aug 2019, at 15:18, Adam Thornton via cctalk <
cctalk at
classiccmp.org> wrote:
>
> https://mvsevm.fsf.net
>
> Currently, the TOPS-10 guest account (42,42) and the Unix v7 account
dmr have
no passwords.
Just logged into TOPS-10 for the first time in many years! Far too much
brain bit
rot but it will encourage me to build up my own RPi emulation
setup!
Thanks.
Does the PI in use have good media storage device?
I suspect TOPS and UNIX swap pages like mad. A little SD card might wear
out in few weeks.
11:31 a.m.
On Thu, Aug 8, 2019 at 12:26 PM Warner Losh <imp at bsdimp.com> wrote:
On Thu, Aug 8, 2019 at 12:20 PM ben via cctalk <cctalk at classiccmp.org>
wrote:
Although to be fair, 50MB/s for 5 days would generate that. I doubt these
old emulated systems could generate that much write traffic via SIMH on a
sustained basis. And if the SD card is any good at all, it will be at least
10x better than that.... Assuming anything decent, and a 5MB/s write rate,
you're looking at years to wear it out with extremely heavy use.
Warner
On 8/8/2019 7:59 AM, Huw Davies via cctalk
wrote:
Even the crappiest of crap SD cards these days aren't that fragile. You'd
need to swap on the order of GB/s to wear it out that fast. Most of the SD
cards can handle hundreds of full drive writes. At 128GB, you're looking at
needing to generate about ~25TB of effective writes before you'd wear them
out. Even with a crazy 10x write amp (typical is 2-3), there's no way you'd
get that through an interface that's measured in the tens of MB/s.
> On 8 Aug 2019, at 15:18, Adam Thornton via cctalk <
cctalk at
classiccmp.org> wrote:
>
> https://mvsevm.fsf.net
>
> Currently, the TOPS-10 guest account (42,42) and the Unix v7 account
dmr have
no passwords.
Just logged into TOPS-10 for the first time in many years! Far too much
brain bit
rot but it will encourage me to build up my own RPi emulation
setup!
Thanks.
Does the PI in use have good media storage device?
I suspect TOPS and UNIX swap pages like mad. A little SD card might wear
out in few weeks.
11:51 a.m.
On 8/8/2019 12:26 PM, Warner Losh wrote:
Even the crappiest of crap SD cards these days
aren't that fragile.
You'd need to swap on the order of GB/s to wear it out that fast. Most
of the SD cards can handle hundreds of full drive writes. At 128GB,
you're looking at needing to generate about ~25TB of effective writes
before you'd wear them out. Even with a crazy 10x write amp (typical is
2-3), there's no way you'd get that through an interface that's measured
in the tens of MB/s.
I would agree if it was doing a whole disk.
A swap or scatch space on magnetic media got used alot back then.
Time sharing back then was having 16KW and swapping pages back and forth
from rotating media while reading or writing cards.
Warner
I am not sure of the memory on a PI,but having a good block cache
for the swap segments on disk would useful.
Well the CPU is the easy part. Now what about the front panel. :)
Ben.
12:04 p.m.
On Thu, Aug 8, 2019 at 12:51 PM ben <bfranchuk at jetnet.ab.ca> wrote:
Agreed. I don't think SIMH forces synchronous writes... A good buffer cache
will mitigate this. I tried to wear our old CF cards w/o disabling the
buffer cache and found it was impossible... But with disabling it, it was
just barely possible, if you did the right things...
On 8/8/2019 12:26 PM, Warner Losh wrote:
The NAND pool in the SD cards is rotated to do wear-leveling and to cope
with the insanely large erase block sizes that we have these days, so the
particular LBAs being re-scribbled doesn't matter, though you might get a
lot of write-amp from doing 512-byte I/O when the underlying page size is
4kb or 16kb. Even so, it would take a lot...
Even the crappiest of crap SD cards these days
aren't that fragile.
You'd need to swap on the order of GB/s to wear it out that fast. Most
of the SD cards can handle hundreds of full drive writes. At 128GB,
you're looking at needing to generate about ~25TB of effective writes
before you'd wear them out. Even with a crazy 10x write amp (typical is
2-3), there's no way you'd get that through an interface that's measured
in the tens of MB/s.
I would agree if it was doing a whole disk.
A swap or scatch space on magnetic media got used alot back then.
Time sharing back then was having 16KW and swapping pages back and forth
from rotating media while reading or writing cards.
Warner
I am not sure of the memory on a PI,but having a good block cache
for the swap segments on disk would useful.
Well the CPU is the easy part. Now what about the
front panel. :)
This is purely the dialup experience, eh? :)
Warner
Ben.
8:15 a.m.
Be careful eh about posting credentials right to a public mailing list like eh
Sent from my iPad
On Aug 8, 2019, at 12:18 AM, Adam Thornton via cctalk
<cctalk at classiccmp.org> wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Adam
10:21 a.m.
Yeah, you might be able to DOS my Raspberry Pi. Maybe you could break out
of simh and start using the Pi itself to mine bitcoins.
That could cost me like literally dozens of cents worth of power.
I suppose I should be a little concerned that once you've broken out of the
simulation into the Pi, then you are on a machine which has access to the
rest of my network. In which case, assuming you can exploit a remote
vulnerability (or crack a credential), then, yeah, you could indeed
somewhat inconvenience me.
Well, here's hoping that there is lower-hanging fruit with more compute
power out there. I suspect there is.
Adam
On Thu, Aug 8, 2019 at 8:15 AM Adrian Stoness <tdk.knight at gmail.com> wrote:
Be careful eh about posting credentials right to a
public mailing list
like eh
Sent from my iPad
On Aug 8, 2019, at 12:18 AM, Adam Thornton via
cctalk <
cctalk at classiccmp.org> wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr
have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not
guaranteed. All systems are hosted on Raspberry Pis (the 36-bit ones on a
Pi 3B+ and the 16-bit and 32-bit ones on a Pi 2B+) on Debian Buster.
Absolutely no guarantee of availability or usability is made.
Adam
10:34 a.m.
On 8/8/2019 12:21 PM, Adam Thornton via cctalk wrote:
Yeah, you might be able to DOS my Raspberry Pi. Maybe
you could break out
of simh and start using the Pi itself to mine bitcoins.
That could cost me like literally dozens of cents worth of power.
I suppose I should be a little concerned that once you've broken out of the
simulation into the Pi, then you are on a machine which has access to the
rest of my network. In which case, assuming you can exploit a remote
vulnerability (or crack a credential), then, yeah, you could indeed
somewhat inconvenience me.
Well, here's hoping that there is lower-hanging fruit with more compute
power out there. I suspect there is.
Adam
Thanks for the chortle in the AM here in IA.? My apologies for not
sending a note of appreciation last night.? I immediately logged into
Unix v7 as dmr and played around for a bit.? (no idea how to break out
of the spell command without any arguments, but that's my bad).? I'm a
CISO equivalent in my day job, and I didn't see any issue with your
initial email.
Jim
11:33 a.m.
On Aug 8, 2019, at 12:21 PM, Adam Thornton via cctalk
<cctalk at classiccmp.org> wrote:
Yeah, you might be able to DOS my Raspberry Pi. Maybe you could break out
of simh and start using the Pi itself to mine bitcoins.
I know you?re joking around, but IoT gadgets and other small devices are being exploited
right now to great effectiveness as attack amplifiers or penetration aids. They won?t mine
coins on your Pi, or attack you directly, but they _will_ use your Pi as a means to attack
others - for which you can be held legally liable in some jurisdictions. And it won?t be a
human who makes the call, it will be an automated attack bot that will slam every port it
can attempting to find something its masters can abuse to make money. I ran public
instances of TOPS-20 and ITS for years, and eventually had to give up due to lack of
interest by anyone except chinese bots using up all the TCP connections with (usually
Cisco) exploit attempts.
This is, after all, The Internet, and The Internet ruins all things as quickly as
possible. This is why we can?t have nice things.
noon
On Aug 8, 2019, at 2:33 PM, Daniel Seagraves via
cctalk <cctalk at classiccmp.org> wrote:
But if the only public port into the device is a TOPS-10 CLI, that's going to stop
your typical criminal pretty well.
paul
On Aug 8, 2019, at 12:21 PM, Adam Thornton via
cctalk <cctalk at classiccmp.org> wrote:
Yeah, you might be able to DOS my Raspberry Pi. Maybe you could break out
of simh and start using the Pi itself to mine bitcoins.
I know you?re joking around, but IoT gadgets and other small devices are being exploited
right now to great effectiveness as attack amplifiers or penetration aids. 
3:19 p.m.
At 10:18 PM 7/08/2019 -0700, you wrote:
WELCOME TO THE ANCIENT COMPUTER MVSEVM
All systems are emulated, on Raspberry Pi and Linux.
1: Multics MR 12.6f (Honeywell 6800 DPS-8/M)
2: TOPS-20 7.1 (PDP-10 KL-10)
3: TOPS-10 7.03 (PDP-10 KA-10)
4: ITS (PDP-10 KA-10)
5: OpenVMS 7.3 (MicroVAX 3900)
6: Unix v7 (PDP-11/70)
7: Unix 2.11bsd (PDP-11/70)
That's very nice.
Though atm my interests are web front-end dev related, and so for me
the nicest part is this:
gotty-bundle.js 325 KB Minified
xterm.css 35 KB
From xterm.css :
* xterm.js: xterm, in the browser
* https://github.com/chjj/term.js
* Originally forked from (with the author's permission):
* Fabrice Bellard's javascript vt100 for jslinux:
* http://bellard.org/jslinux/
https://github.com/chjj/term.js
term.js ( Latest commit Jun 6, 2016 )
A full xterm clone written in javascript. Used by tty.js. -->
https://github.com/chjj/tty.js
This project is no longer maintained. For a maintained fork take a look at
sourcelair/xterm.js. -->
https://github.com/sourcelair/xterm.js
A fascinating rabbit hole!
Guy
4:29 p.m.
On Aug 7, 2019, at 22:18, Adam Thornton wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Thanks. I had a brief look around the 36-bit systems last night.
One of my tests for a Pi 3 B was to build and run the SIMH HP3000 on it. Being able to do
that means having git to get the SIMH source, gcc and gmake to build, (curl|wget) to get
the MPE V/R bits, unzip to extract. The prerequisites were what I wanted for other Pi
stuff and this was a good workflow to find out what was missing from the Jessie Lite image
(yes it was that long ago, in Stretch and Buster I think I have found all those packages
are already present).
One of the things I have found with the Pi is, the low end micro SD cards (P*tr**t and
K*ngst*n would match the ones that did this) are lossy storage. It?s not that they wear
out, it?s that they lose bits. Switching power supply to one sufficient for the Pi did
not solve this problem, they continued to lose bits.
-Frank McConnell
4:52 p.m.
On Aug 8, 2019, at 4:29 PM, Frank McConnell via cctalk
<cctalk at classiccmp.org> wrote:
On Aug 7, 2019, at 22:18, Adam Thornton wrote:
With my RPi2B, my wife tripped a breaker, and scrambled the card. That was a Lexar SD
card. Of course that was also one of the critical systems in my VMS Cluster. I?ve moved
to VM?s, on my VMware cluster, for all my OpenVMS systems. My one Rpi3B runs Multics, the
other TOPS-20. I?m thinking about rebuilding the RPi2B as a PDP-11.
Zane
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Thanks. I had a brief look around the 36-bit systems last night.
One of my tests for a Pi 3 B was to build and run the SIMH HP3000 on it. Being able to
do that means having git to get the SIMH source, gcc and gmake to build, (curl|wget) to
get the MPE V/R bits, unzip to extract. The prerequisites were what I wanted for other Pi
stuff and this was a good workflow to find out what was missing from the Jessie Lite image
(yes it was that long ago, in Stretch and Buster I think I have found all those packages
are already present).
One of the things I have found with the Pi is, the low end micro SD cards (P*tr**t and
K*ngst*n would match the ones that did this) are lossy storage. It?s not that they wear
out, it?s that they lose bits. Switching power supply to one sufficient for the Pi did
not solve this problem, they continued to lose bits.
-Frank McConnell
5:52 p.m.
On Aug 8, 2019, at 16:52, Zane Healy wrote:
On Aug 8, 2019, at 4:29 PM, Frank McConnell via cctalk <cctalk at classiccmp.org>
wrote:
No interruptions of power to running systems, just cards left idle after shutdown -P for
some days/weeks/months while I worked on others.
At first I was running from a weak (5V 1A) power supply and thought that cute
lightning-bolt icon on-screen was its way of indicating that it was writing to its flash
storage. I switched to a recommended power supply that met the stated requirements and
still had data loss in further experiments with (freshly re-installed and left idle) cards
that had lost bits previously.
-Frank McConnell
One of the things I have found with the Pi is,
the low end micro SD cards (P*tr**t and K*ngst*n would match the ones that did this) are
lossy storage. It?s not that they wear out, it?s that they lose bits. Switching power
supply to one sufficient for the Pi did not solve this problem, they continued to lose
bits.
-Frank McConnell
With my RPi2B, my wife tripped a breaker, and scrambled the card. That was a Lexar SD
card. Of course that was also one of the critical systems in my VMS Cluster. I?ve moved
to VM?s, on my VMware cluster, for all my OpenVMS systems. My one Rpi3B runs Multics, the
other TOPS-20. I?m thinking about rebuilding the RPi2B as a PDP-11. 
8:32 p.m.
Zane Healy via cctalk wrote:
I've had two SD card failures in RPis that were continuously on in the
last two years.? One was my appletalk disk server; the other one was a
general NFS server that I used to backup stuff from my Sun+older unix
machines with looming scsi HD failures.? Kingston and Sandisk.? I should
migrate to SCSI2SD, but, this past experience makes me think that the
same failure mode would remain in place.
Perhaps using an USB HD for the served filesystems (or another ARM-based
board with a SATA interface) would be better.? I do have a Hummingboard
with SATA, but I was reserving it for other purposes.
carlos.
On Aug 8,
2019, at 4:29 PM, Frank McConnell via cctalk <cctalk at classiccmp.org> wrote:
On Aug 7, 2019, at 22:18, Adam Thornton wrote:
With my RPi2B, my wife tripped a breaker, and scrambled the card.
That was a Lexar SD card. Of course that was also one of the critical systems in my VMS
Cluster. I?ve moved to VM?s, on my VMware cluster, for all my OpenVMS systems. My one
Rpi3B runs Multics, the other TOPS-20. I?m thinking about rebuilding the RPi2B as a
PDP-11.
Zane https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Thanks. I had a brief look around the 36-bit systems last night.
One of my tests for a Pi 3 B was to build and run the SIMH HP3000 on it. Being able to
do that means having git to get the SIMH source, gcc and gmake to build, (curl|wget) to
get the MPE V/R bits, unzip to extract. The prerequisites were what I wanted for other Pi
stuff and this was a good workflow to find out what was missing from the Jessie Lite image
(yes it was that long ago, in Stretch and Buster I think I have found all those packages
are already present).
One of the things I have found with the Pi is, the low end micro SD cards (P*tr**t and
K*ngst*n would match the ones that did this) are lossy storage. It?s not that they wear
out, it?s that they lose bits. Switching power supply to one sufficient for the Pi did
not solve this problem, they continued to lose bits.
-Frank McConnell
4:53 p.m.
On Aug 7, 2019, at 10:18 PM, Adam Thornton via cctalk
<cctalk at classiccmp.org> wrote:
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and the Unix v7 account dmr have no
passwords.
Please treat the dmr account respectfully.
I will get to account requests?eventually, probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit ones on a Pi 3B+ and the 16-bit and 32-bit
ones on a Pi 2B+) on Debian Buster. Absolutely no guarantee of availability or usability
is made.
Adam
I?m curious, are you doing all that with just two Raspberry Pi?s, or do you have one per
each system you?re emulating?
Zane
11:52 p.m.
Thanks for putting it up. First time I've logged
onto old Unix in decades (should try getting my copy of V6 up on simh).
Have a couple of RasberryPi's kicking around that
just fired up once to play with. Only part that
simulation doesn't let you do is to connect up
all sorts of lab hardware to A/D's and
D/A's. Have lots of PDP-11 code that wrote in
1980's that can't use as no-one has written
additions to PDP-11 emulators which will make one
think one is dealing with 80's era data
acquisition hardware and digital I/O boards which
are far faster on modern microprocessors than there were then.
https://mvsevm.fsf.net
Currently, the TOPS-10 guest account (42,42) and
the Unix v7 account dmr have no passwords.
Please treat the dmr account respectfully.
I will get to account requests
eventually,
probably. TImeliness is not guaranteed. All
systems are hosted on Raspberry Pis (the 36-bit
ones on a Pi 3B+ and the 16-bit and 32-bit ones
on a Pi 2B+) on Debian Buster. Absolutely no
guarantee of availability or usability is made.
Adam
9 Aug
9 Aug
6:35 a.m.
On Aug 9, 2019, at 2:52 AM, Boris Gimbarzevsky via
cctalk <cctalk at classiccmp.org> wrote:
Thanks for putting it up. First time I've logged onto old Unix in decades (should
try getting my copy of V6 up on simh).
Have a couple of RasberryPi's kicking around that just fired up once to play with.
Only part that simulation doesn't let you do is to connect up all sorts of lab
hardware to A/D's and D/A's. Have lots of PDP-11 code that wrote in 1980's
that can't use as no-one has written additions to PDP-11 emulators which will make one
think one is dealing with 80's era data acquisition hardware and digital I/O boards
which are far faster on modern microprocessors than there were then.
I don't think writing SIMH device emulators for things like the DEC Unibus D/A and A/D
devices, or the DR11-A, would be at all hard.
Neat idea, actually. It would let me run the "LABBASIC" I created as an honors
project in college in 1974 -- a version of RT11 BASIC with added statements to drive those
data acquisition devices and the KW-11/P programmable clock.
paul
1:02 p.m.
I don't think writing SIMH device emulators for
things like the DEC Unibus D/A and A/D devices, or the DR11-A, would be at all hard.
Yes; I have also thought it would be nice to have something like a DR11-C in simh, which
could be configured to call into a dynamic library containing user code to emulate a
custom connected device.
--FritzM.
3:34 p.m.
On 09/08/2019 14:35, Paul Koning via cctalk wrote:
I don't think writing SIMH device emulators for
things like the DEC Unibus D/A and A/D devices, or the DR11-A, would be at all hard.
Sounds like a big challenge to me, but one that I would feel blessed
with having on my agenda. If it's not hard for you then you're the kind
of person I'd love to bump in to one day and end up having as a mentor :-)
Andrew
--
OpenPGP key: EB28 0338 28B7 19DA DAB0 B193 D21D 996E 883B E5B9
1745
days inactive
1746
days old
23 comments
17 participants
participants (17)
-
alexandre.tabajara＠gmail.com -
athornton＠gmail.com -
bfranchuk＠jetnet.ab.ca -
boris＠summitclinic.com -
brain＠jbrain.com -
ce.murillosanchez＠gmail.com -
couryhouse＠aol.com -
dseagrav＠lunar-tokyo.net -
fmc＠reanimators.org -
fritzm＠fritzm.org -
guykd＠optusnet.com.au -
healyzh＠avanthar.com -
huw.davies＠kerberos.davies.net.au -
imp＠bsdimp.com -
paulkoning＠comcast.net -
tdk.knight＠gmail.com -
ullbeking＠andrewnesbit.org