On 2012-06-05 17:47, Mark Benson wrote:
NODE LIST AS FOLLOWS:
6.1 STAR69 (Area IV)
*6.2 QUIGON
*6.14
*6.15 PIVAX1
*Denotes part-time node (i.e. it's not on all the time)
PIVAX will be up a good lot of the time, although I take it down occasionally to test RasPi stuff.
So, who is 6.14? Care to present us?
(6.15 registered)
Johnny
--
Johnny Billquist || "I'm on a bus
|| on a psychedelic trip
email: bqt at softjar.se || Reading murder books
pdp is alive! || tryin' to stay hip" - B. Idol
NODE LIST AS FOLLOWS:
6.1 STAR69 (Area IV)
*6.2 QUIGON
*6.14
*6.15 PIVAX1
*Denotes part-time node (i.e. it's not on all the time)
PIVAX will be up a good lot of the time, although I take it down occasionally to test RasPi stuff.
--
Mark Benson
http://DECtec.info
Twitter: @DECtecInfo
HECnet: STAR69::MARK
Online Resource & Mailing List for DEC Enthusiasts.
On 2012-06-05 16:50, Paul_Koning at Dell.com wrote:
On Jun 5, 2012, at 5:29 AM, Mark Benson wrote:
...
I am looking at trying to mount an external hard drive or SSD to handle the disk images instead of the SD card. For some ^$%^$% reason you can't mount a disk as a user in Linux (I might be missing something, admittedly) like you can in RSX and VMS (again, more demonstration that UNIX sucks ;)) so I have to futz about as root to do that.
You can set that to be allowed with the "user" option, see "man fstab". It defaults to not allowed, which is the correct security answer. (I would assume the same is true in DEC operating systems that have protection mechanisms... it certainly is in RSTS.)
Anyone can mount a disk in RSX, assuming you "own" the disk, which basically boils down to knowing the disk label, and have the owner field of the disk structure allowing you.
Same for tapes...
It works nothing like in Unix.
Johnny
On Jun 5, 2012, at 11:03 AM, Mark Benson wrote:
On 5 Jun 2012, at 15:50, <Paul_Koning at Dell.com> <Paul_Koning at Dell.com> wrote:
On Jun 5, 2012, at 5:29 AM, Mark Benson wrote:
...
I am looking at trying to mount an external hard drive or SSD to handle the disk images instead of the SD card. For some ^$%^$% reason you can't mount a disk as a user in Linux (I might be missing something, admittedly) like you can in RSX and VMS (again, more demonstration that UNIX sucks ;)) so I have to futz about as root to do that.
You can set that to be allowed with the "user" option, see "man fstab". It defaults to not allowed, which is the correct security answer. (I would assume the same is true in DEC operating systems that have protection mechanisms... it certainly is in RSTS.)
That means you have to enable it for each volume? That's a pain when you are dealing with dynamic volume devices and USB. In VMS this kind of stuff is decided on the USER account, not on the OS setup. I can see why people love VMS so much now...
I believe you can do it for USB across the board, but I'm not sure of the details.
paul
Having lost my Multinet tunnel to Steve's SG1 node a while back I have been trying sporadically to fix it with no luck. I decided hang it today and went back to using the bridge program (which works fine) so I am happy to say Area 6 is back on HECnet.
There are a few caveats however...
If Chrissie still wants machines on HECnet she need to move the off Area 6. I can no longer offer remote area routing for other people without significant extra work and testing. As I also cannot guarantee my uptime at the moment as I am on a very scrappy ADSL service I think it is for the best if the nodes outside my LAN are relocated. I am sorry for this inconvenience.
As indicated by my comment above, access to my network area will as ever be sporadic, depending on power-cuts, ADSL outages and also might be disturbed by some work on my main server, which houses STAR69 and my bridge.
--
Mark Benson
http://DECtec.info
Twitter: @DECtecInfo
HECnet: STAR69::MARK
Online Resource & Mailing List for DEC Enthusiasts.
On 5 Jun 2012, at 15:50, <Paul_Koning at Dell.com> <Paul_Koning at Dell.com> wrote:
On Jun 5, 2012, at 5:29 AM, Mark Benson wrote:
...
I am looking at trying to mount an external hard drive or SSD to handle the disk images instead of the SD card. For some ^$%^$% reason you can't mount a disk as a user in Linux (I might be missing something, admittedly) like you can in RSX and VMS (again, more demonstration that UNIX sucks ;)) so I have to futz about as root to do that.
You can set that to be allowed with the "user" option, see "man fstab". It defaults to not allowed, which is the correct security answer. (I would assume the same is true in DEC operating systems that have protection mechanisms... it certainly is in RSTS.)
That means you have to enable it for each volume? That's a pain when you are dealing with dynamic volume devices and USB. In VMS this kind of stuff is decided on the USER account, not on the OS setup. I can see why people love VMS so much now...
--
Mark Benson
http://DECtec.info
Twitter: @DECtecInfo
HECnet: STAR69::MARK
Online Resource & Mailing List for DEC Enthusiasts.
On Jun 5, 2012, at 5:29 AM, Mark Benson wrote:
...
I am looking at trying to mount an external hard drive or SSD to handle the disk images instead of the SD card. For some ^$%^$% reason you can't mount a disk as a user in Linux (I might be missing something, admittedly) like you can in RSX and VMS (again, more demonstration that UNIX sucks ;)) so I have to futz about as root to do that.
You can set that to be allowed with the "user" option, see "man fstab". It defaults to not allowed, which is the correct security answer. (I would assume the same is true in DEC operating systems that have protection mechanisms... it certainly is in RSTS.)
paul
...
The doc says that RT and Unix do it differently (no WAIT). I
haven't seen the Unix code but I did see the one for RT (F/B
version), and indeed, no WAIT instruction there. I'm not
sure why not. RT11 S/J seems to just be full of spin loops,
no central idle of any kind that I can see.
paul
Paul,
I think that depends on whether or not the idle loop pattern (for the
console light display) is being used. The lights pattern most certainly
makes use of the WAIT instruction. The SJ monitor is a completely
different beast than FB and friends - no question about it!
-Steve
I was referring to RT V2, which is what I have at hand. But it sounds like it's still the same in later versions (except for having sysgen options -- no sysgen in V2). The thing to remember is that there are multiple ways to tweak the lights, at least on later PDP11s -- WAIT displays R0 (if the Data display switch, if there is one, is set to "Data Paths"), while writing to 177570 sets the Display register (which can also be displayed by setting the switch to Display Register). But some older machines don't have a display register and don't have that switch. The 11/20, if I remember right, always displays R0 at a WAIT and ignores writes to 177570. I forgot what the 11/05 and 11/10 do on a WAIT (since they only have a single display, not a separate address and data display).
paul
Dave McGuire wrote:
... @sys$manager:netconfig
FWIW, that's what I'd do too...
NETCONFIG will start off by purging your existing database anyway, so if
you find yourself needing to do this again you can skip that step...
Bob
On 2012-06-05 11:29, Mark Benson wrote:
On 4 Jun 2012, at 23:15, Johnny Billquist wrote:
Nice.
Yeah, trouble is because I am using my current RasPi to work on VAX stuff I need a second one to run my PDP-11 sim on now :D C'mon Farnell get it together!!
Indeed. The more machines, the merrier... :-)
It would be interesting to hear of a comparison. The "newer" VAXen are really not that bad. But of course, everything is relative...
Like I say, it's not chronically 'slow' it just has a few pregnant pauses. Example: when I an using the TCP/IP configuration script in VMS 7.3 the pauses between menu screens are quite long. Stuff at the DCL terminal on the other hand is very quick (DIR's of long directory lists, etc.). There is always a smidge of additional lag because I am using SSH and the RasPi is having to encode the SSH packets on the fly as it goes. Telnet is much faster so I may switch over to using the DZ11 terminal emulation to work on it. Only problem is that's not the Operator terminal on SimH so I can't do everything through it.
Ok.
Talk about doing a misread. I saw that as an RSX-11 MultiProcessor system. :-)
:D That'd be nice... but tricky ;)
simh don't have the code (yet). e11 do, however...
I would suspect that is because the CPU will still be just waiting for I/O lots of the time. Compilations, as well as running something like SYSGEN in RSX, is really I/O intensive. Not much CPU work in there.
I can tell when the computer is working on the SD card, the LED flashes, and some times it's still loading SD card data (disk images) and sometimes it's thinking really hard :) SYSGENs are snappy except for the time consuming parts (the actually assembly compiles for the Executive, Drivers, etc). The RAM is either DDR or DDR2 so there shouldn't be a speed issue there, the biggest bottleneck is really the SD card.
Even compilation in RSX is a lot of I/O. You can improve things if you do some magic, but reading sources is just a small part of a compilation. It's a lot of work files, and it's a lot of output. And lots of overlays in the compilers...
I am looking at trying to mount an external hard drive or SSD to handle the disk images instead of the SD card. For some ^$%^$% reason you can't mount a disk as a user in Linux (I might be missing something, admittedly) like you can in RSX and VMS (again, more demonstration that UNIX sucks ;)) so I have to futz about as root to do that.
Yes. Mounting in Unix-land have some potentially interesting implications which makes it "unsafe" to allow normal users to do it.
Johnny
