On 2012-12-27 21:56, Dave McGuire wrote:
On 12/27/2012 03:55 PM, Johnny Billquist wrote:
Just don't try using llogin towards an RSX host... There are bugs in the
lat implementation as well. It causes connections to be dropped if
connecting to RSX (for example). And when they are dropped, some bug in
the RSX LAT code is triggered, causing the buffers to be lost forever,
meaning you run out of memory after a while if you use the Linux LAT
package towards RSX.
I've seen this happen sporadically talking to RSTS/E and VMS systems
as well...if it's the same bug, it's not RSX-specific. When it happens
to me, ALL of my current llogin sessions drop at the same moment.
I didn't know that it happened on other systems as well. I assumed they had written the code testing against VMS, and had it running smooth in that scenario. That the same things happens there was news to me.
We really should write a new lat package for Unix. (If I only had time...)
Johnny
Will do- you must have a crapload of mail and stuff, this was like my third request :)
I'm not complaining, your administration of the network is admirable.
Sampsa
On 27 Dec 2012, at 22:59, Johnny Billquist <bqt at softjar.se> wrote:
Give me a couple of minutes to catch up on mail, and do/respond, will ya? :-)
Johnny
On 2012-12-27 21:42, sampsa at mac.com wrote:
Sorry to bug, but since you're around, can you add:
8.100 PPVX01
8.101 PPVX02
8.102 PPVX03
Sampsa
On 27 Dec 2012, at 22:41, Johnny Billquist <bqt at softjar.se> wrote:
On 2012-12-26 09:20, Rok Vidmar wrote:
The DECNET.TXT file obviously had incorrect (very old) information for area
42.
The same is true for Area 3:
Area 3 Patrick Caulfield
Chrissie kindly resigned this area to me.
Done.
Johnny
Give me a couple of minutes to catch up on mail, and do/respond, will ya? :-)
Johnny
On 2012-12-27 21:42, sampsa at mac.com wrote:
Sorry to bug, but since you're around, can you add:
8.100 PPVX01
8.101 PPVX02
8.102 PPVX03
Sampsa
On 27 Dec 2012, at 22:41, Johnny Billquist <bqt at softjar.se> wrote:
On 2012-12-26 09:20, Rok Vidmar wrote:
The DECNET.TXT file obviously had incorrect (very old) information for area
42.
The same is true for Area 3:
Area 3 Patrick Caulfield
Chrissie kindly resigned this area to me.
Done.
Johnny
On 27 Dec 2012, at 15:57, Dave McGuire <mcguire at neurotica.com> wrote:
On 12/27/2012 03:55 PM, Cory Smelosky wrote:
Too bad nobody is maintaining the Linux DECNET packages anymore.
Yeah the module doesn't build on newer kernels, iirc. I tried. ;)
Actually, it does, but procfs and sysfs or whatever have changed too much for it to be configured.
Define "newer"? It runs fine on 3.20.
Well I /was/ running Scientific Linux, so that could explain it. ;)
Do they muck with their kernels? If not, and it's likely they don't,
I don't see that having much to do with it.
I believe they do muck with their kernels, yes.
Maybe i'll reinstall that VM some time all it runs is the script you get when you ssh -p 2222 endpoint at dev.gimme-sympathy.org (with a pass of endpoint123).
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.
On 2012-12-27 19:38, Cory Smelosky wrote:
On 27 Dec 2012, at 06:05, sampsa at mac.com wrote:
Well it worked for about 3 logons. It's borked again.
Too bad nobody is maintaining the Linux DECNET packages anymore.
Yeah the module doesn't build on newer kernels, iirc. I tried. ;)
Actually, it does, but procfs and sysfs or whatever have changed too much for it to be configured.
Which is one of my main gripes with Linux. The whole kernel development is just a mess. They constantly design really bad solutions and implement them. Then they realize they suck, and they change the kernel interfaces to fix things. But that leaves lots of broken code around in different places.
The world would have been a better place without Linux...
Johnny
On 12/27/2012 03:55 PM, Cory Smelosky wrote:
Too bad nobody is maintaining the Linux DECNET packages anymore.
Yeah the module doesn't build on newer kernels, iirc. I tried. ;)
Actually, it does, but procfs and sysfs or whatever have changed too much for it to be configured.
Define "newer"? It runs fine on 3.20.
Well I /was/ running Scientific Linux, so that could explain it. ;)
Do they muck with their kernels? If not, and it's likely they don't,
I don't see that having much to do with it.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
On 12/27/2012 03:55 PM, Johnny Billquist wrote:
Just don't try using llogin towards an RSX host... There are bugs in the
lat implementation as well. It causes connections to be dropped if
connecting to RSX (for example). And when they are dropped, some bug in
the RSX LAT code is triggered, causing the buffers to be lost forever,
meaning you run out of memory after a while if you use the Linux LAT
package towards RSX.
I've seen this happen sporadically talking to RSTS/E and VMS systems
as well...if it's the same bug, it's not RSX-specific. When it happens
to me, ALL of my current llogin sessions drop at the same moment.
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
The LAT is mostly for incoming stuff, mainly a captive account on CHIMPY.
Sampsa
On 27 Dec 2012, at 22:55, Johnny Billquist <bqt at softjar.se> wrote:
Just don't try using llogin towards an RSX host... There are bugs in the lat implementation as well. It causes connections to be dropped if connecting to RSX (for example). And when they are dropped, some bug in the RSX LAT code is triggered, causing the buffers to be lost forever, meaning you run out of memory after a while if you use the Linux LAT package towards RSX.
Johnny
On 2012-12-27 13:21, sampsa at mac.com wrote:
At least LAT works...
sampsa
On 27 Dec 2012, at 13:05, sampsa at mac.com wrote:
Well it worked for about 3 logons. It's borked again.
Too bad nobody is maintaining the Linux DECNET packages anymore.
Sampsa
On 27 Dec 2012, at 11:56, sampsa at mac.com wrote:
I noticed a while back that incoming CTERM connections to PYFFLE:: (my Waffle BBS clone written in Python, http://en.wikipedia.org/wiki/Waffle_(BBS_software) ) weren't working.
So I reinstalled DECNET-Linux on the machine and it's all working again.
Happy days.
sampsa
On 27 Dec 2012, at 15:53, Dave McGuire <mcguire at neurotica.com> wrote:
On 12/27/2012 01:38 PM, Cory Smelosky wrote:
Well it worked for about 3 logons. It's borked again.
Too bad nobody is maintaining the Linux DECNET packages anymore.
Yeah the module doesn't build on newer kernels, iirc. I tried. ;)
Actually, it does, but procfs and sysfs or whatever have changed too much for it to be configured.
Define "newer"? It runs fine on 3.20.
Well I /was/ running Scientific Linux, so that could explain it. ;)
Someone really needs to pick it up and continue its maintenance. I'd
do it if I had some time and knew the protocols better. There really is
a need.
Yeah, maybe i'll bother some friends with protocol documentation and see if they're interested in picking it up. They probably wouldn't be interested though. :(
-Dave
--
Dave McGuire, AK4HZ
New Kensington, PA
--
Cory Smelosky
http://gewt.net/ Personal stuff!
http://gimme-sympathy.org/ My permanently-a-work-in-progress pet project.