Johnny,
     I did notice one more unusual thing (I seem to always have the unusual things happen to me ;) during might reboot of CBVRP3 this morning. 

     Two nodes, STYX (Mark Curtis) and ONAPI4 (Wilm Boerhout) did not come up like the others and they had been up before the reboot.

     Checking the command file that brings up those HECnet circuits I noticed they cost had COSTS set explicitly with NCP:

NCP SET CIR IP-0-7 LEV ONE COST 9
NCP SET CIR IP-0-7 LEV TWO COST 9

    I commented these two lines that specified costs for each link and rebooted RSX again. Then these links then came up normally with the others.

>TYP NTD.SCR
3-MAR-2026  07:48:40 Node: CBVRP3(30.15) CBVRP3 RASPBERRYPI RSX11M+ V4.6
                 RSX-11M-Plus, Area Routing Node, DECnet V4.6
  LNKS   2 / 10 XXXXXXXX!-------------------------------   Alloc Fails:
  CCB   14 / 45 XXXXXXXXXXXX!---------------------------      0
  SDB    0 / 25 ----------------------------------------      0
  LDB    9 / 45 XXXXXXXX!-------------------------------      0 :    0

          Known Circuits:                     Network Processes/Tasks:
Circuit   Cost Size  Adj Node/Status        Name   TI  Links MBX XMT RCV

UNA-0       3        CBVRP3(30.15)         NTD...  CO0:    1   0   0   0
IP-0-0      6        On-starting           NTDT3   TT3:    1   0   0   1
IP-0-1      6   576  NICHTS(30.60)
IP-0-2      6   576  PYHEC ( 1.1022)
IP-0-3      6   576  PYRTR (31.3)
IP-0-4      6   576  PYTHON(41.1)
IP-0-5      6   576  KNIGHT(17.1023)
IP-0-6      6        On-starting
IP-0-7      6   576  STYX  (29.700)
IP-0-8      6   576  ONAPI4(29.206)
IP-0-9      6        On-starting               

   The last three On-Starting links are to nodes that I know are not online at the moment. 

Best,
Mark


On Mar 3, 2026, at 3:47 AM, Johnny Billquist <bqt@softjar.se> wrote:

On 03/03/2026 05.40, John H. Reinhardt wrote:
On 3/2/2026 6:17 PM, Johnny Billquist wrote:
On 2026-03-02 18:42, John H. Reinhardt wrote:
On 3/2/2026 10:44 AM, Johnny Billquist wrote:
Hm. The command "MNC SET CIR IP-0-5 MODE DDCMP" should have given a warning that this has been deprecated. In essence, it does nothing.

All links are just multinet links, which is DDCMP, with a small header to just get whole packets when using TCP.

  Johnny


That explains the "MLTNET" in Marks error log.  So, if I set up the phDECnet circuit as a Multinet one is should work?

Sortof yes, but no.
The errors are because RSX is receiving packets without that multinet header.

I misunderstood.  But figured because changing to Multinet on pyDECnet side did not make any difference.  Mark just gets a different error
> 10:45:36  MLTNET - IP-0-5 TCP ACCEPT failure. IOSB=-37
Oh, well.  Time to fire up a local SimH RSX node and see if I can figure out what I'm doing wrong.

Now, *that* is a really weird error. My best guess would be that configuration was changed, and things brought up while other/old links were still in the process of getting shut down, so it's in a bit of a messy state.

Probably a restart of things would clear it up.

Just fyi:

.err -37
000333 (-37): %I/O-F-IE.NLN, no file accessed on LUN
.


 Johnny

_______________________________________________
HECnet mailing list -- hecnet@lists.dfupdate.se
To unsubscribe send an email to hecnet-leave@lists.dfupdate.se