13 Sep
2022
13 Sep
'22
5:08 p.m.
On Sep 13, 2022, at 10:55 AM, Robert Armstrong
<bob(a)jfcl.com> wrote:
FWIW, I have a very similar problem on ZITI with DECnet/Linux. Every night when Paul’s
mapper would run it would leave a zombie NML process hanging around, and after a couple of
weeks I’d have to reboot to get rid of them. I just stopped allowing NML access on ZITI
for that reason.
I’m not saying that Paul’s mapper is to blame – DECnet/Linux has tons of bugs and I’m
sure this is just another mistake somewhere. But it might be that the mapper is doing
something slightly different or unusual and that’s why we see these symptoms.
The mapper is very simple. Quoting from pydecnet/doc/mapper.txt:
The network scanner looks at every node mentioned in the node database
plus every node it can see as reachable. Roughly speaking, it makes a
NICE connection (remote network management protocol) to each node,
then asks it three questions: "show exec characteristics", "show
active circuits" and "show active nodes". The first questions answers
"what type of node is this" and also gets the node name if we don't
already know it. The other two give the adjacent nodes, i.e., the
connections from this node to its neighbors. Often but not always,
that information includes the node type of the adjacent nodes.
So to see what impact mapper polling would have on any given NML, you can simply use NCP,
issue "set exec node <nodename>" then issue the three commands shown
above.
paul