Last Sunday Tom WA7TBP alerted me to a problem with W7PFB-11 where it was allowing connections but reporting it could not connect to a CMS.

Rebooting the Raspberry Pi that hosts W7PFB-11 did not solve the problem. Further, I could not SSH into the Raspberry Pi to try to diagnose the problem, and until this morning I had no free time to do more than desultory debugging.

This morning I first tackled the problem in earnest. After a bunch of dorking around I finally realized that all of the other computers that route through one specific 100Base-T switch also seemed to have no connectivity to the internet (or, indeed, to any of the rest of my home network.) This took me an embarrassingly long time to discover (more than an hour of head scratching).

So I checked on the switch, which was essentially hidden in the bottom drawer of a locked filing cabinet in a disused lavatory with a sign on the door reading “Beware of the Leopard”. The switch was alive in the sense that the power LED was illuminated but dead in the sense that it was not routing packets. Power cycling the switch changed the switch from “mostly dead” to “working but now highly suspect and not to be trusted”.

Anyway, the RMS station W7PFB-11 now appears to be working fine. When I get a moment I will order a replacement switch and also route the wiring for the rack so that it goes directly to my main switch instead of going higgledy-piggely thru some indeterminate number of switches.

Memo to self: it does not help if you have backup power at the equipment rack to keep the RMS station running, and backup power at the main switch and network link to keep that running, if your traffic from the RMS will be routed through a switch that is not connected to some sort of backup power! Bad Paul! No biscuit!

Eventually W7PFB-10 will be restored to service but not until I screw up my courage to wrestle with BPQ configuration again. Shawn K7ATA will attest to the horrors that lurk there.

-p W7PFB
73, Don’t forget to smile and have fun!