Some ports on a WS-12-250-DC unresponsive

[kiwitinker]

On a solar powered site we had/have an issue with very low battery levels due to lack of sun. The center switch is a WS-12-250DC (running 1.5.5) normally driving between 5 and 7 antennas on 24V.
This has been very reliable in the past. To conserve battery power I switched 2 off completely and set 3 of the other 5 ports to switch off during the night. This worked without a hitch for a couple of days. Then I introduced a low voltage switch off as well (well above the Netonix's limits at 18.9V and back on at 20.0V).
I watched the Netonix to switch off last night and then this morning as the sun was shining on the panels, the ports did power up when reaching 20.0V. So far so good.
Drama started, when the Ethernet side of things did not go well. It appeared that the antennas were running (using the correct amount of power), but I could not reach any of the three (even on a low level mac-ping).
I had no choice, but to drive out and climb the pole retrieving the antennas for a bench test. They worked flawless in the lab. Back to the site , installed the antennas plugged them in ... and no response. Then I plugged them into a different port that I had powered up previously and the antennas booted up plus the Ethernet worked fine.
Here is a part of the log when The Netonix switched the power on automatically:

Jun 30 09:32:39 switch[852]: Enabling PoE power
(24V) on port 1 (APSectorNorth) due to Input Voltage (20.0V) greater
than user specified Power Up voltage (20.0V)
Jun
30 09:32:39 switch[852]: Enabling PoE power (24V) on port 2
(PtgSchnapper) due to Input Voltage (20.0V) greater than user specified
Power Up voltage (20.0V)
Jun 30 09:32:39
switch[852]: Enabling PoE power (24V) on port 4 (PtgScotts) due to Input
Voltage (20.0V) greater than user specified Power Up voltage (20.0V)
Jun 30 09:32:41 sSMTP[1010]: Sent mail for sender@xxxxxxxx (221 2.0.0 Bye) uid=0 username=admin outbytes=713
Jun 30 09:32:41 sSMTP[1013]: Sent mail for sender@xxxxxxxx (221 2.0.0 Bye) uid=0 username=admin outbytes=711
Jun 30 09:32:41 sSMTP[1016]: Sent mail for sender@xxxxxxxx (221 2.0.0 Bye) uid=0 username=admin outbytes=705


Because of the lack of connectivity I manually tried to reboot the antennas:
Jun 30 09:40:46 UI: Configuration changed by admin (::ffff:10.10.10.7)
Jun 30 09:40:46 UI: Port 1 PoE: changed from '24V' to 'Off'
Jun 30 09:41:15 UI: Configuration changed by admin (::ffff:10.10.10.7)
Jun 30 09:41:15 UI: Port 1 PoE: changed from 'Off' to '24V'
Jun 30 09:41:19 switch[852]: PoE enabled on port 1, PoE Smart is starting cable check
Jun 30 09:47:49 UI: Configuration changed by admin (::ffff:10.10.10.7)

Later (still back at the lab) I pre-powered the spare ports on the Netonix (no antenna plugged in at this stage) - see this log

Jun 30 14:29:55 UI: Port 5 PoE: changed from 'Off' to '24V'
Jun 30 14:29:55 UI: Port 7 PoE: changed from 'Off' to '24V'
Jun 30 14:29:56 UI: Port 9 PoE: changed from 'Off' to '24V'
Jun 30 14:30:00 switch[852]: PoE enabled on port 5, PoE Smart is starting cable check
Jun 30 14:30:00 switch[852]: PoE enabled on port 7, PoE Smart is starting cable check
Jun 30 14:30:00 switch[852]: PoE enabled on port 9, PoE Smart is starting cable check
Jun 30 14:30:05 switch[852]: Port 5 cable check results: Open, Open, Open, Open
Jun 30 14:30:05 switch[852]: Port 5 passed PoE Smart cable check, turning on power
Jun 30 14:30:05 switch[852]: Port 7 cable check results: Open, Open, Open, Open
Jun 30 14:30:05 switch[852]: Port 7 passed PoE Smart cable check, turning on power
Jun 30 14:30:05 switch[852]: Port 9 cable check results: Open, Open, Open, Open
Jun 30 14:30:05 switch[852]: Port 9 passed PoE Smart cable check, turning on power

The big difference are the messages of checking the cable etc. These did NOT occur after the automatic power on earlier in the day.

Then back at the site. I plugged in the antennas into the 'original' ports and again nothing happened (they did power up and boot properly, but no connectivity!) and no log entry. After plugging in the cable into a pre-powered port (I know I should not, but I was desperate) the following log entries occoured:


Jun 30 15:39:36 Port: link state changed to 'up' (1G) on port 5
Jun 30 15:39:36 STP: msti 0 set port 5 to discarding
Jun 30 15:39:39 STP: msti 0 set port 5 to learning
Jun 30 15:39:39 STP: msti 0 set port 5 to forwarding
Jun 30 15:39:49 Port: link state changed to 'down' on port 5
Jun 30 15:39:49 STP: msti 0 set port 5 to discarding
Jun 30 15:39:52 Port: link state changed to 'up' (1G) on port 5
Jun 30 15:39:52 STP: msti 0 set port 5 to discarding
Jun 30 15:39:54 STP: msti 0 set port 5 to learning
Jun 30 15:39:54 STP: msti 0 set port 5 to forwarding
----------------------------------
Now finally my first question:
Currently I have 3 or 4 ports in the limbo state and no 'spare' left to play. What can I do to reactivate the logic of the port to re engage connectivity? In the Netonix web config panels ALL ports are set to equal values (except the name).


I am a little scared to switch off the antennas now as you can imagine. I would have to purchase a new switch and get it up to the site to solve any issues....??
Any ideas?
Cheers

[kiwitinker]

Just to confirm: the Netonix in question here is NOT configured to use VLAN. But having said that, the issue I found (detailed above) might be connected?