Minor Bug in 1.5.5 - ping watchdog
Posted: Tue Dec 17, 2019 9:22 pm
So today I upgraded a couple of WS-150-DC-8's from 1.5.2 to 1.5.5 and I got a crapload of "port bounce" emails from one of the switches where that particular port(s) was NOT powered up.
The second switch had no ports powered down, so I can't claim it happens always.. But on the one switch where the issue did occur, two of the ports have been shut-off for about 2 weeks. We were having issues with persistent cloud cover and the two devices were not needed, so I powered them off. (having to worry about my solar power budget in San Diego is not a common occurrence).
The port-bounce messages (ping failure) were coming in every 5 minutes for about 3 hours. I reiterate, for ports that were NOT powered up. I disabled the ports in the Netonix and am waiting to see if they continue. I also disabled to Watchdog functionality for those two ports as well, but never had an issue before with the Watchdog tripping on ports that were powered off. (Freaked me out when I got back from working on that tower, after having verified everything was operating correctly, before I left, and then I return to the office to about 30 emails re: watchdog failure. Looked out the window at the pitch-black of night and thought "aww frick.. now I have to go back to that tower)..
Anyhow.. This didn't / has not occured on any other versions of the firmware that I am aware of, and all of my towers have at least 1 port shutdown (management loco m2) that I don't need to be powered up unless I am actually at the tower. The other upgrade today I powered up that port before driving to the tower so I'd have internet / network access).
For clarity, my previous upgrades to 1.5.5 did not involve towers with ANY ports off, because the weather was good. We've only had crap weather for the last couple of weeks. So this might only occur during an upgrade, but it certainly persists after the upgrade is complete.
The second switch had no ports powered down, so I can't claim it happens always.. But on the one switch where the issue did occur, two of the ports have been shut-off for about 2 weeks. We were having issues with persistent cloud cover and the two devices were not needed, so I powered them off. (having to worry about my solar power budget in San Diego is not a common occurrence).
The port-bounce messages (ping failure) were coming in every 5 minutes for about 3 hours. I reiterate, for ports that were NOT powered up. I disabled the ports in the Netonix and am waiting to see if they continue. I also disabled to Watchdog functionality for those two ports as well, but never had an issue before with the Watchdog tripping on ports that were powered off. (Freaked me out when I got back from working on that tower, after having verified everything was operating correctly, before I left, and then I return to the office to about 30 emails re: watchdog failure. Looked out the window at the pitch-black of night and thought "aww frick.. now I have to go back to that tower)..
Anyhow.. This didn't / has not occured on any other versions of the firmware that I am aware of, and all of my towers have at least 1 port shutdown (management loco m2) that I don't need to be powered up unless I am actually at the tower. The other upgrade today I powered up that port before driving to the tower so I'd have internet / network access).
For clarity, my previous upgrades to 1.5.5 did not involve towers with ANY ports off, because the weather was good. We've only had crap weather for the last couple of weeks. So this might only occur during an upgrade, but it certainly persists after the upgrade is complete.