Closed walterav1984 closed 2 years ago
Given we haven't seen any complaints about this in the past two weeks, I am closing this as resolved.
On a side note, the underlying issue is bytes being dropped on the RX side. The packetizer won't catch that, and will keep reading packets with an off-by-n offset. If we see that in the wild, then we'll have to open a new bug.
ZyXEL GS1900-24HPv1 realtek-poe versions from the beginning till CI41 affected
After a couple of PoE device port swaps (re)connecting the
ubus call poe info
won't reflect/update the current correct status for example if port 13 and port 14 were connected delivering PoE and I swap over devices to port 15 and 16 the status would still report port 13 and port 14 being active and delivering power. The PoE mechanism is still active though and new (re)connected devices still receive power.The only remedy for this like the all port status
unknown
issue #10 is to restart not reload the/etc/init.d/poe restart
service. The root cause might even be the same as issue #10 because that eventually is also a hang/stuck on realtime refresh of port status.Sometimes you can port swap more than 24 for ports before this issue shows up, and sometimes after 2 or 3 port swaps.