Remote follows fail. restart required.
Hi,
We have noticed that remote follows fails on our server and we need to restart Pleroma. After the restart things work for some days, but then it stops working. Log is full of this.
Sep 14 20:42:05 my_host pleroma[709]: 20:42:05.679 [warn] Pool worker for https:framatube.org:443: Client #PID<0.31413.87> died before releasing the connection with :normal
Sep 14 20:42:05 my_host pleroma[709]: 20:42:05.681 [error] GenServer {Pleroma.Gun.ConnectionPool, "https:framatube.org:443"} terminating
Sep 14 20:42:05 my_host pleroma[709]: ** (ArgumentError) argument error
Sep 14 20:42:05 my_host pleroma[709]: :erlang.demonitor(nil)
Sep 14 20:42:05 my_host pleroma[709]: (pleroma 2.1.0) lib/pleroma/gun/connection_pool/worker.ex:86: Pleroma.Gun.ConnectionPool.Worker.handle_call/3
Sep 14 20:42:05 my_host pleroma[709]: (pleroma 2.1.0) lib/pleroma/gun/connection_pool/worker.ex:49: Pleroma.Gun.ConnectionPool.Worker.handle_cast/2
Sep 14 20:42:05 my_host pleroma[709]: (stdlib 3.12.1) gen_server.erl:637: :gen_server.try_dispatch/4
Sep 14 20:42:05 my_host pleroma[709]: (stdlib 3.12.1) gen_server.erl:711: :gen_server.handle_msg/6
Sep 14 20:42:05 my_host pleroma[709]: (stdlib 3.12.1) proc_lib.erl:259: :proc_lib.wake_up/3
Sep 14 20:42:05 my_host pleroma[709]: Last message: {:DOWN, #Reference<0.3939071935.232521731.178536>, :process, #PID<0.31413.87>, :normal}
Sep 14 20:42:05 my_host pleroma[709]: State: %{client_monitors: %{}, key: "https:framatube.org:443", timer: nil}
Pleroma 1.2.0 OTP
Debian 10
We are using config :tesla, :adapter, Tesla.Adapter.Gun
to resolve this issue #2093 (closed)
Any thought would be appreciated.