Ignoring unexpected message in NotificationStream: hello
anyone have an idea?
I’ve seen the same messages, I’m wondering if this is anyway related to the playback issues we’re having on sonos.
I don’t have sonos…
And Plex hasn’t responded…
Some Plex clients send literal “hello” messages periodically to keep the communication channel alive. This probably could be filtered out by the server, but it isn’t unexpected or a problem.
It was a problem… hammering my system…
Are you using PlexAmp or PlexDash?
Were the log messages causing a problem themselves, or were you having another issue, and that’s why you were looking at the logs in the first place?
A normal healthy Plex system is quite chatty about logging. I wouldn’t even blink over a few log messages per second.
It would be good to know what problem it actually causing specifically. From what I know it is those apps trying to keep the websocket alive.
1 Like
I have PlexDash on my phone, but it’s not usually being used unless I need to check something when I’m away… Not sure the last time I checked if I was still getting the errors since no one responded… In the meantime I have set to block those messages to my servers…
The problem it was causing it was using nearly 90% cpu on the servers. Log doesn’t show all the hits, it was happening about every 600ms… Remote users couldn’t connect. why I was checking…
Anyway I solved it myself…
that seems very wrong.
perhaps a bugged client somewhere?
How did you block those messages? Reverse proxy, or just firewall the specific clients sending them?
To be clear, I’m not saying that you didn’t have a problem.
I’m saying that those messages themselves don’t appear to indicate a problem (because it’s been confirmed by Plex that they’re a normal part of a functioning system) and because Plex normally logs a LOT. I didn’t assume “high CPU” or “problem” when I read your initial message, only “What’s this funny log entry?”
I have a tap on the segment…
I pass traffic I know, block all other traffic including some Plex.TV ip’s
Plexamp and Plex Dash send these messages to keep the web-socket alive (and know it’s alive) every 15 seconds; they are an insignificant amount of data, and use trivial processing time on the server which should able to process literally thousands of websocket messages a second.
I find it bad that I posted this Jun 1st and had no replies for 2 months.
Not sure why this is of importance now when it wasn’t 2 months ago.
anyway I have my fix… and my idle server do sleep like they should.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
