I use Tautulli monitoring.
For the past 4-5 weeks across 2-3 different server versions I’ve tried, Tautulli will randomly email me and state remote access is down. When this starts, it seem to ‘bounce’ notices several times in a row for a few hours ‘up’ ‘down’ ‘up’ ‘down’.
Sometimes, not always, if I go to the local server - settings - server - remote menu - it will state it’s not connected. If I click ‘retry’, most times, but not always, I’ll get the green checkmark back.
Viewing the Plex logs results in hundreds of the following appearing. WARN PubSub: Received notifyConnectivity event with incorrect async identifier <-- followed by a hex key.
Though my search of the forum found a few mentions of the same error, I’m unable to find work-around or solution.
Can anyone tell me what it is and how to correct it?
I’ve attempted rebooting all net equipment, and the server’s been restarted 2-3 times since this began.
Server is on ethernet, static ip (no dhcp), Win10, PMS is 1.13.3.5208, but 2-3 past versions also occurred.
I can look into the issue if you provide logs zip covering the time when the connection becomes unavailable
The async identifier errors is a known problem - a timing problem - but from the evidence seen so far, the impact is only the accuracy of the remote access availability indicator. There is no evidence to prove that it actually has an impact on server availability remotely
I’ll attempt to get logs the next time Plex (Tautulli) reports loss of remote - unfortunately, by the time I can get to logs, it’s been 4+ hours (during my sleep or at work), and due to the hundreds of async errors along with standard info - the actual time it lost connection is long gone from the 5 log file entry roll.
@sa2000 and @johnm_ColaSC
After almost 2 weeks of no issue, it finally happened again.
Logs attached. Network connection was up/operation entire time, as I could access Tautulli with no issue (and is how I restarted the server remotely using a script, as indicated in the DownUpTimes (that log is from Tautulli log.)
Server was updated recently, and a few plugins were updated, both requiring restart.
I’m not sure if it’s what I did last time, but once it’s started, it seems like restarting PMS is the only thing to correct it. (I’ve restarted at least 2-3 times in the past few days since this last occurred, so I don’t believe it’s a simple ‘run x days, you need to restart due to some single bug’.
[EDIT attachement removed due to not containing proper info - new logs attached later in posts]
I have the same issue. My server is on Ubuntu 16.04 LTS and is behind a Ubiquiti UniFi Security Gateway. Here are my log files. Plex Media Server Logs_2018-07-16_22-05-34.zip (955.9 KB)
@sa2000 - Debug enabled logs attached
Sorry they don’t have a clean start point once I enabled debug, but, the tautulli log DownUp text file should make it reasonably easy to tell what points to look inside the plex logs
I think I had more notifications by time I woke up this morning and through out the day than I did yesterday. Archive.zip (864.6 KB)
@sa2000 - just reporting in - since uploading the debug logs yesterday, I’ve had no issues with reported remote access, nor actually experienced any issue. (I did the day before during the up/down)
As stated, it’s seemingly random.
I have been looking at the logs. I am not really happy that tautulli is making a refreshReachability request every minute - that puts a big load on the pubsub servers - much more than what would be anticipated.
In future please always restart the server after enabling debug logging - to enable it in mid flight would not capture startup sequence
I will let you know when i conclude my investigation.
I also am having this exact issue at this very moment… PMS remote server disconnecting then reconnecting over and over… it is happening constantly. I have added a comment to a more recent thread here -
The problem happened before the period covered by the log you provided
Also I am not happy with the reachability checks being requested every minute by Tautulli. I do not believe our pubsub servers were sized to allow for every server to do a remote connectivity test via plex.tv and pubsub servers every minute of the day.
Though Tautulli could be over doing it on pubsub end, what would prevent me from actually being able to access my server at least 20% of the times Tautulli reports it is down? I’ve had 18 notifications today. During the last, i opened plex android and it made me do a reconnect to my server. Twice, to connect.
Thanks for the log
I can see that Plex Media Server set the port to 0 which may have led to the issue. I do not know why that arose - i have referred it to the development team
Jul 19, 2018 07:15:38.928 [0x7f361b7ff700] DEBUG - MyPlex: Sending Server Info to myPlex (user=xxxxxxxxxxxxxx@xxxx.xxx, ip=, port=0)
As mentioned in my PM I need to establish if the repeated connectivity requests resulting from the minute by minute Tautulli refreshReachability are giving rise to the loss of remote access (due to some timing bugs) - could you run without Tautulli for a few days - if you still lose remote access, please let me have a look at the logs
