Massive Packet Loss with PMS is running

Plex Media Server Desktops & Laptops
1

Server Version#:1.24.4.5080
Player Version#:

Noticed some random buffering so I jumped on the box to check it out, inbound to the server and it has NO packet loss, Outbound has terrible packet loss with PMS running. All resolves if I stop PMS.

Downgraded to PMS version 1.23.3.4707 and issues persist (that was the farthest back my update folder had available to me.

Reviewed some of the other posts, disabled DLNA, GDM, no luck.

Windows 10 is up to date
Drivers / Firmware are up to date
Dell is about 5 years old running the onboard NIC

I’ve been using Plex for maybe 10 years, maybe longer, this may be my first issue. Let me know if I need to provide any additional information / logs etc.

2

Changed NICs, changed Ethernet cables; issues persist.

3

What CPU are you running?
How are you measuring the packet loss?

4

just ping to the default gateway and Google DNS servers.

Intel Core i7-6700

5

Is Plex streaming (transcoding) anything when you encounter packet loss?

What are your Plex Transcoder settings set to?

Do you have any Conversions in-progress?

Can you enable debug logging, then go to troubleshooting and download logs., then upload them here?

6

The packet loss is active even if there is no streaming period.

Transcoder is setup to Automatic.

No conversions

Yes here are the logsPlex Media Server Logs_2021-10-17_16-38-55.zip (3.7 MB)

7

I even rebuilt the PC here this afternoon thinking it was just some junk Windows installation issue; its bee running flawlessly for a couple of years now.

PC rebuilt, updated etc. Thought I was in the clear, no dice…

8

What do you mean “inbound” and “outbound” have packet loss?

Reboot router?

9

Inbound - pinging TO the PMS from another PC on my network - no loss
Outbound - pinging FROM the PMS to that same PC on my network - 60% loss

I’m not 100% on this, but the issues have subsided for the moment but have done two things.

  1. Changed the LOCAL IP address of my PMS box; I don’t think it had anything to do with a duplicate IP as it was just a DHCP reservation within my Eero so a conflict really shouldn’t have been possible as I don’t really set STATIC IPs anymore.

  2. This is what I believe to be the culprit; when I changed my LOCAL IP, I also disabled Remote Access. I don’t have any logs here but I think I may have been flooded with INBOUND traffic from a bot on the open port.

Killing the PMS server always fixed the packet loss, changing to a new IP locally would have broken my inbound rules, and then finally, having left the inbound rules OFF, no issues.

Not ideal for the long term but I’ll randomize my inbound port, and if I get port scanned and it happens again, at least I’ll know what the issue is and can call my ISP and request a new public IP from them.

Anyways, I think I went to alot of work for basically nothing; but hey, maybe this thread will help someone else?

Anyways, for now, I think that is all she wrote.

Thanks everyone!

1 Like
10

What’s interesting is that every ping requires bidirectional traffic. Packet goes, response comes back.

I do kinda like your theory that it was exacerbated by inbound requests/dos/spam/hackers.

Please update if you ever learn more.

11

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.