[GH-ISSUE #4229] proxies randomly die over time but ~20% of them stay up #3332

Closed
opened 2026-05-05 14:08:57 -06:00 by gitea-mirror · 2 comments
Owner

Originally created by @modpotato on GitHub (May 21, 2024).
Original GitHub issue: https://github.com/fatedier/frp/issues/4229

Bug Description

after a few days (or hours) of running frp it will randomly close a large amount of connections

frpc Version

0.56.0

frps Version

0.56.0

System Architecture

linux/amd64

Configurations

server:

# frps.toml
bindPort = 7000
kcpBindPort = 7000
vhostHTTPSPort = 8080
allowPorts = [
    {single = 2022},
    {start = 4000, end = 5000},
    {single = 8080}
]
# The default value is 127.0.0.1. Change it to 0.0.0.0 when you want to access it from a public network.
webServer.addr = "0.0.0.0"
webServer.port = 7500
# dashboard's username and password are both optional
webServer.user = "admin"
webServer.password = "admin"

auth.additionalScopes = ["HeartBeats"]
auth.method = "token"
auth.token = "authtoken"

client

# frpc.toml
serverAddr = "15.204.234.35" # Replace with frps server IP
serverPort = 7000
transport.protocol = "kcp"

auth.additionalScopes = ["HeartBeats"]
auth.method = "token"
auth.token = "authtoken"


# TCP port range mapping
{{range $val := parseNumberRangePair "4000-5000" "4000-5000"}}
[[proxies]]
name = "tcp-{{$val.First}}"
type = "tcp"
localPort = {{$val.First}}
remotePort = {{$val.Second}}
{{end}}

# Additional TCP ports
[[proxies]]
name = "tcp-2022"
type = "tcp"
localPort = 2022
remotePort = 2022

[[proxies]]
name = "tcp-8080"
type = "https"
customDomains = ["website.com"]

[proxies.plugin]
type = "https2http"
crtPath = ""
keyPath = ""
localAddr = "127.0.0.1:8080"

# UDP port range mapping
{{range $val := parseNumberRangePair "4000-5000" "4000-5000"}}
[[proxies]]
name = "udp-{{$val.First}}"
type = "udp"
localPort = {{$val.First}}
remotePort = {{$val.Second}}
{{end}}


Logs

im only going to cite relevant logs, lmk if more are needed (10k lines plus)

May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.855 [W] [client/control.go:127] [22197ea8e7593c30] start new connection to server error: session shutdown
May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.348 [I] [proxy/udp.go:142] [22197ea8e7593c30] [udp-4021] writer goroutine for udp work connection closed
May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.348 [W] [proxy/udp.go:128] [22197ea8e7593c30] [udp-4178] read from workConn for udp error: EOF

all 3 of these are spammed (aside from these the others are just random requests that didnt reach an endpoint)

Steps to reproduce

  1. load configs
  2. use ovhcloud vps for server
  3. use local for client
  4. wait

Affected area

  • Docs
  • Installation
  • Performance and Scalability
  • Security
  • User Experience
  • Test and Release
  • Developer Infrastructure
  • Client Plugin
  • Server Plugin
  • Extensions
  • Others
Originally created by @modpotato on GitHub (May 21, 2024). Original GitHub issue: https://github.com/fatedier/frp/issues/4229 ### Bug Description after a few days (or hours) of running frp it will randomly close a large amount of connections ### frpc Version 0.56.0 ### frps Version 0.56.0 ### System Architecture linux/amd64 ### Configurations # server: ``` # frps.toml bindPort = 7000 kcpBindPort = 7000 vhostHTTPSPort = 8080 allowPorts = [ {single = 2022}, {start = 4000, end = 5000}, {single = 8080} ] # The default value is 127.0.0.1. Change it to 0.0.0.0 when you want to access it from a public network. webServer.addr = "0.0.0.0" webServer.port = 7500 # dashboard's username and password are both optional webServer.user = "admin" webServer.password = "admin" auth.additionalScopes = ["HeartBeats"] auth.method = "token" auth.token = "authtoken" ``` # client ``` # frpc.toml serverAddr = "15.204.234.35" # Replace with frps server IP serverPort = 7000 transport.protocol = "kcp" auth.additionalScopes = ["HeartBeats"] auth.method = "token" auth.token = "authtoken" # TCP port range mapping {{range $val := parseNumberRangePair "4000-5000" "4000-5000"}} [[proxies]] name = "tcp-{{$val.First}}" type = "tcp" localPort = {{$val.First}} remotePort = {{$val.Second}} {{end}} # Additional TCP ports [[proxies]] name = "tcp-2022" type = "tcp" localPort = 2022 remotePort = 2022 [[proxies]] name = "tcp-8080" type = "https" customDomains = ["website.com"] [proxies.plugin] type = "https2http" crtPath = "" keyPath = "" localAddr = "127.0.0.1:8080" # UDP port range mapping {{range $val := parseNumberRangePair "4000-5000" "4000-5000"}} [[proxies]] name = "udp-{{$val.First}}" type = "udp" localPort = {{$val.First}} remotePort = {{$val.Second}} {{end}} ``` ### Logs im only going to cite relevant logs, lmk if more are needed (10k lines plus) May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.855 [W] [client/control.go:127] [22197ea8e7593c30] start new connection to server error: session shutdown May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.348 [I] [proxy/udp.go:142] [22197ea8e7593c30] [udp-4021] writer goroutine for udp work connection closed May 21 01:58:26 frpc[70729]: 2024-05-21 01:58:26.348 [W] [proxy/udp.go:128] [22197ea8e7593c30] [udp-4178] read from workConn for udp error: EOF all 3 of these are spammed (aside from these the others are just random requests that didnt reach an endpoint) ### Steps to reproduce 1. load configs 2. use ovhcloud vps for server 3. use local for client 4. wait ### Affected area - [ ] Docs - [ ] Installation - [X] Performance and Scalability - [ ] Security - [X] User Experience - [ ] Test and Release - [ ] Developer Infrastructure - [ ] Client Plugin - [ ] Server Plugin - [ ] Extensions - [ ] Others
gitea-mirror 2026-05-05 14:08:57 -06:00
Author
Owner

@xqzr commented on GitHub (May 21, 2024):

#4223

<!-- gh-comment-id:2122327537 --> @xqzr commented on GitHub (May 21, 2024): #4223
Author
Owner

@github-actions[bot] commented on GitHub (Jun 12, 2024):

Issues go stale after 21d of inactivity. Stale issues rot after an additional 7d of inactivity and eventually close.

<!-- gh-comment-id:2161846351 --> @github-actions[bot] commented on GitHub (Jun 12, 2024): Issues go stale after 21d of inactivity. Stale issues rot after an additional 7d of inactivity and eventually close.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/frp#3332
No description provided.