Listen queue overflow

bigsy · May 25, 2020, 6:31 AM

Running pfBlockerNG-devel 2.2.5_32 on 2.4.5-RELEASE (amd64).

I've noticed my logs are filling up with messages along the lines of:

May 25 07:24:33 192.168.64.1 kernel: sonewconn: pcb 0xfffff800663993a0: Listen queue overflow: 193 already in queue awaiting acceptance (605 occurrences)
May 25 07:25:39 192.168.64.1 kernel: sonewconn: pcb 0xfffff800663993a0: Listen queue overflow: 193 already in queue awaiting acceptance (5 occurrences)
May 25 07:26:39 192.168.64.1 kernel: sonewconn: pcb 0xfffff800663993a0: Listen queue overflow: 193 already in queue awaiting acceptance (915 occurrences)

I think this is related to the ports DNSBL is listening on:

[2.4.5-RELEASE][root@x]/root: netstat -Lan
Current listen queue sizes (qlen/incqlen/maxqlen)
Proto Listen                           Local Address
tcp4  0/0/128                          10.99.99.1.443
tcp4  193/0/128                        *.8443
tcp4  163/0/128                        *.8081
tcp4  0/0/128                          127.0.0.1.953
tcp4  0/0/128                          *.53
tcp6  0/0/128                          *.53
tcp6  0/0/128                          *.80
tcp4  0/0/128                          *.80
tcp6  0/0/128                          *.443
tcp4  0/0/128                          *.443
tcp4  0/0/128                          *.22
tcp6  0/0/128                          *.22

Any pointers to what's going on?

hammer73 · Jun 26, 2021, 5:23 AM

@bigsy I see the same errors...

jdeloach · Jun 26, 2021, 6:14 PM

This post is deleted!

hammer73 · Jun 26, 2021, 6:14 PM

@jdeloach FWIW I am running the latest versions. PFS+ 21.05 Release (on a 2100) and pfBlockerNG-devel 3.0.0_16.

While I was troubleshooting a connectivity issue on one of my two WANs (PPPoE) I noticed the Listen Queue errors. I'm not aware that they were related to my issue but I did switch from DNSBL Unbound Python Mode (which I understand is beta) to Unbound Mode nd rebooted. Currently don't see the error appear and connectivity is restored.

This change also seemed to have stopped my disk volume from slowly filling up. While using Unbound Python, changing log settings (or even clearing logs) didn't help. Only a reboot would bring the disk usage back to normal. I tried to locate the directory where the storage was being hogged but curiously, I could not find that either. (Note my skills are 'fair', certainly not 'expert'. ) df and du sorted by largest etc didn't help.

Currently all is good so ... all the above... FWIW..
Tnx!

bigsy · Jun 26, 2021, 6:20 PM

@jdeloach I posted that message in May 2020.