Complete new install to PFsense CE 2.8.1-RELEASE, soon PHP-FPM fails, pfsense webgui unavailable 50x Error

Follow
Chris Lawsonn
  • Edited

Complete new install to PFsense 2.8.1-RELEASE Community Edition, pfsense configuration is complete absolute initial install factory defaults (of course, followed instructions, Hardware Offloading disabled). Only 1 Wan, 1 Lan interface, no vlans, nothing else configured, Zenarmor was immediately installed to test and trial.

dash.zenarmor.com shows engine  Version: 2.4 - Feb 25, 2026 9:03 AM

Zenarmor has been set and tested using both Deployment modes:

Routed Mode (L3 Mode, Reporting + Blocking) with native netmap driver Native netmap

Routed Mode (L3 Mode, Reporting + Blocking) with emulated netmap driver Native netmap

 

The pfsense hardware: Intel nics, 8 CPUs, 32GB ram, 500gb hdd.

A while after Zenarmor is running soon PHP-FPM fails, pfsense webgui unavailable 50x Error.

Restarting PHP-FPM temporarily resolves issue.

How to find root cause, what is happening why Zenarmor causing issue? Reviewing pfsense logs cannot see anything abnormal. What directories, what traces, what to look for to investigate this issue further?

Thank you

0

Comments

2 comments

Date Votes
  • Official comment
    SVN Support Team

    Hi,

    Thank you for reaching out and bringing this to our attention.

    Could you provide details about the resource usage when the issue occurred? Was there a spike in CPU or RAM usage?

    Best regards

    Comment actions Permalink
  • Chris Lawsonn


    Hi, I appreciate you responding. 

    I cannot find any spike in CPU or RAM usage when issue occurred. The machine has 4 physical cpu cores (xeon e3) with hyperthreading total 8 logical cores and 32gb of ecc ram of which only 5% ram used when idling. Pfsense and zenarmor are running directly on the physical hardware, no virtualization, no hypervisors.

    The system is a standalone machine with a vanilla factory install of pfsense with practically no configuration changes so I have not yet setup any additional monitoring or logging to send off syslog or metrics to another system for more granular inspection. Using the default built-in pfsense monitoring tools:
    monitoring > system > processor > Processes shows ~480 average processes running.

    The machine is not under any heavy load, only 1 device connected to it for testing performing downloading random binaries, uploading files, streaming youtube etc using a WAN connection less than 500mbit.  

    I don't yet have a more precise point in time picture of resource usage when the issue occurred other than what pfsense monitoring logs show which pfsense status monitoring graph data does not identify any outlier activity, however after a day passes pfsense loses more granular historical data points. I am leaving the system running to reproduce the event again and will try to setup resource monitoring and alerting to hopefully catch the event again and capture more detailed resource usage if/when occurs.

    I reviewed system logs further and did find the following abnormal activity...


    Feb 28 02:12:04     nginx         2026/02/28 02:12:04 [error] 76241#100363: *48953 connect() to unix:/var/run/php-fpm.socket failed (61: Connection refused) while connecting to upstream, client: 192.168.1.10, server: , request: "GET / HTTP/2.0", upstream: "fastcgi://unix:/var/run/php-fpm.socket:", host: "192.168.1.1" 


    Feb 28 02:12:05     nginx         2026/02/28 02:12:05 [error] 76241#100363: *48953 connect() to unix:/var/run/php-fpm.socket failed (61: Connection refused) while connecting to upstream, client: 192.168.1.10, server: , request: "GET / HTTP/2.0", upstream: "fastcgi://unix:/var/run/php-fpm.socket:", host: "192.168.1.1" 


    Feb 28 02:12:04     check_reload_status     26754     Could not connect to /var/run/php-fpm.socket
    Feb 28 02:12:04     check_reload_status     26754     Could not connect to /var/run/php-fpm.socket
    Feb 28 02:12:04     check_reload_status     26754     Could not connect to /var/run/php-fpm.socket
    Feb 28 02:12:04     check_reload_status     26754     Could not connect to /var/run/php-fpm.socket 


    Feb 28 02:11:08     kernel         sonewconn: pcb 0xfffff8000ad2e780 (local:/var/run/php-fpm.socket): Listen queue overflow: 193 already in queue awaiting acceptance (232 occurrences), euid 0, rgid 0, jail 0 


    Mar 2 22:57:44     php-fpm     18443     /rc.linkup: The command '/usr/local/sbin/unbound -c /var/unbound/unbound.conf' returned exit code '1', the output was '[1772513864] unbound[48843:0] warning: setsockopt(..., SO_SNDBUF, ...) was not granted: No buffer space available [1772513864] unbound[48843:0] warning: so-sndbuf 4194304 was not granted. Got 57344. To fix: start with root permissions(linux) or sysctl bigger net.core.wmem_max(linux) or kern.ipc.maxsockbuf(bsd) values. or set so-sndbuf: 0 (use system value). [1772513864] unbound[48843:0] error: bind: address already in use [1772513864] unbound[48843:0] fatal error: could not open ports' 


    Mar 2 22:58:00     php-fpm     94607     /rc.linkup: The command '/usr/local/sbin/unbound -c /var/unbound/unbound.conf' returned exit code '1', the output was '[1772513880] unbound[49455:0] warning: setsockopt(..., SO_SNDBUF, ...) was not granted: No buffer space available [1772513880] unbound[49455:0] warning: so-sndbuf 4194304 was not granted. Got 57344. To fix: start with root permissions(linux) or sysctl bigger net.core.wmem_max(linux) or kern.ipc.maxsockbuf(bsd) values. or set so-sndbuf: 0 (use system value). [1772513880] unbound[49455:0] error: bind: address already in use [1772513880] unbound[49455:0] fatal error: could not open ports' 

     

    Mar 2 22:58:02     php-fpm     438     /rc.linkup: The command '/usr/local/sbin/unbound -c /var/unbound/unbound.conf' returned exit code '1', the output was '[1772513882] unbound[68503:0] warning: setsockopt(..., SO_SNDBUF, ...) was not granted: No buffer space available [1772513882] unbound[68503:0] warning: so-sndbuf 4194304 was not granted. Got 57344. To fix: start with root permissions(linux) or sysctl bigger net.core.wmem_max(linux) or kern.ipc.maxsockbuf(bsd) values. or set so-sndbuf: 0 (use system value). [1772513882] unbound[68503:0] error: bind: address already in use [1772513882] unbound[68503:0] fatal error: could not open ports' 


    Mar 2 22:58:21     php-fpm     42529     /rc.linkup: The command '/usr/local/sbin/unbound -c /var/unbound/unbound.conf' returned exit code '1', the output was '[1772513901] unbound[12355:0] warning: setsockopt(..., SO_SNDBUF, ...) was not granted: No buffer space available [1772513901] unbound[12355:0] warning: so-sndbuf 4194304 was not granted. Got 57344. To fix: start with root permissions(linux) or sysctl bigger net.core.wmem_max(linux) or kern.ipc.maxsockbuf(bsd) values. or set so-sndbuf: 0 (use system value). [1772513901] unbound[12355:0] error: bind: address already in use [1772513901] unbound[12355:0] fatal error: could not open ports'

    Mar 2 22:58:21     php-fpm     94607     /rc.newwanip: The command '/usr/local/sbin/unbound -c /var/unbound/unbound.conf' returned exit code '1', the output was '[1772513901] unbound[13013:0] warning: setsockopt(..., SO_SNDBUF, ...) was not granted: No buffer space available [1772513901] unbound[13013:0] warning: so-sndbuf 4194304 was not granted. Got 57344. To fix: start with root permissions(linux) or sysctl bigger net.core.wmem_max(linux) or kern.ipc.maxsockbuf(bsd) values. or set so-sndbuf: 0 (use system value). [1772513901] unbound[13013:0] error: bind: address already in use [1772513901] unbound[13013:0] fatal error: could not open ports' 

     

     

    0
    Comment actions Permalink

Please sign in to leave a comment.

Didn't find what you were looking for?

New post
Powered by Zendesk