I have just upgraded to 25.1.10 last night, and immediately after rebooting, the memory usage keeps increasing.
Free memory declines until the system is unusable, and I need to hard reboot the box.
I am seeing lots of the following in System-Diagnostics-Activity:
331 root 20 0 220M 130M wait 3 0:00 0.44% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 20 0 220M 130M wait 0 0:00 0.40% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 20 0 220M 130M wait 0 0:00 0.27% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 68 0 220M 130M accept 2 0:01 0.25% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 20 0 220M 130M wait 2 0:00 0.21% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 68 0 220M 130M wait 2 0:00 0.20% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 20 0 220M 130M wait 0 0:00 0.19% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
331 root 68 0 220M 130M wait 3 0:00 0.13% /usr/local/bin/python3 /usr/local/opnsense/service/configd.py console (python3.11){python3.11}
I am not sure how to get a handle on this.
I am running the following services, this has been the case for a long time. All updates are applied:
acme ACME client
configd System Configuration Daemon
cron Cron
ddclient ddclient
dhcpd DHCPv4 Server
login Users and Groups
monit Monit System Monitoring
nginx Reverse Proxy and Web Server
ntpd Network Time Daemon
openssh Secure Shell Daemon
openvpn OpenVPN server MyServer
pf Packet Filter
routing System routing
sysctl System tunables
syslog-ng Syslog-ng Daemon
telegraf telegraf agent
unbound Unbound DNS
webgui Web GUI
FYI: Did a full reinstall from USB Live, imported my config. same issue.
Looks like disabling the WAN_DHCP6 gateway did the trick. Not sure why this was enabled (could this be enabled by default and not have caused issues in prev. builds?).
https://forum.opnsense.org/index.php?topic=46184.msg231245#msg231245
Hello, it s my 1st time that I see my VM 25.11.1 rebooting within my Proxmox. I have not yet been able to understand why (no crash core dump).
I have disabeled IPv6 GW for now but I still see my mem 90% used in Proxmox which is OK but can"t seem to understand why it crashed.
root@OPNsense:~ # top -o res
last pid: 18374; load averages: 0.16, 0.14, 0.15 up 0+00:32:05 16:18:44
83 processes: 1 running, 82 sleeping
CPU: 0.0% user, 0.0% nice, 0.0% system, 0.0% interrupt, 100% idle
Mem: 303M Active, 425M Inact, 1448M Wired, 9687M Free
ARC: 613M Total, 121M MFU, 424M MRU, 15M Anon, 4797K Header, 47M Other
489M Compressed, 1400M Uncompressed, 2.86:1 Ratio
PID USERNAME THR PRI NICE SIZE RES STATE C TIME WCPU COMMAND
62477 root 15 36 0 1401M 116M kqread 3 0:04 0.00% crowdsec
90642 unbound 4 20 0 120M 64M kqread 2 0:00 0.00% unbound
69639 root 10 20 0 1212M 57M kqread 3 0:01 0.00% crowdsec-firewall-b
98397 root 1 20 0 79M 52M nanslp 1 0:05 0.00% php
377 root 1 68 0 88M 46M accept 0 0:02 0.00% python3.11
22958 root 1 20 0 66M 40M accept 0 0:00 0.00% php-cgi
83348 root 1 23 0 51M 40M nanslp 1 2:25 0.00% python3.11
28331 root 1 20 0 64M 39M accept 0 0:00 0.00% php-cgi
28026 root 1 20 0 64M 39M accept 2 0:00 0.00% php-cgi
25618 root 1 20 0 68M 36M accept 2 0:00 0.00% php-cgi
22629 root 1 20 0 64M 36M accept 2 0:00 0.00% php-cgi
24001 root 1 20 0 62M 34M accept 0 0:00 0.00% php-cgi
25014 root 1 20 0 62M 34M accept 3 0:00 0.00% php-cgi
26897 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
26019 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
26714 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
26207 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
19937 root 1 68 0 56M 28M wait 0 0:00 0.00% php-cgi
25999 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
25255 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
25367 root 1 68 0 56M 28M accept 1 0:00 0.00% php-cgi
21145 root 1 68 0 56M 28M wait 3 0:00 0.00% php-cgi
23418 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
23220 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
24674 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
20534 root 1 68 0 56M 28M wait 1 0:00 0.00% php-cgi
21377 root 1 68 0 56M 28M wait 3 0:00 0.00% php-cgi
30005 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
29462 root 1 68 0 56M 28M accept 2 0:00 0.00% php-cgi
28931 root 1 68 0 56M 28M accept 0 0:00 0.00% php-cgi
62981 nobody 6 20 0 1233M 21M kqread 0 0:01 0.00% node_exporter
93059 root 1 20 0 32M 18M nanslp 1 0:01 0.00% python3.11
375 root 1 68 0 27M 15M wait 0 0:00 0.00% python3.11
16232 root 1 20 0 28M 15M select 2 0:00 0.00% python3.11
17419 root 3 20 0 45M 15M kqread 0 0:00 0.00% syslog-ng
15648 root 1 20 0 27M 14M select 2 0:00 0.00% python3.11
71043 redis 4 20 0 37M 12M kqread 1 0:01 0.00% redis-server
36847 dhcpd 1 20 0 28M 10M select 2 0:00 0.00% dhcpd
19715 root 1 20 0 23M 10M kqread 0 0:00 0.00% lighttpd
17134 root 1 68 0 24M 10M wait 0 0:00 0.00% syslog-ng
54091 root 1 20 0 20M 9060K select 2 0:00 0.00% sshd-session
16346 root 1 20 0 20M 9052K select 0 0:00 4.96% sshd-session
77473 root 2 20 0 21M 8756K nanslp 1 0:00 0.00% monit
53631 root 1 37 0 20M 8748K select 2 0:00 0.00% sshd-session
Is there any idea?
Your problem has nothing to do with the OP's problem. Please open new a new thread in such cases.
Obviously, you have allocated 10 GByte of virtual memory, of which most is free. Memory usage of FreeBSD within a Proxmox VM will always be shown as ~90% in Proxmox, regardless of how much is really in use within the VM.
If your VM stops working, then it is because you overcommitted the memory and then, the OOM reaper will stop VM processes as needed. You should be fine with much less allocated memory for your OpnSense VM. 4 GByte should usually be enough, so your overall memory consumption will be 6 Gbytes less and if that can be met, there will be no more process killing.