Hi all,
I've been experiencing an issue with my OPNsense firewall where the system becomes unresponsive approximately every 2–3 weeks. The most recent incident occurred on April 5th, with the system locking up around 01:00 and remaining unresponsive until I manually rebooted it just after 06:00.
System Info:
OPNsense Version: 25.1.1 (amd64) with Zenarmor 1.18.6 - Feb 10, 2025
FreeBSD 14.2-RELEASE-p2
CPU: Intel Core i5-3450
Memory: 16 GB
Storage: 256 GB SSD (SMART tests passed)
ZFS file system
What I've Done So Far:
Log Collection & Review:
Collected /var/log/system/system_20250405.log, dmesg.today, and other logs before/during the outage.
No clear errors or kernel panics noted in dmesg or /var/log/system.
Performed dmesg | grep -iE "memory|panic|fault|fail|segfault|ECC" — no suspicious output.
Disk Health:
Ran SMART self-tests on the SSD — all tests passed with no bad sectors or errors.
SSD life left is at 81%, and write cycles are well within tolerance.
Unbound Service:
Investigated Unbound as a possible cause, including verifying its runtime state, config consistency, and control interface. No errors or conflicts were observed.
Suricata and Zenarmor:
Checked Suricata logs — no evidence of high resource usage or crash during the outage.
Zenarmor logs were not found in the default directory; may not be configured for log persistence.
System Performance:
Resource usage (CPU, memory, swap) appears healthy before and after reboot.
No abnormal spikes or saturation were seen in interface statistics via Flow Interface Totals.
Despite all this, the root cause remains elusive. I would greatly appreciate it if anyone with a similar experience or insight into possible culprits (e.g., hardware compatibility, driver issues, kernel bugs, plugin conflicts) could share their thoughts or recommendations.
Thanks in advance!
Rockyuk
leave the grep for now and scan log files to spot what perhaps wasn't in the grep.