Unbound fails to restart after DNSBL feed update
-
Have and issue that seems to be plaguing us with some frustrating symptoms.
Our Netgate SG-5100 is running pfSense 21.05, with several separate networks, and OpenVPN.
Issues
-
DNSBL does not display the number of packets dropped, Gives the 'SQLite database missing, Force reload DNSBL to recover' error which does not clear no matter how many times its reloaded or restarted
-
DNSBL fails to update occasionally and causes the Unbound service to fail. Here is the error from the pfblogerng.log
Clearing all DNSBL Feeds Stopping Unbound Resolver.............................. Additional mounts (DNSBL python): No changes required. Starting Unbound Resolver. DNSBL enabled FAIL *** Fix error(s) and a Force Reload required! *** ==================== [1623373312] unbound[60921:0] error: bind: address already in use [1623373312] unbound[60921:0] fatal error: could not open ports ==================== Stopping Unbound Resolver..............................- There have also been instances where the DNSBL service seems to stop functioning, Typically when the device has reached +70% storage capacity (well before storage is truly exhausted). Unbound still resolves for devices on OpenVPN because DNSBL is not enabled on that interface, but all other interfaces with DNSBL will fail to resolve external IP.
Unbound fails after the device has been running for an extended period of time, generally about a month, but we recently had DNS failures after 3 days. Restarting the Unbound or DNSBL services does not solve the issue, rebooting the firewall is the only thing that solves the issue, and it is successful every time with no other changes. To elaborate, I can resolve the DNS issue without clearing large log files or taking any other troubleshooting steps. I am baffled by these symptoms, its very hard to troubleshoot since it occurs only about once a month and I have to get it back online ASAP.
Any help would be appreciated, let me know what kind of additional information I can provide.
-
-
@kurlee said in Unbound fails to restart after DNSBL feed update:
Gives the 'SQLite database missing, Force reload DNSBL to recover' error which does not clear no matter how many times its reloaded or restarted
You have to clear the error yourself.
That is : clean out the 'log file' that triggers this message / warning :
( I guess it's the error.log file that contains the error message )
edit : see also https://forum.netgate.com/topic/164305/py_error-log-errors-maxmindb-and-_sqlite3-modules-not-found as it could be related.
