Messages

I added a cron job to restart Unbound DNS every night, and for the past month I've only had one instance where DNS resolution stopped working for LAN addresses. 

Same rate of errors, e.g. once-a-month-ish, but I'd say the past month has been the "quietest" for me in regards to Opnsense issues so far.

Happened again yesterday.

Only a restart of Unbound DNS would resolve the issue.

Normally, everything is working great (25.7.4), but once in a while I'll notice that certain programs on the LAN stop working, and their ability to ping other LAN machines by name is broken (even when that name is their own).  When that occurs, pinging by IP address still works fine, and programs using the Internet still work fine.

Interestingly, on one machine that runs a VPN, I can ping other machines by name but it uses the 10.x subnet through the VPN (which is set to allow LAN traffic) instead of the 192.x subnet that all machines run on normally.

If I restart Unbound DNS then everything gets back to normal.  It had been up and running for 22 days yesterday when I had the problem, and I'd say this happens about once a month-ish.

Should I just automate a process to reboot Unbound DNS every night?

Widget is still OK for me.

Good advice.  Thank you for the link.

Thank you for the reply.

I ran the scrub and all of the error counts were zero.

I opted not to set the tunable, and I'll live with the error for now.

No panics in the log.

Older entries contained a typical complaint about the inability to increase the buffer size, but lately there are only info messages about GOMAXPROCS and GOMEMLIMIT, and warn messages about "No files matching import glob pattern".

RAM is soldered on, so that's not going to be possible.  Maybe I'll look at the SSD at some point, but if it reboots by button and not by system "restart" command that doesn't feel like a hardware issue to me?

How is your Maltrail configured now, e.g. are you using firewall rules to block packets on WAN and LAN, IN and/or OUT, and on which interfaces?

Thank you for responding.  Memory utilization runs in the low 30% range, and ZFS has been installed from day 1.

Good catch, I am running 25.7.3_7 now (and for about a week), but I was running 1_1 back when I started this thread on 8/22.

Zpool scrub reports 0 errors and 0 blocks repaired.

Awesome, thank you so much!  I'll look into that and report back.

One of the three reverse proxies stopped working yesterday, so I need to look into what is going on with those, I thought I had it fixed permanently.

Any suggestions on where to look or troubleshooting tips to try when Caddy stops responding?

Yesterday, one of my reverse proxy services stopped working.  As part of the troubleshooting I restarted Caddy and Unbound DNS, but that did not help, so I then used the GUI to reboot the firewall, and it did not come back up, again.

Since the problems with reboots mentioned earlier in this thread I have installed a KVM with full-time EDID, thinking the box might not like being headless.

The firewall had the power button illuminated, but nothing would come up on the monitor.  I pressed the button quickly and the light extinguished (usually you have to hold it for a while to get the box to shutdown / power cycle, I thought).  Then I pressed it again and the box booted normally.

Not sure what else to do in troubleshooting this issue, but the box is set to boot on power failure, and it's not been a problem prior to 25.7 series.

For not being a ZFS developer that's a pretty good find, and I agree with you, that code calls the same function, but throws the b_slow flag/error.

I just wish it didn't stop logging, as I'd like to comb through my boot logs to see if everything is OK, but the log file stops right at that error message.

Thank you for the strong response!

After updating to 25.7.3 I have noticed that the legend for the "Interface Statistics" graph on the dashboard is missing.

It's a simple "Lan" / "Wan" legend, but it's supposed to be there for a reason, I suppose.