@marinsnb said in CGNAT and pfSense: but my WAN gets a 100.xxx.xx.x IP according to my pfSense. Yeah 100.64/10 or 100.64-127.x.x is cgnat range. If that is what your isp is using there not much you can do about that other than contacting to see if they can give you a actual public IP, possible more $$.. Do you also get a IPv6 - that should be a global address and public, I would hope they would do a prefix delegation of /56 or /48 even which would allow for not natting when doing IPv6. Problem with cgnat, is even getting say a hurricane electric IPv6 tunnel isn't going to work.

Yes probably should be the CGNAT space, 100.64.0.0/10.

@cool_corona Thank you -- can you point me to any online discussion/thread where this is discussed so I can follow along?

@bassit said in Any ideas? dns gone rouge.: I hadn't changed anything in resolver or forwarder. Well pfsense was just resolving - that is how it is out of the box. I resolve for like 10 some years now, never an issue. What stopped working is your blocking it via your IPS is most likely.. Or you had an issue resolving "something" and noticed the log entries and went down a rabbit hole that has nothing to do with anything ;) If your just in IDS mode and monitoring and not blocking. DNS (resolving) can fail when you can not talk to a ns in the path to getting to the authoritative NS for the domain your wanting to lookup.. Or can not talk to the authoritative ns for some reason, or dnssec fails, etc. and you have that enabled (its enabled out of the box).. But if yoru going to forward you should uncheck that. When something fails to resolve, and your resolving - you need to figure out why.. dig +trace is good start to see what your having an issue talking too.. If your forwarding, and something fails - your at the mercy of why you didn't get an answer to where you forwarded too.

@johnpoz Done. Thanks for your help!

@st6 said in Connection Dropped After Gateway Down: I asked here and people told me that this feature will only clear states of the link that becomes down don't affecting other links connection. That's not true. It will kill states on all gateways when any gateway goes down. It's not enabled by default because it's very disruptive but can be helpful in some situations. When that feature was added it was not possible to kill states by gateway so it was all or nothing. Steve

@johnpoz wow thanks very much, i am new to pfsense so im not realy sure at what i am seeing, it seemed odd that something pre programmed would contain syntax that is wrong.

@stephenw10 - Your assumption is probably correct, the original install would have been done 5 years ago with whatever version was current at the time. And, I have not connected a monitor to it since. I also re-installed using the UEFI option and that's what it is in now. Still does not explain why it never booted after being shutdown. Guess we will never know, but thanks again for all your input and time. Greg

If it isn't running the ure(4) driver you may need to load the module. That may pass at full speed.

@stephenw10 said in NUT Question: It might be better to prevent the OMV shutting down until it really has to. This sounds like the best solution. Rather then shutdown the OMV (NUT Client) in 15 minutes set it to low battery, like the NUT Server, then both will shutdown together and hopefully both come back on when power is restored.

Looks like some things may have been missing from that post. But same question; what states do you see when you run these tests? @joeschmoe said in TLS pr_end_of_file_error (non proxy related): pr_end_of_file_error message in squid appears Where are you actually seeing that? In the Squid logs? Connections from the pfSense CLI should not go via Squid. If they are being redirected somehow that would show in the states created. Do you still see this issue if you disable Squid? Steve

Thanks a lot!

Mmm, the same port version though? 5.0.1? From what I can see that fix should be in 5.0.1_1 and that hasn't arrived in FreeBSD yet. How are you querying the agent?

@admintkh said in Hugh CPU Load after Upgrade to 2.6.0: https://redmine.pfsense.org/issues/12045 You mean these commits? b5d787d93b3d83f28e87e1f8cc740cb160f8f0ac 0020c845a086766b3315372f006363f8ad76ac54 d97753b5c8f1d32fbcdcbb0d129b49f808245865 3bea7b5b05f200df4cabee12e405b8feade16f0e 89d5cbb82294c8624e66f920d50353057ccab14b That's shouldn't be necessary or even possible in 2.6. Steve

@stephenw10 I have a Cisco switch here and port mirroring with it is a pain. I created a data tap, with a cheap 5 port managed switch.

@stephenw10 said in DNS Resolver: It looks like you have something set that means it's to do a reverse lookup on all IPs. Anything that doesn't resolve is shown in that form which is why you're seeing a whole bunch of 'in-addr.arpa' logs. I think you are on to something here, "all the IP's"... which do seem to be spread across the world, and made me think of pfBlocker... Which in fact seems to be what is behind most of it (could Suricata be the other??). Anyway, I quickly found that two of the IP's which came up 4-5 times in the last 15 minutes, and do not resolve, are on pfBlocker lists, like "pfB_Top_v4"... Yes, if an address has a PTR record then I'd expect it to show a domain there. Steve Interestingly it seems the issue with internal IP's and/or my employers site are no longer there. Not sure what I did but I was going through the settings earlier today and I think had "ignore remote DNS servers" active (under system > general). For sure it's at the default setting now, and I have to go back several hours in the logs to find those IP's in the list as in-addr.arpa... Could that setting change have made this difference? Anyhow, seem the problem is solved, or at least I have a better understanding now. I do still have to look for a simple way to skip those items in the analyzer... The output of Diag > DNS Lookup shows all configured DNS servers not just those in use by the system or by clients. That's expected. So I guess I shouldn't worry then... In fact I can of course remove those servers completely on that page and things still work.

The file is not especially large. Like of the order of 10MB. It really does seem to be the file since in the original use case the rest of the folder can be passed fine without it.

When you add policy routing by setting a gatewau (or gateway group) on the rules you force all traffic to use that route. But here you want traffic between local subnets to use the system routing not go out the WAN. So you need to add a rule above the policy routing rule to pass local traffic only. Create an alias Local_Subnets and put in it all your locally connected subnets. Then add a rule at the top of the list to pass from LANnet to Local_Subnets without a gateway set. See: https://docs.netgate.com/pfsense/en/latest/multiwan/policy-route.html#bypassing-policy-routing Steve

@peter_apiit said in Full Command to restart the networking in pfsense: become slower, so need refresh the connection Not really a thing - but ok ;) Pfsense been running for 72 days, just as fast as it was 72 days ago for "internet"..

@johnpoz fair enough then. thanks!