This may sounds confusing, but you shape downloads by controlling uploaded data.  Each packet you receive is in response to a request you sent.  If you slow down the send requests, the received data will also slow down.  That's how shaping works.  You can't shape traffic that is already hitting your WAN.

Again, thanks for the info :) Yep, my UnRaid is set up to use 802.3ad but I've never actually tested it to see if it's working or not, so probably not if the switch doesn't support it. Like you've said, probably no great loss and I'll happily just use the single link with the new switch.

Problem was probably caused by RDM (Remote Desktop Manager) inserting data in a field that was at least not visible to me. Still not clear how it can generate that message.

No pfsense is not going to be the master browser for the moronic windows network neighborhood. I would suggest determine who your master browser is..  Here is a simple tool to do that with.. https://scottiestech.info/2009/02/14/how-to-determine-the-master-browser-in-a-windows-workgroup/ Why do you care about the stupid browse list anyway??  Do you not know the name of your NAS?  Why do not just access it via is fqdn ie nas.yourdomain.tld ?? But to be clear, unless you installed some samba package on pfsense, it has ZERO to do with participation in maintaining the brwose list.  And it being your gateway, it has ZERO to do with broadcast traffic between devices on the same network.  You do understand that the browselist can take long time to fully populate.  Had you recently turned off machines, say your master browser and now a new election was held, then all the clients have to register themselves with the master browser, etc. etc.. While I do not use that stupid list, I do make sure it works on my network for examples for people that do want to use it ;)  What I would suggest is you turn off the computer browser service on all boxes that you do not want to be the master browser.  You should pick a box that is on 24/7/365 or as close to that as possible to be your master browser..  For example I have my linux box as my master browser running samba. $ nbtstat -A 192.168.9.7 Ethernet: Node IpAddress: [0.0.0.0] Scope Id: [] Host not found. Local: Node IpAddress: [192.168.9.100] Scope Id: [] NetBIOS Remote Machine Name Table Name              Type        Status     –-------------------------------------------     UBUNTU        <00>  UNIQUE      Registered     UBUNTU        <03>  UNIQUE      Registered     UBUNTU        <20>  UNIQUE      Registered     ☻MSBROWSE☻<01>  GROUP      Registered     LOCAL          <00>  GROUP      Registered     LOCAL          <1D>  UNIQUE      Registered     LOCAL          <1E>  GROUP      Registered

Sorry to bump this thread but I have the same symptoms (v2.3.2 crashing everyday, multiple times): https://forum.pfsense.org/index.php?topic=120513

Cache/Proxy forum Anything in squid's log?  Anything in System log when it happens?  By 'crashes', you mean the service stops and must be manually restarted?

Thanks

Dude I did check.. And your forcing the connection out your BL, capture 1.2 For only stuff that is in mis group.  You have not other rules that would allow outbound at all to your wan IP.

Hi. I have already checked the execution order of the 3 pfSense options to launch a command or script at startup. Is the next: 1- /usr/local/etc/rc.d/.sh 2- <earlyshell>3- <shellcmd>4- /usr/local/etc/rc.d/.sh</shellcmd></earlyshell> It seems that sh scripts in /usr/local/etc/rc.d/ run twice, first order and again, after shellcmd and earlyshellcmd I defined in config.xml <earlyshellcmd>echo "I am earlyshell" >> /order.tmp</earlyshellcmd> <shellcmd>echo "I am shellcmd" >> /order.tmp</shellcmd> I created a script with execute permissions in /usr/local/etc/rc.d/order.sh with #!/bin /sh echo "I am /usr/local/etc/rc.d/order.sh" >> /order.tmp; And the dump of /order.tmp cat /order.tmp I am /usr/local/etc/rc.d/order.sh I'm earlyshell I am shellcmd I am /usr/local/etc/rc.d/order.sh Regards

If that old router can run openwrt/dd-wrt you might be able to have it run multiple SSIDs over VLANs. Steve

No it's not getting responses to DHCPREQUESTs or DHCPDISCOVERs so it used what it had cached from the last time it got 192.168.0.28 from somewhere.

Well that makes me feel stupid. It looks like that got it to work by just not selecting a provider. Thank you!

@MicheMuche: For several reasons, those objects can not be wired => Wireless communication Do you power all devices by battery as well? There are countless protocols not bound to ethernet with only one or two wires needed for communications.

Until Vodafone gives you public IP all the configurations make no sense. Most likely you have private IP starting with 10. or 100.

@Derelict: It's in your packet capture. It's coming from somewhere. Check the MAC address tables in your switches, etc. Wireshark out on a mirror port on the physical network. Something. ;D  "Something" ok will do.

Look at your rules on WAN. What is passed there? It is perfectly normal for you to be able to bring up the web gui from the inside using the outside IP address unless you specifically block that. That behavior is governed by the rules on LAN. Being able to connect from the outside is governed by the rules on WAN.

well…... anonymity:  where I live the Government insists on keeping a record of everything done online so using the ISP DNS server will have every request logged  (not that the like NSA dont have a direct link to pretty much everything) And google is not much better. But some kinda separation is comforting as is an anonymous VPN  - So really anything that will make data collection harder an more costly has to be good! Actually not concerned over cashing - just thought DNSBL was an useful addition to blocking stuff- maybe Ive missed something? Thanks for your help!