Topics

Hi all,

long time no see. I just came back as I'm still thinking of switching back to my custom device with OPNsense. In the past I had issues with IPv6 not updating the prefixes after WAN connection lost (due to ASSIA profile enhancement or nightly reconnect). I've spent a lot of time of testing during a few 23.x releases but at the end I sadly had no more time for further testing and stopped testing back in June 2023.
So how is the current situation? Do you experience still problems? I'm running a dual stack connection from Deutsche Telekom.

Thanks in advance.  :)

Hi there,

last days I have spent many hours setting up an e-mail alert with monit for getting e-mails after the WAN has reconnected (or changed it's IP addresses).
Test notifications are working. I have cloned the default "gateway_alert" and changed the test to a created changedstatus one, but mails are not getting out. The only time I'm getting e-mails is during reboot if the monit instance changes.

How is it possible? Any ideas? Thanks in advance.

Cyberturtle

Hi there,

a few weeks ago I've recognized that die temperature widget seems to show higher temperatures than reading via CLI. I just have had a look into the sources and it seems that the reading of the widget uses a different way, so I started comparing the results.

Using

Code Select Expand

sysctl -a dev.cpu | grep temperature
returns lower temperatures

Code Select Expand

dev.cpu.3.temperature: 41.0C
dev.cpu.2.temperature: 41.0C
dev.cpu.1.temperature: 41.0C
dev.cpu.0.temperature: 41.0C


than

Code Select Expand

sysctl -a | grep temperature
which returns

Code Select Expand

hw.acpi.thermal.tz0.temperature: 40.1C
dev.cpu.3.temperature: 45.0C
dev.cpu.2.temperature: 45.0C
dev.cpu.1.temperature: 45.0C
dev.cpu.0.temperature: 44.0C


The second variant takes a bit longer and creates higher load on the CPU. My Firewall device uses a Intel J4125. What do you think about the results?

Best regards
Cyberturtle

Hi there,

currently I have a problem. I recognized something yesterday when I was starting a huge download on my Xbox Series X. Suddenly my wife's and my laptop lost their VPN connections to our work and we haven't been able to work with our RDP connections reliably. Every now and then our connections timed out.

I already have setup FQ-CoDel like recommended by many howtos in the WWW. Bufferbloat tests show A or A+. After the connection drops occured I set up additional prio queues for TCP ack, DNS and of course the network address of our work. After that I reduced the bandwidth in the pipes as well. The shaper status shows that the rules are being used.

I can reproduce this issue by starting a huge download on any client in our home network. The Ubuntu Iso download is a good test for reproduction. As soon as I start it and the bandwidth gets saturated, the issue occurs. I need many tries to establish a VPN connection on my client. It seems as some packets are getting lost or so.

My hardware is:
VENEON P09B
Intel I-225V B3
Intel J4125 CPU
Crucial 16 GB RAM DDR4
Kingston SSD 256 GB

My provider is Deutsche Telekom with dual stack. PPPoE is being handled by OPNsense 23.1.7_3. I'm using a Speedport Smart 4 as modem. If I'm using a Fritzbox 4060 as test device, everythin is working fine. My line is synced with 179 Mbit/s down and 46,9 Mbit/s up. The pipes are configured with 160 Mbit/s down and 43 Mbit/s up. I also tried toggling ECN.

By the way a huge download don't affect watching Disney+ or Netflix. It seems to affect only our working VPN.

Sadly, I don't know what else I could try.

I'm so sad as I like OPNsense and all the stuff really much.

Thanks in advance,
Cyberturtle