Don't always blame IPFire (or the Internet)!

Over the past day I have been experiencing a really severe slowdown with my networks. I noticed it about one hour after I updated from CU 174 to CU 175. And it took a day to figure out!

NOTE: normal is 800 Mbps / 20 Mbps

Spoiler Alert (don't read)


My streaming videos would not play properly on two different streaming services.

I first checked my iPhone on the BLUE network.

I checked a GREEN network device.

It was slow also! Definitely the @#&$%&! Internet…

So I logged into the IPFire device and ran the speedtest add-on. It ran near 300 Mbps / 10 Mbps. Much slower than normal (800 Mbps / 20 Mbps). BUT faster then 4 Mbps / 17 Mbps seen in the image above.

So I started debugging the update.

Nothing made sense! Why do I have two reboots? I should only have one!

The first reboot turned out to be a bug I knew all about! Ouch! :crazy_face: I set this one aside. The 2nd reboot was mine at the end of the update process.

I looked at the message logs and the CPU graph, the memory graphs, the media graphs, the firewall logs, etc, etc, etc!

There were a few error/fail/warn type messages in the log. But nothing seemed related to the slow down.

Since nothing made sense I rebooted the firewall, and the Wi-Fi AP and the gateway (cable modem). Everything seemed to boot fine. No errors in the logs. Same result - very slow networks.

after a late night troubleshooting it was time for bed…

The next day I re-scrubbed all of the logs. I found one 3 line entry that didn’t look right. I stopped that service. no joy…

I did a few iperf3 tests and and confirmed “slow”.

I kept going back to the update and I considered going back to CU 174. Or rebuilding everything with CU 175.


After staring at this for many hours (on & off) it finally dawned on me (19 hours later).

When I restarted the network (i.e., firewall, Wi-Fi AP, gateway) I failed to restart my unmanaged switches.

After restarting my unmanaged switches everything started to work perfectly! :smiling_face_with_tear:

My fast network speeds all returned to normal.

Moral to my long story: Don’t always blame the IPFire box. And don’t always blame the Internet. And when you restart (power down, wait, power up) make sure to include hubs and switches!


Nice to hear, I find strange posts in this forum already I deem totally unrelated, as far as I can tell, to IPFire as such but still causing potential issues with IPFire.

So what was the issue with the switches? They just needed the restart due to the IPFire upgrade?

I don’t really know. Since they are unmanaged there is no logging and no web interface to review.

1 Like

I maintain hardware installations of both Testing and Stable. If I re-plug the gigabit switch from one to the other it goes into unknown state, continually flashing the LEDs for all ports, whether or not the connected device is powered up.

I now make a point of powering down the switch, before swapping IPFire boxes. On powering up again, all is fine.

BTW, the reason I have a hard installation of Testing is that the virtual hardware in KVM bears little resemblance to recent, real hardware and won’t always identify kernel module issues.


At least once I’ve have to deal this same issue. Restarting unmanaged switch was only solution for my slowed connection. It might have been switch with somekind of QoS properties but not sure anymore. Can’t even remember what I was doing when this happened, since it happened ages ago.

1 Like