Messages

I'm the only admin and there is only one device running OPNsense under my control, which rules both of those out unfortunately!

So strange that there are no logs or anything anywhere, mind...

Just going to bite the bullet and rebuild the config changes and be a bit stricter with my backup cadence.

System: Configuration: History would shed some light on when and where these parts were removed (by whom).



Cheers,
Franco

Hey, thanks for the reply.

Unfortunately it doesn't.

Context:
I had initially set up WAN traffic through a single gateway group with double failover (x2 ProtonVPN tunnels x1 Mullvad) and then progressed that to x3 gateway groups each with one failover tunnel for a total of 6 tunnels. I had aliases built out to define traffic flows with PBR fw rules depending on usecase for clean separation of concerns.

History shows absolutely none of this configuration post-single gateway group. No creation of wireguard peers, interfaces, gateways, aliases, Unbound DNS changes etc. etc., let alone their deletion.

Ran firmware update this morning and had 3 VPN tunnel configs, interfaces, gateways, fw policies, associated aliases and gw groups wiped from my system on reboot. Snapshot is gone, last backup is dated 22/03/2026.

Absolutely no idea how/why this would happen, but not too happy about it.

Anyone else had a similar experience and managed a recovery somehow or should I just slug through the setups again? How can I be certain this won't happen again?

EDIT:

Figured this out.

I need a hand setting up shaping to help prioritize media streaming from a Jellyfin and Navidrome server > content downloading over a wireless bridge.
When streaming media on my Nvidia Shield (192.168.3.20) served from my Jellyfin instance (192.168.2.70) while concurrently downloading content that saturates the wireless bridge, I experience playback lag.

Network Topology:

OPNsense is running on an N100 box.

Two subnets on dedicated interface ports:
    - 192.168.2.0/24 [StudyLab]
    - 192.168.3.0/24 [LivingRoom]

OPNsense box is physically located in my actual living room.

LivingRoom subnet consists solely of clients physically connected using an unmanaged switch. On this subnet is my streaming client, an Nvidia Shield (192.168.3.20).

StudyLab has all of my physical server hardware located in my study under a HPE J9774A switch (192.168.2.2), also currently unmanaged.
This is bridged to OPNsense using an ASUS AX56U in MediaBridge mode with an AX58U functioning as the upstream router in AP mode which in turn is plugged in to a dedicated interface port. Wired backhaul is not possible in this setup unfortunately.
Jellyfin and Navidrome are running in a compose stack at 192.168.2.70
Download clients are also at this address.


I tried to give this a go by creating a single pipe to encapsulate all up/down traffic across the bridge instead of dedicated pipes for each as it's the connection bandwidth being saturated that's causing me grief and that's bi-directional.
Unfortunately I've had no luck and the official docs aren't the best at covering this use case (or indeed, a lot of the variables in general).

Some help understanding the correct logic for traffic shaping for my scenario would be amazing.

Attached are some screenshots: