Multi WAN, reloading filter issue

[treg]

Hi,

I implemented an OpnSense multi WAN setup following the official guide + https://www.thomas-krenn.com/pl/wiki/Multi_WAN_w_OPNsense . I also configured some firewall rules to policy route certain traffic directly on the second interface.

The primary interface is a PPPoE connection which receives it's public IPv4 from an upstream DSL system. (tier1)
The second interface is an ethernet interface which receives it's IP from another router upstream. (tier2)
OpnSense version: 21.10.3_1-amd64

Once every X hours (sometimes 6h, sometimes 40h), the following happens in the backend log:

Code: [Select]

2022-06-22T07:06:19 configd.py[42803] message 4b8f2e22-4c6f-4313-8898-057592af1416 [filter.refresh_aliases] returned {"status": "ok"}
2022-06-22T07:06:19 configd.py[42803] [d036238f-80fc-41d5-ba2f-626dc710bd99] updating dyndns WAN_PPPOE
2022-06-22T07:06:19 configd.py[42803] [4b8f2e22-4c6f-4313-8898-057592af1416] refresh url table aliases
2022-06-22T07:06:19 configd.py[42803] OPNsense/Filter generated //usr/local/etc/filter_geoip.conf
2022-06-22T07:06:19 configd.py[42803] OPNsense/Filter generated //usr/local/etc/filter_tables.conf
2022-06-22T07:06:19 configd.py[42803] generate template container OPNsense/Filter
2022-06-22T07:06:19 configd.py[42803] [635fb816-930b-4bd2-ba5f-eb458013aa0b] generate template OPNsense/Filter
2022-06-22T07:06:18 configd.py[42803] [cc320e1e-94f5-4724-80e8-32101a777efe] request pf current overall table record count and table-entries limit
2022-06-22T07:06:18 configd.py[42803] [20c7c750-e08b-4869-8f9e-b7d0877d3997] Reloading filter
2022-06-22T07:05:42 configd.py[42803] message 54e803db-2416-4e0e-a316-7d0f7cf4525d [filter.refresh_aliases] returned {"status": "ok"}
2022-06-22T07:05:41 configd.py[42803] [171d1b18-e1f2-48fa-8550-b9dc9dad695f] updating dyndns WAN_PPPOE
2022-06-22T07:05:41 configd.py[42803] [54e803db-2416-4e0e-a316-7d0f7cf4525d] refresh url table aliases
2022-06-22T07:05:41 configd.py[42803] OPNsense/Filter generated //usr/local/etc/filter_geoip.conf
2022-06-22T07:05:41 configd.py[42803] OPNsense/Filter generated //usr/local/etc/filter_tables.conf
2022-06-22T07:05:41 configd.py[42803] generate template container OPNsense/Filter
2022-06-22T07:05:41 configd.py[42803] [33f5b1e4-f33c-48e3-b5bb-b10714dc5ad3] generate template OPNsense/Filter
2022-06-22T07:05:40 configd.py[42803] [e348bdba-26f7-4a90-baa1-4f3a44a6fdc7] request pf current overall table record count and table-entries limit
2022-06-22T07:05:40 configd.py[42803] [a555886c-b0d5-4176-8e5d-7078eeb57652] Reloading filter

After this moment, outgoing connections will not flow anymore thru the PPPoE interface, while the interface is still up, no PPPoE re-negotiations happened.

Why is this "reloading filter" happening? This was not requested manually. Based on the logs I cannot correlate it to other events.

Does this behavior ring a bell with anyone? I crawled the whole forum and Google, but couldn't find a good match.

Thx, T.