LHG connections unstable

After trying hAP ax3 + wAP R passthrough, I switched the wAP for a LHG with R11e-LTE modem. Internet speed has almost tripled with the directional antenna, which is a much appreciated improvement.
The connection is, however, unstable to say the least. It works, but with drops happening often and in an irregular frequency.

The LHG is configured after being updated to 7.10.2, with a complete reset, basically following this guide: https://youtu.be/IZFAeLbujso
The modem’s firmware has been updated to the latest MikroTik_CP_2.160.000_v021.
Passthrough works this time, in a vlan net + vlan man setup.

Issues observed:

  • Opening a website takes a long time. It takes quite some time before receiving a first response to start downloading and showing something.
  • Opening a website times out completely with messages like the following, depending on the platform and browser: “Address Not found”, “net::ERR_NAME_NOT_RESOLVED”, “We’re having trouble finding that site”
  • Sometimes only half of the images on a site are loading, the others showing up a lot later, if at all
  • Connection to the LHG through the web UI jumping back to the login screen with the message “ERROR: Internal Server Error”
  • Connection to the LHG through the Android app jumping back to the login screen with the message “Connection timed out”

The website issues happen during the same browser session and in new ones. I.e. visiting the same website works one moment and possibly doesn’t when coming back to it.

Besides the minimal changes to perform as explained in the guide, I tried the following to see if it would improve things, to no avail:

  • Adding static DNS servers
  • Removing eth1 from the LHG’s LAN list instead of having both eth1 and vlan10-man on the list
  • Fixing the passthrough to a specific MAC address
  • Using the same ntp servers as the hAP

The routing entries do not show me anything out of the ordinary.
hAP:

[admin@MikroTik ax3] /ip/route> print detail  
Flags: D - dynamic; X - disabled, I - inactive, A - active; c - connect, s - static, r - rip, b - bgp, o - ospf, d - dhcp, v - vpn, m - modem, y - bgp-mpls-vpn; H - hw-offloaded; + - ecmp 
   DAd   dst-address=0.0.0.0/0 routing-table=main pref-src="" gateway=10.46.173.112 immediate-gw=10.46.173.112%vlan20-net distance=1 scope=30 target-scope=10 vrf-interface=vlan20-net suppress-hw-offload=no 
   DAc   dst-address=10.46.173.96/27 routing-table=main gateway=vlan20-net immediate-gw=vlan20-net distance=0 scope=10 suppress-hw-offload=no local-address=10.46.173.111%vlan20-net 
   DAc   dst-address=192.168.1.0/24 routing-table=main gateway=bridge immediate-gw=bridge distance=0 scope=10 suppress-hw-offload=no local-address=192.168.1.1%bridge

LHG:

[admin@MikroTik LHG] /ip/route> print detail  
Flags: D - dynamic; X - disabled, I - inactive, A - active; c - connect, s - static, r - rip, b - bgp, o - ospf, d - dhcp, v - vpn, m - modem, y - bgp-mpls-vpn; H - hw-offloaded; + - ecmp 
   DAd   dst-address=0.0.0.0/0 routing-table=main pref-src="" gateway=192.168.1.1 immediate-gw=192.168.1.1%vlan10-man distance=1 scope=30 target-scope=10 vrf-interface=vlan10-man suppress-hw-offload=no 
   DAc   dst-address=10.46.173.96/27 routing-table=main gateway=vlan20-net immediate-gw=vlan20-net distance=0 scope=10 suppress-hw-offload=no local-address=10.46.173.112%vlan20-net 
   DAc   dst-address=192.168.1.0/24 routing-table=main gateway=vlan10-man immediate-gw=vlan10-man distance=0 scope=10 suppress-hw-offload=no local-address=192.168.1.20%vlan10-man

The config is as basic as possible at the moment and the LTE connection signal is good.

RSRP -86 dBm (good)
SINR 24 dB (excellent)
RSRQ -8.5 dB (excellent)

The (default) logs on both devices do not show anything related to the drops.
Attached you’ll find the exports of both devices.

What could be the cause of the problem?
Is there anything I can do to have the log tell me more about those “ERROR: Internal Server Error”?
hap.passthrough.export.rsc (7.77 KB)
lhg.passthrough.export.rsc (6.27 KB)

After digging some more, it looks like I found the cause: having vlan10-man on eth1, with both eth1 and the vlan being a bridge port. I will do some more testing, but so far it looks like taking eth1 out of the bridge config fixes the problem.

I need eth1 as a bridge port as well though, with this interface being connected to a switch which has the LHG connected, as well as other devices.
The help suggests to simply set the vlan on the bridge interface. Is this really the right way to go and would it work just like that?
I suppose the alternatives would be to make sure everything connected to the switch is already tagged somehow or having the hAP tag everything untagged through eth1, wouldn’t it?