AImesh node requests dhcp, but never receives one...

[Pergola Fabio]

Hi, running an AX88u, 3004.88.1 , one of my ac86u nodes, sometimes goes offline , at that point my syslog is flooded like these below, reconnecting cable doesnt work, rebooting the node doesnt work
only fix is to reboot my router? why is that?

Mar  1 08:49:33 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10 
Mar  1 08:49:35 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10 
Mar  1 08:49:35 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10 
Mar  1 08:49:36 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10 
Mar  1 08:49:38 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10 
Mar  1 08:49:41 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10

Below log from a bit sonner before it goes offline

Mar  1 08:48:09 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) e6:1e:5f:c8:b5:70
Mar  1 08:48:09 dnsmasq-dhcp[2273]: DHCPOFFER(br0) 192.168.0.101 e6:1e:5f:c8:b5:70
Mar  1 08:48:09 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.101 e6:1e:5f:c8:b5:70
Mar  1 08:48:09 dnsmasq-dhcp[2273]: DHCPACK(br0) 192.168.0.101 e6:1e:5f:c8:b5:70 Pixel-7-Fabio
Mar  1 08:49:00 wlceventd: wlceventd_proc_event(685): eth6: Auth 6A:6E:6A:51:0F:21, status: Successful (0), rssi:0
Mar  1 08:49:00 kernel: br0: received packet on eth6 with own address as source address
Mar  1 08:49:00 hostapd: eth6: STA 6a:6e:6a:51:0f:21 IEEE 802.11: associated
Mar  1 08:49:00 wlceventd: wlceventd_proc_event(695): eth6: ReAssoc 6A:6E:6A:51:0F:21, status: Successful (0), rssi:-73
Mar  1 08:49:00 kernel: CFG80211-ERROR) wl_cfg80211_change_station : WLC_SCB_AUTHORIZE sta_flags_mask not set
Mar  1 08:49:00 hostapd: eth6: STA 6a:6e:6a:51:0f:21 RADIUS: starting accounting session 9C808A05B4EA4CFE
Mar  1 08:49:00 hostapd: eth6: STA 6a:6e:6a:51:0f:21 WPA: pairwise key handshake completed (RSN)
Mar  1 08:49:02 wlceventd: wlceventd_proc_event(662): eth6: Disassoc 6A:6E:6A:51:0F:21, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Mar  1 08:49:02 wlceventd: wlceventd_proc_event(662): eth6: Disassoc 6A:6E:6A:51:0F:21, status: 0, reason: Disassociated because sending station is leaving (or has left) BSS (8), rssi:0
Mar  1 08:49:02 hostapd: eth6: STA 6a:6e:6a:51:0f:21 IEEE 802.11: disassociated
Mar  1 08:49:02 hostapd: eth6: STA 6a:6e:6a:51:0f:21 IEEE 802.11: disassociated
Mar  1 08:49:04 wlceventd: wlceventd_proc_event(685): eth7: Auth E6:1E:5F:C8:B5:70, status: Successful (0), rssi:0
Mar  1 08:49:04 kernel: br0: received packet on eth7 with own address as source address
Mar  1 08:49:04 wlceventd: wlceventd_proc_event(695): eth7: ReAssoc E6:1E:5F:C8:B5:70, status: Successful (0), rssi:-74
Mar  1 08:49:04 hostapd: eth7: STA e6:1e:5f:c8:b5:70 IEEE 802.11: associated
Mar  1 08:49:04 kernel: CFG80211-ERROR) wl_cfg80211_change_station : WLC_SCB_AUTHORIZE sta_flags_mask not set
Mar  1 08:49:05 hostapd: eth7: STA e6:1e:5f:c8:b5:70 RADIUS: starting accounting session 1C633ECCC2577503
Mar  1 08:49:05 hostapd: eth7: STA e6:1e:5f:c8:b5:70 WPA: pairwise key handshake completed (RSN)
Mar  1 08:49:06 wlceventd: wlceventd_proc_event(685): eth6: Auth 6A:6E:6A:51:0F:21, status: Successful (0), rssi:0
Mar  1 08:49:06 kernel: br0: received packet on eth6 with own address as source address
Mar  1 08:49:06 wlceventd: wlceventd_proc_event(695): eth6: ReAssoc 6A:6E:6A:51:0F:21, status: Successful (0), rssi:-61
Mar  1 08:49:06 hostapd: eth6: STA 6a:6e:6a:51:0f:21 IEEE 802.11: associated
Mar  1 08:49:06 kernel: CFG80211-ERROR) wl_cfg80211_change_station : WLC_SCB_AUTHORIZE sta_flags_mask not set
Mar  1 08:49:06 hostapd: eth6: STA 6a:6e:6a:51:0f:21 RADIUS: starting accounting session E278D786F361764C
Mar  1 08:49:06 hostapd: eth6: STA 6a:6e:6a:51:0f:21 WPA: pairwise key handshake completed (RSN)
Mar  1 08:49:18 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:18 dnsmasq-dhcp[2273]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:24 rc_service: cfg_server 1312:notify_rc update_sta_binding
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 50:02:91:bc:ba:39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPOFFER(br0) 192.168.0.65 50:02:91:bc:ba:39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 50:02:91:bc:ba:39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPOFFER(br0) 192.168.0.65 50:02:91:bc:ba:39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.65 50:02:91:bc:ba:39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPACK(br0) 192.168.0.65 50:02:91:bc:ba:39 ESP_BCBA39
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:27 dnsmasq-dhcp[2273]: DHCPACK(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:28 wlceventd: wlceventd_proc_event(685): eth6: Auth 98:F4:AB:FB:9A:BA, status: Successful (0), rssi:0
Mar  1 08:49:28 wlceventd: wlceventd_proc_event(722): eth6: Assoc 98:F4:AB:FB:9A:BA, status: Successful (0), rssi:-65
Mar  1 08:49:28 hostapd: eth6: STA 98:f4:ab:fb:9a:ba IEEE 802.11: associated
Mar  1 08:49:28 kernel: CFG80211-ERROR) wl_cfg80211_change_station : WLC_SCB_AUTHORIZE sta_flags_mask not set
Mar  1 08:49:28 hostapd: eth6: STA 98:f4:ab:fb:9a:ba RADIUS: starting accounting session 86735683AA6163AA
Mar  1 08:49:28 hostapd: eth6: STA 98:f4:ab:fb:9a:ba WPA: pairwise key handshake completed (RSN)
Mar  1 08:49:28 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 98:f4:ab:fb:9a:ba
Mar  1 08:49:33 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:35 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:35 dnsmasq-dhcp[2273]: DHCPREQUEST(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:36 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:38 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10
Mar  1 08:49:41 dnsmasq-dhcp[2273]: DHCPDISCOVER(br0) 192.168.0.2 24:4b:fe:be:7c:10

 

[Pergola Fabio]

tried doing below, but doesnt work, i really need to reboot the router

killall dnsmasq
rm /var/lib/misc/dnsmasq.leases
service restart_dnsmasq

 

[bbunge]

Upgrade your router firmware. Make sure your nodes are on the most recent Asus firmware. May be a good idea to factory reset everything and start over!

 

[Pergola Fabio]

All running latest, main router is Merlin. Nodes are stock

 

[bbunge]

All running latest, main router is Merlin. Nodes are stock

But you stated 3004.88.1 for the router. Most recent Merlin for the AX88U is 3004.388.11

Could be that the AC86U is failing. Remove it from the mesh, Hard factory reset it then try adding it back to the mesh. If it works fine. If not, recycle it.

 

[Pergola Fabio]

That was a type error
Hmm, don't think it's an hardware issue

 

[Tech9]

one of my ac86u nodes

You have no signature (we don't know your hardware), you provide incomplete information (there is no AX88u model and 3004.88.1 firmware, perhaps RT-AX88U and 3004.388.11 firmware), your main router is running 3rd party firmware in unknown configuration (guest networks, scripts, anything else?), we can only guess your AiMesh is wired (reconnecting a cable, which one?), your unknown number RT-AC86U are known to fail (radios, ports, voltage regulators, weird intermittent things). Based on the limited information provided all I can tell is - something is wrong. Reset everything, run stock Asuswrt on all units, test again.

 

[Pergola Fabio]

sorry, more info, all nodes are indeed wired, all running latest firmware:
nodes are connected to the ports 2-4 from the ax88u

i have configured in the AX , that the the nodes get a dhcp with mac reservation, so they get 192.168.0.2 - 192.168.0.4
i want to have them as a static ip, thats why i make dhcp reservations for the nodes ...
maybe i can give the a static IP instead for the nodes? but i cant find where i can do that? i cant login in those aimesh, if i browse to 192.168.0.2 for example, it redirects me to the main router

RT-AC86U
24:4B:FE:BE:7C:10
192.168.0.2 Default

RT-AC86U
0C:9D:92:46:4E:A0
192.168.0.3 Default

RT-AC86U
B0:6E:BF:63:55:08
192.168.0.4 Default

Current Version : 3004.388.11 Manual firmware update : Upload

Current Version : 3.0.0.4.386_52294-ga575528 Manual firmware update : Upload

Current Version : 3.0.0.4.386_52294-ga575528 Manual firmware update : Upload

Current Version : 3.0.0.4.386_52294-ga575528 Manual firmware update : Upload

[td]AiMesh router[/td]

[th width="246.438px"]

RT-AX88U​

[/th]​

[td]AiMesh Node[/td]

[th width="246.438px"]

RT-AC86U ( B0:6E:BF:63:55:08 )
Location : Yard​

[/th]​

[th width="246.438px"]

RT-AC86U ( 0C:9D:92:46:4E:A0 )
Location : Attic​

[/th]​

[th width="246.438px"]

RT-AC86U ( 24:4B:FE:BE:7C:10 )
Location : Hallway​

[/th]​

​

 

[Tech9]

i want to have them as a static ip, thats why i make dhcp reservations for the nodes ...

You don't do this with AiMesh. It's a closed source software with own requirements that may change on firmware update. Remove all your optimizations around it and try again.

 

[Pergola Fabio]

what optimizations? i dont have any, its a quite a basic setup

 

[jksmurf]

Pergola Fabio said:
i want to have them as a static ip, thats why i make dhcp reservations for the nodes ...

Actually I do exactly this (node IP reservation) too, like the OP. Makes managing access to the nodes so much easier, especially running scripts, putty (SSH), WinSCP access and the like.
Keeps it neat and I have been doing that for several years now, on two systems.

I do not personally believe the random subnet IP (last octet) that you get upon adding a Mesh node is critical to the functioning of AiMesh, but if someone knows that it is for a fact and what the ill effects are, I'm happy to be proven wrong.

I have no comment on other optimizations, out of my sphere of knowledge.

 

[Tech9]

what optimizations?

Like reserving IP addresses for black box AiMesh. You are having DHCP issues with it now, no? Reverse the changes you've made in DHCP server, reboot the system and see if the issue is resolved. If still not resolved remove the node from AiMesh and re-add it. If still not resolved this particular device may have issues.

Actually I do exactly this

You help @Pergola Fabio to resolve the DHCP issue then...

 

[Pergola Fabio]

my guess its still an issue on main router
reconnecting cables doesnt work, i dont even see in syslog a link down/up ... rebooting the node doesnt work
rebooting main router is the only solution

 

[Tech9]

Reserving IP addresses for AiMesh devices is not needed nor recommended. If you insist on doing it for whatever reason - continue rebooting the main router as the only solution.

 

[Pergola Fabio]

Why is this not recommended? Anyway making a Mac reservation or not, should not be an issue

 

[Tech9]

You've made changes to DHCP server on your router and have issues with DHCP on your node managed by closed source constant work-in-progress state AiMesh. The first troubleshooting step is reversing the changes you've made on a blind without knowing what AiMesh needs or expects to have. If you don't want to do it - figure it out. I can't help you any further.

 

[bbunge]

I agree with @Tech9. Using manual reservations (note that this is not the same as static IP addresses) with mesh nodes is not a good idea. Manual reservations are not a great feature and can cause problems. If a client needs to have the same IP address give it a static IP address! My cams, NVR, NAS, Pi-Hole, managed switch and printer all have static IP address. They have been like this across three new routers and the only setting I need to change is the DHCP pool starting address.
And yes, the AC86U could have hardware issues. Many of us have experienced this with the AC86U.

 

[Pergola Fabio]

hmmm, all my important devices are below 100 , dhcp starts from 100, there are 0% issues for this, nodes can have perfect dhcp reservations, there is no difference then configuring another device as a mac reservation, the problem is related to something else

 

[jksmurf]

@Pergola Fabio

Let's take a step back, stop firing a stream of consciousness at it, and try and break this down a wee bit:

• It seems that your mesh works OK on 2 out of 3 nodes.
• I seems the problematic node is the Hallway node, the one you assigned the IP address 192.168.0.2 to MAC 24:4b:fe:be:7c:10. Let's assume for now that manual address assignment is fine, because your network actually works on 2 of 3 nodes, using that approach (which I also use as noted above).
• So how about you physically disconnect the two good nodes, Yard and Attic. You don't have to remove them from your mesh setup on MAIN, the AiMesh WebGUI will just show them as "offline" for now.
• I then strongly suggest you delete that problematic node from the MAIN router's AiMesh WebGUI and then HARD reset the problem node.
• To do this, turn off the power on the Node using the switch (but leave the connector plugged in), hold the WPS button down and while holding (do NOT let it go), turn it on, wait for the power light (LED) to come on, wait patiently (even it it blinks a few times) for the power LED to go OFF, then release the WPS button, then switch the unit off; wait a second or two, then switch it on again. Then cable up and RE-ADD that node as normal.
• Your static IP assignment should give you the same 192.168.0.2 address. DOUBLE CHECK however that you have the correct MAC address for that Node; because if you used the dropdown in the LAN DHCP page, for when it was connected (unknown to you) via 5Ghz or 2.4Ghz it MAY not be the correct MAC for the Ethernet Backhaul.
• You did not say what your settings were for the backhaul. There are effectively "four"settings, from 'tightest' too 'loosest':
• (1) The "Ethernet Backhaul Mode" AiMesh, System Settings) is NOT not recommended by most as if you do have any failures, that's it, it will not drop back to wireless backhaul.
• Then, from the menu AiMesh, Node, Management, Backhaul Connection Priority, there are options (4) Auto, (3) 5Ghz Wifi First, (2) 1G WAN first (sometimes 2.5G as well) in number order.
• With a wired LAN I would suggest you choose 1G WAN first (as AiMesh nodes all get the Etherent cable plugged in to the WAN port). If it drops, it will drop back to AUTO, then may recover to 1G later. Auto, as I understand it, SHOULD choose 1G WAN if it is a solid connection, but might choose 5Ghz and not move, or worse, flop between them.
• In any case I do not think this will help (much), as it sounds to me to be an issue with the physical connection; or the NODE itself, by elimination.
• So that means MAIN Ethernet Port, NODE Ethernet Port, Ethernet CABLE. So try a GOOD cable and do NOT plug it in the same (1/2/3/4) Ethernet Port it WAS in (I am guessing you used Port 2 to match... 0.2, 0.3 and 0.4).
• Select a known "Good" port from another Node (that is unplugged as above). Check the WAN port of the NODE (it is the ONLY place you can plug in to) for dirt, dust, grease. Use isopropyl alcohol to clean it and the cable pins with a Q-tip.
• You did not say if you have any switches or worse cable connectors. If yes, take these out of the mix completely. Go direct; this is a must.
• If your wired backhaul has MAIN Router Ethernet Port X→Ethernet Wall Socket→Ethernet Wall Socket→Hallway Node WAN, EITHER (a) just use a DIRECT cable MAIN→Node to test OR (b) Put the Hallway unit at the location of one of the other two working Nodes i.e. use their (know good) Wall sockets.
• Run it like that for a while, see if it stays connected.
• If it still drops, it may simple be a faulty unit OR the power supply on the unit is going. So you COULD then try using the PS from one of your known GOOD Nodes, on this wonky unit and test that. That would be a last resort.

This is what I would do to troubleshoot. At least you will have ruled out a few things.
When you have done all that, come back and report your findings. HTH.

k.

 

[Pergola Fabio]

ok, will try it, it can take some time , sometimes its running good for weeks..., then i try indeed another cable
yes, nodes are directly connected to port 2 3 and 4 , and on nodes, they use the WAN port
In aimesh setup its configured to "1Gwan Only" , seems grayed out, cant chage it