XT8 Asus ZenWiFi cannot detect mesh nodes if in Access Point mode #542
Comments
|
This is the exact same issue I am seeing on my XT8's and I have 5 nodes (primary + 4 secondary mesh). All of my mesh nodes are wired, are yours wired or Wi-Fi backhaul? |
|
Yes my 2 mesh nodes are on wired backhaul. |
|
If you press f12 do you see any error in the developer console of chrome? |
|
@robbbaxley I will have to probably try that when my wife is asleep. Lol! I have always been able to do dirty flash and it works fine. I just did the downgrade only on the main/controller and it worked flawlessly. |
|
@gnuton I didn't try that but I will test again later tonight. |
|
@craigbailey18 , were you able to try the erase NVRAM per @gnuton ? I haven't been able to try it out yet. |
|
I have been busy so didn't get a chance to do it yet. |
|
I'm also having odd issues with the XT8s (mostly in WiFi backhaul). All devices are now running 3004.388.6-gnuton1. It seems that now the 'hidden network' that the mesh sets up to manage itself is disabled. They seem to insist on using the weak user-facing 2.4GHz network for their backhaul channel.
|
|
Just to be clear, before reporting this I did perform the full WPS-button-pushing reset. However, I didn't think to check the various radio settings of each XT8 after re-adding them to the network. My bad. I'm back on the official firmware for these XT8s until I have more time to poke around. Please let me know if there are any specific things I should directly check at that time. |
|
I tested the new @gnuton pre-release, 3004.388.6_0-gnuton1, this morning on my XT8's and the remote/slave nodes updated OK without any obvious issues (wired backhaul). However, the primary/master node still exhibits the same issue where, when the 3004.388.6 firmware is loaded, all remote/slave nodes become inaccessible (invisible) to the primary/master XT8 node. First, I did a dirty upgrade, then for my second attempt, I performed a 'clean' NVRAM upgrade, and there was no difference for me. Also, for the first time, downgrading failed until I performed a NVRAM erase. I was then able to upload the old configuration and perform a downgrade to the 3004.388.5_1 Gnuton firmware without any further issues. So right now, my remote/slave XT8 nodes are running the pre-release, and my primary node is on the previous stable @gnuton 3004.388.5_0-gnuton1. There's definitely something going on with the XT8's and this firmware version. |
|
Yes I finally had a chance to try again this morning and I have the same issue on the primary node. I have the pre-release on the nodes and the official on the primary node. |
|
Me to second node led flashing blue so im back with .5_gnu firmware(xt8) I got this with the latest .6 |
|
@gnuton ....do you have any idea what could be going on here? It appears that the XT8's have an issue with this version of the firmware on the primary/master node. |
|
No idea yet, I will update u as soon as I het what is causing the problem |
|
Ok thanks @gnuton |
|
Is this affecting et8 too? |
|
@robbbaxley I have 4 x XT8's and dirty flashed 3004.388.6_0-gnuton1 on all the units over official FW 3.0.0.4.388_24609. Haven't experienced any issues whilst flashing. I always flash the furthest away node first and work my way up to the the primary router. Is this what you do too? |
|
@Dodgydrains , yes that's my exact procedure as well. One thing I do notice is that all of us are in "AP" mode and have "Wired Backhaul". Is that your setup too? If it is, I will try what you did and dirty flash over the Asus original firmware. I did try a dirty and clean NVRAM flash...both failed for me but I was coming from the previous @gnuton version. |
|
@robbbaxley My primary router is connected to my ISP provided modem. All my XT8's are wireless backhaul. Could that be the difference as you are wired backhaul? |
|
@Dodgydrains , it's certainly possible and I hope it points @gnuton in the directly of whatever has changed between this version and the previous version to cause this issue. |
|
Yeah you guys could be right. |
|
@robbbaxley flashing stock and back makes config just more dirty and unpredictable. |
|
Me i have wifi backhaul(2x xt8) and the problem is the second xt8 (node) not connect and flashing blue and the primary dont see him 😭 |
|
@FlashLight34 , it does appear that this issue/bug is related somehow. I am going to attempt @gnuton's latest request to reset all of the devices and set them up as if they were all brand new but using the latest firmware. |
|
Ran a test yesterday and my xt8 was able to connect to the node via 2,4 GHz, but not via 5ghz. Maybe the issue is there. No idea why this happens |
|
@gnuton @FlashLight34 @craigbailey18 @Dodgydrains @gurple Ok, so I did the last thing @gnuton recommended, which is defaulting the firmware of all devices and resetting up the system from scratch. What I did tonight was:
In the coming days I will move the system back to AP mode from Router mode to test to see if that was the problem...unless someone wants to try that first? :D |
|
@gnuton @FlashLight34 @Dodgydrains @gurple |
|
@craigbailey18 good to hear it worked for you. Always best to do factory reset if you encounter issues to rule this out. |
|
@hulleyrob I am sorry to hear it happened to you as well. I am hoping the new 388_7 addresses the bug. |
|
For my setup as in picture Details
All mesh routers have direct LAN to a main switch then to main router but they still show using WIFI or LAN to second router. |
|
Are you sure you are using gnuton firmware on a ZenWiFi XT8 / RT-AX95Q? The picture shows a ET8 and a non gnuton firmware |
|
@gnuton @robbbaxley The issue remains on the new 3004.388.6_2-gnuton0_beta1 firmware. I guess I have to stick to 3004.388.5_0-gnuton1 until the issue with the AP mode wired/wireless backhaul is figured out. I think there was a change between firmware 3004.388.5_0-gnuton1 and Unstable: 3004.388.6_0-gnuton0_beta1 that caused this bug. |
|
Thanks for the warning I hadn't got round to trying it out yet so I can save the hassle and wait for a new release. |
|
You're welcome @hulleyrob |
gnuton
changed the title
gnuton
changed the title
I have 4x ET8's in router mode which are connected wirelessly. Just dirty flashed 3004.388.6_2-gnuton1 and no obvious issues. |
|
I have 3 ET8's and I seem to be experiencing this issue on 3004.388.6_2-gnuton1. |
I can confirm downgrading to 3004.388.5_0-gnuton1 on my ET8's allowed me to set up AIMesh. |
|
Does it work ok with Ethernet link?
I gave up on WiFi between the 2 ET8s and bought some TP-Link power line
gigabit adapters. Has done wonders to keep upstairs stable.
…On Thu, Apr 11, 2024, 3:26 PM adampk17 ***@***.***> wrote:
I have 3 ET8's and I seem to be experiencing this issue on
3004.388.6_2-gnuton1.
I can confirm downgrading to 3004.388.5_0-gnuton1 on my ET8's allowed me
to set up AIMesh.
—
Reply to this email directly, view it on GitHub
<#542 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABUJKQ6UOLPR2HRBE6TY5OTY44E3BAVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJQGY2TOOBWGQ>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
I am using a wired backhaul currently and it's working as expected. |
|
Awesome, thank you. Does 388.6_2 feel any "better" or more stable than
388.4?
…On Thu, Apr 11, 2024 at 3:52 PM adampk17 ***@***.***> wrote:
Does it work ok with Ethernet link? I gave up on WiFi between the 2 ET8s
and bought some TP-Link power line gigabit adapters. Has done wonders to
keep upstairs stable.
… <#m_8880272170058351076_>
On Thu, Apr 11, 2024, 3:26 PM adampk17 *@*.*> wrote: I have 3 ET8's and I
seem to be experiencing this issue on 3004.388.6_2-gnuton1. I can confirm
downgrading to 3004.388.5_0-gnuton1 on my ET8's allowed me to set up
AIMesh. — Reply to this email directly, view it on GitHub <#542 (comment)
<#542 (comment)>>,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ABUJKQ6UOLPR2HRBE6TY5OTY44E3BAVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJQGY2TOOBWGQ
<https://github.com/notifications/unsubscribe-auth/ABUJKQ6UOLPR2HRBE6TY5OTY44E3BAVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJQGY2TOOBWGQ>
. You are receiving this because you are subscribed to this thread.Message
ID: @.*>
I am using a wired backhaul currently and it's working as expected.
—
Reply to this email directly, view it on GitHub
<#542 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABUJKQ2BNHTXEEFN5CLNTZ3Y44HZ3AVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJQGY4DCNJSG4>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
For my use case 388.4 was more stable than 388.6. I had issues all day today with calls made over Microsoft Teams. Reverted back to stock official Asus FW 3.0.0.4.388_24621 to resolve this problem. I will wait till the next release. Looks like 388.6 is not working well on my XT8 setup. |
|
Bummer, but thanks for the update. I've had to reboot every few days
lately. It feels like there's a memory leak and if everyone's doing a
certain type of streaming And/or game playing it eventually the chokes
itself. But I also have these TP-Link power line adapters which also need
to be re-plugged when they've seen heavy traffic. So yeah. Seems to be a
common issue...? Every time I look to another brand to try to overcome this
need to reboot weekly, The comments from those customers make me happy that
I'm where I'm at. Except I'm not happy with these ET8s either. LOL I mean
problem is probably choosing the wrong hardware for 50 to 60 devices
connected simultaneously... 😞
…On Fri, Apr 12, 2024, 10:54 AM Dodgydrains ***@***.***> wrote:
For my use case 388.4 was more stable than 388.6. I had issues all day
today with calls made over Microsoft Teams. Reverted back to stock official
Asus FW to resolve this problem. I will wait till the next release. Looks
like 388.6 is not working well on my XT8 setup.
—
Reply to this email directly, view it on GitHub
<#542 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ABUJKQ3SY5JXTLU43LJRSETY5ANVJAVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDANJSGIYTMMZTHE>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
Has anyone tried 3004.388.7_1 to see if this issue is resolved? |
|
Not yet resolved. I am not sure it will be anytime soon. You may just have to leave the old firmware on the master node and user the new firmware on the points for now. |
Why do you think it won't be fixed any time soon? |
|
Well the bug has been present for close to 3 months that's all I was saying. I appreciate all the hard work that @gnuton does but this seems like a more complex issue. |
|
AIMesh is a closed component and it come as prebuilt binary from Asus. Tldr I am waiting for a fix from ASUS. |
|
I understand @gnuton. That's why I was recommending to just use the older firmware for now. |
|
INVESTIGATION LOG The reason why there are no mesh nodes dispayed comes from http://192.168.50.12/require/modules/amesh.js an hook in web.c is defined for this, this somehow returns null. if you run the httpd server you will see this errors gnuton@ZenWiFi_XT8-6490:/www# httpd Other values can be fetched from the shared memory. { "get_cfg_client_info", ej_get_cfg_client_info}, This configuration comes from cfg_mnt which is a closed source component asuswrt-merlin.ng/release/src/router/cfg_mnt/prebuild/RT-AX95Q$ ls I have reverted all these files back to previous version. UPDATE1 Tried to run cfg_server via SSH which displays several errrors reference to these files are only part of the cfg service there are no other binaries or code referncing them Somehow the cfg_server in the prebuild doesn't match the one in the new image created. gnuton@ZenWiFi_XT8-6490:/www# md5sum /usr/sbin/cfg_server |
|
Is this expected to be fixed in the latest release? |
|
No this should be still broken since it s based on the same Asus GPL
…On Mon, Jun 10, 2024, 17:08 Rob Hulley ***@***.***> wrote:
Is this expected to be fixed in the latest release?
—
Reply to this email directly, view it on GitHub
<#542 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAEMYRGCFZAVB5JGUKQCWULZGW6PPAVCNFSM6AAAAABDQOEH4OVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNJYGYYTQMJSGY>
.
You are receiving this because you were mentioned.Message ID:
***@***.***>
|
Router Model Affected
[ASUS ZenWiFi AX (XT8)]
Firmware Version Affected
[RT-AX95Q_3004_388.6_0-gnuton0_beta1_puresqubi.w] and [RT-AX95Q_3004_388.6_0-gnuton0_beta2_puresqubi.w]
Is this bug present in upstream Merlin releases too?
No, just both RT-AX95Q_3004_388.6 beta 1 and beta 2
Describe the bug
The Mesh nodes are not detected on the main router on this firmware. I have the RT-AX95Q and the firmware updates but the nodes don't show up. I have to downgrade the main router back to the older stable version for it to detect them again. All of my 3 nodes are the same model.
To Reproduce
Expected behavior
I usually update it node's firmware first and then the main router. I followed the same process here and the nodes/mesh points never get detected. I even reset all the nodes and started from scratch and the issue remained. I reverted the main router/controller to the stable [3004.388.5_0-gnuton1] and the mesh points were detected again.
Screenshots
RT-AX95Q_3004_388.6 Beta 2 Screenshot
RT-AX95Q_3004_388.5 Stable Screenshot
The text was updated successfully, but these errors were encountered: