What's new

384.17 loosing wan connection

  • SNBForums Code of Conduct

    SNBForums is a community for everyone, no matter what their level of experience.

    Please be tolerant and patient of others, especially newcomers. We are all here to share and learn!

    The rules are simple: Be patient, be nice, be helpful or be gone!

nihon

Regular Contributor
I encounter stability issues with latest 384.17 FW.
Wan connection goes down and does not recover unless I restart the router.
LAN still works. It is not an ISP issue (connecting a laptop instead of the router works).

Main router is AC-2900 (AC-86U FW version 384.17 Merlin) and a aimesh node based on a AC-68U (384.17 Merlin).

Only addon enabled is YazFi with one YazFi guest network enabled.

I attached the syslog. I really hope someone can help to shine some light upon what is going on.
I have been a Merlin user for many years and with the AC68U as main router I never had a single stability issue.
But since I changed to AC86U for the main router it has not been stable a single day.
 

Attachments

  • syslog.txt
    492.8 KB · Views: 161
Pixelserv shows up in your log. Is Diversion enabled? There is so much noise about WiFi auto channel changing and the device 98: D8:63:01:8B:54 flapping up and down on the guest network.

Your syslog is also missing a lot of expected entries. Nothing I could see suggests any WAN connection going down, so you should provide more specifics about when the WAN appears down because it’s not obvious from that log file.
 
Thanks Dave for going trough the log,
I have installed Diversion indeed, but disabled it. Pixelserv might still be running though I suppose.
Yes I only had Error log level logged. I will turn on most verbose log level and upload it again when it has hang next time then.

When I woke up this morning the WAN was down. I then saved the log first thing. So exactly when it happend during the night I cannot tell.
Yesterday evening I streamed a movie file to a chromecast. That worked fine (but I have had problems now and then that chromecasts and nest devices lost wifi connection, the chromecast I used yesterday is connected through cable though due to not been stable). So I am afriad I cannot give an exact time when it happened. I hoped the log would tell something but since it does not I suppose I have to wait until next hang.
 
Not sure if this is linked, the only WAN instability that I have experienced since I use Merlin is with 384.13_8 on my RT-AC87U.
I noticed that when I start OpenVPN Server, the WAN becomes crazy (CPU usage as well) until the moment I run a health scanner on my external USB drive plugged in the rear of the router. When the disk check is done, WAN & CPU comes back to normality ... :eek:
 
I don't have an answer yet but I have a very similar issue with the wan connection dropping - but it is very intermittent. It happens once after 5-10 days of normal operation. I haven't seen anything in the logs to help.

The only clue I have is that when this issue happens, the wan connections fails to update its DHCP lease from my internet modem (NBN connection box in Oz).

I have tried another router (apple AirPort Extreme) and I did not have any issues.

I will update this thread if I find anything more in the next week or so.
 
I don't have an answer yet but I have a very similar issue with the wan connection dropping - but it is very intermittent. It happens once after 5-10 days of normal operation. I haven't seen anything in the logs to help.

The only clue I have is that when this issue happens, the wan connections fails to update its DHCP lease from my internet modem (NBN connection box in Oz).

I have tried another router (apple AirPort Extreme) and I did not have any issues.

I will update this thread if I find anything more in the next week or so.
Look for ISP dhcp failed or something along those lines, then look up the contious mode thread, there is instructions there that might help, if all else fails pm jack Cheng on whirlpool about it, I'm a fellow Aussie with dhcp from my ISP Optus FTTN.
 
I've got exactly the same setup, 86U router and 68U mesh node both running Merlin 384.17, and conversely the entire setup has been solid since I moved to it last weekend.

Prior to that, I just used the 86U with 384.12 and the 68U was collecting dust on a shelf the last few years and was still on 380.69 before I updated it to use as a mesh node.
 
Not sure if this is linked, the only WAN instability that I have experienced since I use Merlin is with 384.13_8 on my RT-AC87U.
I noticed that when I start OpenVPN Server, the WAN becomes crazy (CPU usage as well) until the moment I run a health scanner on my external USB drive plugged in the rear of the router. When the disk check is done, WAN & CPU comes back to normality ... :eek:

Yes I have heard about USB drives causing issues, especially if health check is enabled. At the moment I only have a USB drive connected to store the addons like diversion and Yazfi on. And the automatic health check is disabled for the USB drive. VPN server is not in use.
 
Okay I got some sort of hang again now. But this time the router stopped being accessible (web gui did not respond) and WAN / LAN still worked.
Fortunatelly I now had switched on all log levels. I upload the log here....
Some strange stuff seems to be in the log with module failures, some YazFi messages, my doorbell seems to also to cause some messages and the date jumps to say May 5th which seems strange. Also it seems it has detached and scanned a USB drive even though I have disabled automatic health checks in the WebUI for the USB drive. Also there are a lot of deauth and other error messages. Diversion seems to have become enabled as well even though I disabled it (I did disable it even before I updated to the latest FW version). I verified it and diversion is in fact running.

....but I cannot really read out exactly what is going wrong causing the problem from the log. Hopefully someone more familar with reading these logs and the merlin fw can help out?
 

Attachments

  • syslog.txt
    469.4 KB · Views: 136
Last edited:
Okay I got some sort of hang again now. But this time the router stopped being accessible (web gui did not respond) and WAN / LAN still worked.
Fortunatelly I now had switched on all log levels. I upload the log here....
Some strange stuff seems to be in the log with module failures, some YazFi messages, my doorbell seems to also to cause some messages and the date jumps to say May 5th which seems strange. Also it seems it has detached and scanned a USB drive even though I have disabled automatic health checks in the WebUI for the USB drive. Also there are a lot of deauth and other error messages. Diversion seems to have become enabled as well even though I disabled it (I did disable it even before I updated to the latest FW version). I verified it and diversion is in fact running.

....but I cannot really read out exactly what is going wrong causing the problem from the log. Hopefully someone more familiar with reading these logs and the merlin fw can help out?

I have an RT-AC5300 with Merlin .17 I was having issues with one of the two 5 Ghz dropping and also losing internet or very low speeds. Both issues were not at all Merlin related. The 5 Gig issue is known across multiple routers and can be fixed by turning off auto channel select My intermittent internet issues were caused by a bad attenuator attached to my cable modem. It took me too long to figure this out including purchasing a new RT-AC5300 and painfully refiguring all my clients.... when all along all I had to do was remove the attenuator and choose a channel for 5 gig (I returned the 2nd RT-AC5300)
 
Okay I got some sort of hang again now. But this time the router stopped being accessible (web gui did not respond) and WAN / LAN still worked.
Fortunatelly I now had switched on all log levels. I upload the log here....
Some strange stuff seems to be in the log with module failures, some YazFi messages, my doorbell seems to also to cause some messages and the date jumps to say May 5th which seems strange. Also it seems it has detached and scanned a USB drive even though I have disabled automatic health checks in the WebUI for the USB drive. Also there are a lot of deauth and other error messages. Diversion seems to have become enabled as well even though I disabled it (I did disable it even before I updated to the latest FW version). I verified it and diversion is in fact running.

....but I cannot really read out exactly what is going wrong causing the problem from the log. Hopefully someone more familar with reading these logs and the merlin fw can help out?
There's still something "filtered" about your log. For example, I see log messages from YazFi saying the firewall was restarted, but I don't see any rc events with restart_firewall. Or your router reboots (and the time/date resets to May 5) but there are no useful log messages about WAN or ISP DHCP, etc.

Is this what your Syslog page looks like? This is what you need to capture useful information.
upload_2020-5-31_20-14-16.png
 
There's still something "filtered" about your log. For example, I see log messages from YazFi saying the firewall was restarted, but I don't see any rc events with restart_firewall. Or your router reboots (and the time/date resets to May 5) but there are no useful log messages about WAN or ISP DHCP, etc.

Is this what your Syslog page looks like? This is what you need to capture useful information.
View attachment 23851

Hm, indeed it has switched to info level only now when I checked. I enabled all logging before.
I have changed it again now to default (message log level) and all (Log only messages more urgent than).
Hopefully it keeps the settings this time.
 
I have an RT-AC5300 with Merlin .17 I was having issues with one of the two 5 Ghz dropping and also losing internet or very low speeds. Both issues were not at all Merlin related. The 5 Gig issue is known across multiple routers and can be fixed by turning off auto channel select My intermittent internet issues were caused by a bad attenuator attached to my cable modem. It took me too long to figure this out including purchasing a new RT-AC5300 and painfully refiguring all my clients.... when all along all I had to do was remove the attenuator and choose a channel for 5 gig (I returned the 2nd RT-AC5300)

Hm, I cannot switch off the auto channel select. I assume it is "Control Channel" under Wireless settings?
It can only be set to auto?
If one is expected to be able to change that to something else than auto, maybe it is forced to auto by YazFi?
 
Hm, I cannot switch off the auto channel select. I assume it is "Control Channel" under Wireless settings?
It can only be set to auto?
If one is expected to be able to change that to something else than auto, maybe it is forced to auto by YazFi?

Do you have ‘Smart Connect’ enabled?
From memory, I think that sets channels to ‘auto’.
(Might be wrong, it’s been a while since I messed with smart connect):confused:
 
Yes I have smart connect (2.4 and 5Ghz sharing SSID) enabled.

Skickat från min SM-G973F via Tapatalk
 
Look for ISP dhcp failed or something along those lines, then look up the contious mode thread, there is instructions there that might help, if all else fails pm jack Cheng on whirlpool about it, I'm a fellow Aussie with dhcp from my ISP Optus FTTN.
Just wanted to thank you for this info as this is the exact issue that I have been experiencing.

I haven't changed to continuous mode (yet) as I spoke to Optus as I could see my WAN IP was swapping between 2 addresses when using my Asus router and I thought this was the causing the issue. Optus have set my account to a fixed ip over dhcp and I am happy about this as I prefer this config.

Since changing to a fixed IP I haven't had the WAN DHCP issue for 5 days. I'm going to leave everything the same for another week to make sure the issue is resolved. If not I will try 'continuous mode'

for those experiencing this issue here are some helpful threads

https://forums.whirlpool.net.au/thread/38kwm449?p=10
https://www.snbforums.com/threads/a...or-isps-dhcp-did-not-function-properly.61907/
 
Just wanted to thank you for this info as this is the exact issue that I have been experiencing.

I haven't changed to continuous mode (yet) as I spoke to Optus as I could see my WAN IP was swapping between 2 addresses when using my Asus router and I thought this was the causing the issue. Optus have set my account to a fixed ip over dhcp and I am happy about this as I prefer this config.

Since changing to a fixed IP I haven't had the WAN DHCP issue for 5 days. I'm going to leave everything the same for another week to make sure the issue is resolved. If not I will try 'continuous mode'

for those experiencing this issue here are some helpful threads

https://forums.whirlpool.net.au/thread/38kwm449?p=10
https://www.snbforums.com/threads/a...or-isps-dhcp-did-not-function-properly.61907/

If it helps I recently posted about enabling "Continuous" mode on the 384.18 Alpha Merlin build - https://www.snbforums.com/threads/384-18_alpha-builds-testing-all-variants.63954/page-24#post-592893 (also AC86U)
 

Sign Up For SNBForums Daily Digest

Get an update of what's new every day delivered to your mailbox. Sign up here!
Top