What's new
  • 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!

AdGuardHome AdGuardHome stops working after every router reboot - have to un-/reinstall it every time

jtbx

Occasional Visitor
Hi,

As the title says, I've been having this issue with AdGuardHome. If I'm remembering correctly, I first noticed it when I updated amtm a couple months ago (don't remember to which version unfortunately). I rebooted my router (I do it weekly, just to keep things fresh) and then after a while I found out AdGuardHome wasn't running.

The AdGuardHome installer module in amtm doesn't report anything special, and instead points me to the web ui as it normally does:

Code:
Info:  Detected GT-AX11000 router.
Info:  Detected Linux platform.
Info:  Detected ARMv8 architecture.
Info:  JFFS custom scripts and configs are already enabled.
Info:  DNS Environment is Ready.
Info:  AI_VERSION=v1.7.4
Info:  ADGUARDHOME_BUILD=Release
Info:  ADGUARDHOME_VER=v0.107.54
Info:  Run Option 1 to if you want to change builds or upgrade when updates are available.
Info:  To visit AdGuardHome,
Info:  please go to http://10.0.0.1:3000.

Of course the web ui isn't running either. Doing a ps in the terminal, there's at least 10-20 entries of the following processes:

Bash:
 6444 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start
...
39383 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start

spread across the whole list with various pid's, from early ones , right after the reboot, to the latest.

Trying to start the service manually with the recommended commands doesn't do anything (besides spawning more of the above processes):

Code:
20:16 /tmp/home/root :: pidof AdGuardHome
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: service start_AdGuardHome
Done.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome start
20:19 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.

The only way to get AdGuardHome to start running again is to completely uninstall it via amtm and then reinstall it, I haven't figured out any other solution.

When it runs, it works fine and without any issues, until the next reboot. Neither do I have anything else malfunctioning in the router. I've tried even removing AdGuardHome's left-overs after the uninstall (e.g. its configuration file and anything else I can find), but it doesn't make a difference.

Neither is there any issue with its installation/uninstallation, they work as they should. Regarding versions of stuff, I keep everything updated as soon as amtm notifies me, so assume I have the latest versions of firmware, AMTM and the rest of the scripts.

It's not a huge deal to reinstall it after each reboot, it's just a bit weird. Though once, after a reboot, it wouldn't start even after reinstalling it, the installer would get stuck at the "Checking AdGuardHome... dead", at least during I waited for 30-60 mins. It suddenly started after the 5th-6th un- and reinstall.

Anyway, any help or pointer to the right direction would be greatly appreciated :D
 
Hi,

As the title says, I've been having this issue with AdGuardHome. If I'm remembering correctly, I first noticed it when I updated amtm a couple months ago (don't remember to which version unfortunately). I rebooted my router (I do it weekly, just to keep things fresh) and then after a while I found out AdGuardHome wasn't running.

The AdGuardHome installer module in amtm doesn't report anything special, and instead points me to the web ui as it normally does:

Code:
Info:  Detected GT-AX11000 router.
Info:  Detected Linux platform.
Info:  Detected ARMv8 architecture.
Info:  JFFS custom scripts and configs are already enabled.
Info:  DNS Environment is Ready.
Info:  AI_VERSION=v1.7.4
Info:  ADGUARDHOME_BUILD=Release
Info:  ADGUARDHOME_VER=v0.107.54
Info:  Run Option 1 to if you want to change builds or upgrade when updates are available.
Info:  To visit AdGuardHome,
Info:  please go to http://10.0.0.1:3000.

Of course the web ui isn't running either. Doing a ps in the terminal, there's at least 10-20 entries of the following processes:

Bash:
 6444 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start
...
39383 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start

spread across the whole list with various pid's, from early ones , right after the reboot, to the latest.

Trying to start the service manually with the recommended commands doesn't do anything (besides spawning more of the above processes):

Code:
20:16 /tmp/home/root :: pidof AdGuardHome
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: service start_AdGuardHome
Done.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome start
20:19 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.

The only way to get AdGuardHome to start running again is to completely uninstall it via amtm and then reinstall it, I haven't figured out any other solution.

When it runs, it works fine and without any issues, until the next reboot. Neither do I have anything else malfunctioning in the router. I've tried even removing AdGuardHome's left-overs after the uninstall (e.g. its configuration file and anything else I can find), but it doesn't make a difference.

Neither is there any issue with its installation/uninstallation, they work as they should. Regarding versions of stuff, I keep everything updated as soon as amtm notifies me, so assume I have the latest versions of firmware, AMTM and the rest of the scripts.

It's not a huge deal to reinstall it after each reboot, it's just a bit weird. Though once, after a reboot, it wouldn't start even after reinstalling it, the installer would get stuck at the "Checking AdGuardHome... dead", at least during I waited for 30-60 mins. It suddenly started after the 5th-6th un- and reinstall.

Anyway, any help or pointer to the right direction would be greatly appreciated :D
What other scripts do you have installed?
 
What other scripts do you have installed?
Hi, sorry for the delayed reply, was away from home.

Those are my scripts currently:

Code:
Skynet            7.6.4
YazFi            4.4.4
scribe            3.2.0
connmon            3.0.2
vnStat            2.0.6
RTRMON            2.1.4
ntpMerlin        3.4.5
scMerlin        2.5.8
spdMerlin        4.4.5
uiScribe        1.4.5
YazDHCP            1.0.6
AdGuardHome        1.7.5
AGH bin            v0.108.0-b.62
WICENS            4.04
VPNMON-R3        1.3.10

I've installed AdGuardHome's beta to see if it has the same issue, and it does.
I think the pid file for AdGuardHome is somehow related, maybe it gets stuck waiting for the file to be created/available, but it's just speculation, and I don't know how to investigate it further.
 
Hi, sorry for the delayed reply, was away from home.

Those are my scripts currently:

Code:
Skynet            7.6.4
YazFi            4.4.4
scribe            3.2.0
connmon            3.0.2
vnStat            2.0.6
RTRMON            2.1.4
ntpMerlin        3.4.5
scMerlin        2.5.8
spdMerlin        4.4.5
uiScribe        1.4.5
YazDHCP            1.0.6
AdGuardHome        1.7.5
AGH bin            v0.108.0-b.62
WICENS            4.04
VPNMON-R3        1.3.10

I've installed AdGuardHome's beta to see if it has the same issue, and it does.
I think the pid file for AdGuardHome is somehow related, maybe it gets stuck waiting for the file to be created/available, but it's just speculation, and I don't know how to investigate it further.
Automatically I can tell you adguardhome has some compatability issues with syslog-ng on entware setups. You may want to try removing scribe. See if it resolves your issue.
 
Sorry to chime in OPs post but when this happens to me I just disable, IPV6, UPNP, Remove Skynet and then disable Asus Firewall then do anotber clean boot and works then renable everything. I guess Adguard don't like other stuff hogging port 80 is it?
 
Sorry to chime in OPs post but when this happens to me I just disable, IPV6, UPNP, Remove Skynet and then disable Asus Firewall then do anotber clean boot and works then renable everything. I guess Adguard don't like other stuff hogging port 80 is it?
oh, it uses port 80? thats a different story. Might as well turn the router off then, otherwise nothing will work.
 
Same/similar issue, I think it's related to uncorrectly handled port change for dnsmasq or something close to that.
Issue is visible after router restart.

Code:
Apr 25 01:30:10 AdGuardHome[30330]: 2025/04/25 01:30:10.756321 [info] dnsproxy: warning: binding attempt=1 err="listen tcp 0.0.0.0:53: bind: address already in use"

Also seems it doesn't play ince with Guest Network Pro (hotspot):
Code:
Apr 25 02:13:49 rc_service: watchdog 3116:notify_rc start_dnsmasq 1
Apr 25 02:13:49 dnsmasq[85597]: failed to create listening socket for 192.168.52.1: Address already in use
Apr 25 02:13:49 dnsmasq[85597]: FAILED to start up
Apr 25 02:13:49 custom_script: Running /jffs/scripts/service-event-end (args: start dnsmasq)


snz@GT-AX6000-CFD8:/jffs/scripts# cat /etc/dnsmasq.conf |grep 52.1
dhcp-option=br52,6,192.168.52.1


Ultra temporary fix is:

Code:
killall -9 dnsmasq
/mnt/ROUTER_DATA/entware/etc/init.d/S99AdGuardHome restart

CC: @SomeWhereOverTheRainBow
 
Last edited:
Hi,

As the title says, I've been having this issue with AdGuardHome. If I'm remembering correctly, I first noticed it when I updated amtm a couple months ago (don't remember to which version unfortunately). I rebooted my router (I do it weekly, just to keep things fresh) and then after a while I found out AdGuardHome wasn't running.

The AdGuardHome installer module in amtm doesn't report anything special, and instead points me to the web ui as it normally does:

Code:
Info:  Detected GT-AX11000 router.
Info:  Detected Linux platform.
Info:  Detected ARMv8 architecture.
Info:  JFFS custom scripts and configs are already enabled.
Info:  DNS Environment is Ready.
Info:  AI_VERSION=v1.7.4
Info:  ADGUARDHOME_BUILD=Release
Info:  ADGUARDHOME_VER=v0.107.54
Info:  Run Option 1 to if you want to change builds or upgrade when updates are available.
Info:  To visit AdGuardHome,
Info:  please go to http://10.0.0.1:3000.

Of course the web ui isn't running either. Doing a ps in the terminal, there's at least 10-20 entries of the following processes:

Bash:
 6444 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start
...
39383 myusername    3424 S <  {S99AdGuardHome} /bin/sh /opt/etc/init.d/S99AdGuardHome monitor-start

spread across the whole list with various pid's, from early ones , right after the reboot, to the latest.

Trying to start the service manually with the recommended commands doesn't do anything (besides spawning more of the above processes):

Code:
20:16 /tmp/home/root :: pidof AdGuardHome
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: service start_AdGuardHome
Done.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.
20:18 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome start
20:19 /tmp/home/root :: /opt/etc/init.d/S99AdGuardHome check
 Checking AdGuardHome...              dead.

The only way to get AdGuardHome to start running again is to completely uninstall it via amtm and then reinstall it, I haven't figured out any other solution.

When it runs, it works fine and without any issues, until the next reboot. Neither do I have anything else malfunctioning in the router. I've tried even removing AdGuardHome's left-overs after the uninstall (e.g. its configuration file and anything else I can find), but it doesn't make a difference.

Neither is there any issue with its installation/uninstallation, they work as they should. Regarding versions of stuff, I keep everything updated as soon as amtm notifies me, so assume I have the latest versions of firmware, AMTM and the rest of the scripts.

It's not a huge deal to reinstall it after each reboot, it's just a bit weird. Though once, after a reboot, it wouldn't start even after reinstalling it, the installer would get stuck at the "Checking AdGuardHome... dead", at least during I waited for 30-60 mins. It suddenly started after the 5th-6th un- and reinstall.

Anyway, any help or pointer to the right direction would be greatly appreciated :D
Hi, sorry for the delayed reply, was away from home.

Those are my scripts currently:

Code:
Skynet            7.6.4
YazFi            4.4.4
scribe            3.2.0
connmon            3.0.2
vnStat            2.0.6
RTRMON            2.1.4
ntpMerlin        3.4.5
scMerlin        2.5.8
spdMerlin        4.4.5
uiScribe        1.4.5
YazDHCP            1.0.6
AdGuardHome        1.7.5
AGH bin            v0.108.0-b.62
WICENS            4.04
VPNMON-R3        1.3.10

I've installed AdGuardHome's beta to see if it has the same issue, and it does.
I think the pid file for AdGuardHome is somehow related, maybe it gets stuck waiting for the file to be created/available, but it's just speculation, and I don't know how to investigate it further.
Same/similar issue, I think it's related to uncorrectly handled port change for dnsmasq or something close to that.
Issue is visible after router restart.

Code:
Apr 25 01:30:10 AdGuardHome[30330]: 2025/04/25 01:30:10.756321 [info] dnsproxy: warning: binding attempt=1 err="listen tcp 0.0.0.0:53: bind: address already in use"

Also seems it doesn't play ince with Guest Network Pro (hotspot):
Code:
Apr 25 02:13:49 rc_service: watchdog 3116:notify_rc start_dnsmasq 1
Apr 25 02:13:49 dnsmasq[85597]: failed to create listening socket for 192.168.52.1: Address already in use
Apr 25 02:13:49 dnsmasq[85597]: FAILED to start up
Apr 25 02:13:49 custom_script: Running /jffs/scripts/service-event-end (args: start dnsmasq)


snz@GT-AX6000-CFD8:/jffs/scripts# cat /etc/dnsmasq.conf |grep 52.1
dhcp-option=br52,6,192.168.52.1


Ultra temporary fix is:

Code:
killall -9 dnsmasq
/mnt/ROUTER_DATA/entware/etc/init.d/S99AdGuardHome restart

CC: @SomeWhereOverTheRainBow
I greatly appreciate @muffintastic 's early satire and help on this thread, but i also want to share a known compatibility issue with newer syslog-ng version used by scribe which causes adguarhome to fail to start on router reboots because adguardhome does not detect the syslog. Post in thread 'AdGuardHome New Release's 2023' https://www.snbforums.com/threads/adguardhome-new-releases-2023.85191/post-890313 I would appreciate if anybody posting issues in this thread would try uninstalling scribe to see if their issue goes away. Please report back here. The OP of this thread has scribe installed, and this issue has only been reported on newer model routers running scribe. The old generation AC models are uneffected.
 
Same/similar issue, I think it's related to uncorrectly handled port change for dnsmasq or something close to that.
Issue is visible after router restart.

Code:
Apr 25 01:30:10 AdGuardHome[30330]: 2025/04/25 01:30:10.756321 [info] dnsproxy: warning: binding attempt=1 err="listen tcp 0.0.0.0:53: bind: address already in use"

Also seems it doesn't play ince with Guest Network Pro (hotspot):
Code:
Apr 25 02:13:49 rc_service: watchdog 3116:notify_rc start_dnsmasq 1
Apr 25 02:13:49 dnsmasq[85597]: failed to create listening socket for 192.168.52.1: Address already in use
Apr 25 02:13:49 dnsmasq[85597]: FAILED to start up
Apr 25 02:13:49 custom_script: Running /jffs/scripts/service-event-end (args: start dnsmasq)


snz@GT-AX6000-CFD8:/jffs/scripts# cat /etc/dnsmasq.conf |grep 52.1
dhcp-option=br52,6,192.168.52.1


Ultra temporary fix is:

Code:
killall -9 dnsmasq
/mnt/ROUTER_DATA/entware/etc/init.d/S99AdGuardHome restart

CC: @SomeWhereOverTheRainBow
Do you have a better fix ?
After upgrading my AX6000 from 3004 to 3006 i have the same problem, AGH don't change dnsmasq port to 553 and the is a conflict.
Thanks !
 
Add me to the list of users with this issue: AGH worked perfectly under the 3004 series of firmware, but completely chokes under this new 3006 series (at least on an GT-AX6000). Round and round failures after rebooting, if you can get it working and installed at all. I gave up and restored back to 3004.388.8_4 (thanks to BACKUPMON, it was a piece of cake, which was welcome relief after banging my head against the wall fighting 3006 for hours).
 
Last edited:
Add me to the list of users with this issue: AGH worked perfectly under the 3004 series of firmware, but completely chokes under this new 3006 series (at least on an GT-AX6000). Round and round failures after rebooting, if you can get it working and installed at all. I gave up and restored back to 3004.388.8_4 (thanks to BACKUPMON, it was a piece of cake, which was welcome relief after banging my head against the wall fighting 3006 for hours).
sorry for the delays, I am traveling for work at the moment. If I was going to give you a time table to deal with this issue, it will probably be in early june. If someone is kind enough to point me at what changes I should be looking at to rectify this issue, I may be able to do the fixes while on my travels, but this patch will rely on individuals here testing it before I release it. I am still on older firmware on my main router, I do not feel comfortable flashing my router with current firmware from the other side of the U.S. I will however adjust the script in an attempt to patch based on feedback of my peers i.e. @dave14305 , @thelonelycoder , @Martinski , @Viktor Jaep , or anyone else who doesn't mind providing helpful direction.
 
@dave14305 @RMerlin
To your knowledge, how does
  • dnsmasq.conf
  • dnsmasq-INDEX.conf (for SDN, only .add is supported, INDEX = SDN index number)
change the dnsmasq behaviors. Wouldn't it have made more sense to set a universal behaviors *.conf, then point back to the universal behaviors *.conf for each SDN as --conf-file=? That way dnsmasq.postconf could universally manage the certain behaviors shared by all SDN?
 
Last edited:
Wouldn't it have made more sense to set a universal behaviors *.conf, then point back to the universal behaviors *.conf for each SDN as --conf-file=? That way dnsmasq.postconf could universally manage the certain behaviors shared by all SDN?
I don’t think there are many universal settings that should be applied to both primary and SDN networks. From what I remember, AGH assumes it can listen on 0.0.0.0:53, but these extra SDN dnsmasq instances are already occupying port 53 on their dedicated bridge/vlan.

You either need to restrict AGH to the main br0 or move all the SDN dnsmasq instances to a different port via dnsmasq-sdn.postconf, but that could break DNS for those networks if done wrong.
 
I don’t think there are many universal settings that should be applied to both primary and SDN networks. From what I remember, AGH assumes it can listen on 0.0.0.0:53, but these extra SDN dnsmasq instances are already occupying port 53 on their dedicated bridge/vlan.

You either need to restrict AGH to the main br0 or move all the SDN dnsmasq instances to a different port via dnsmasq-sdn.postconf, but that could break DNS for those networks if done wrong.
Thank you for your general, but helpful response. My other alternative is forgo independent client logging, and just send all client traffic on port 53 to AdGuardHome. i guess a dirty way to do this is IPtables where I only worry about the interface and the port. Or i can just forward dnsmasq request on port 53, to adguardhome on some other chosen port (e.g. 553). I am just trying to figure out the best pathway. A easy quick fix would be to restrict AGH to the main br0, but then I would (or users) would need special rules to allow bridged(or otherwise private) vlans to communicate to br0 for dns.

The only other solution I could think of is just to keep killng dnsmasq and all its instances (that would be a complete useless waste), or set all dns ports in dnsmasq and dnsmasq SDN instances to port=0 but how will this impact dnsmasq-dhcp for AIMesh and client configurations. Setting each dnsmasq instance to port=0 will allow users to manage client dns information strictly inside Adguardhome, but i don't know if it will impact AIMESH functionality meaning i do not know how dependent Aimesh is with dnsmasq for its configuration. @dave14305 In case you are wondering, I am just brainstorming. If you have any other helpful suggestions please feel free to present your ideas. Thank you for any contribution.
 
Last edited:
I have a very similar yet different issue which just started a few weeks ago. At random times I am seeing logs in the router showing errors related to AdGuard over and over which eventually crashes the router to the point all devices lose connectivity to the internet. The only thing that fixes it is to cycle the power switch. I was using the Beta version and have switched to the stable release version but am still seeing instances happening.

I have not upgraded the firmware yet.

The error is an interesting one is due to the fact that the router uses a different DNS servers and only specific devices use the AdGuard DNS servers. Yet it takes down the entire network.

A couple of the errors I am seeing are:

" connect: connection timed out"
"i/o timeout"

It looks like it threw 1000s of errors messages in the logs.

This is on a GTAX6000 running 3004.388.8_4
 

Support SNBForums w/ Amazon

If you'd like to support SNBForums, just use this link and buy anything on Amazon. Thanks!

Sign Up For SNBForums Daily Digest

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