Issues with Netduma R2 since April 2025

[adro]

Hello,

I've had my Netduma R2 for many years now with no issues at all! I disconnected the internet on the 4th of April and when I returned and reconnected on the 8th of April I noticed some issues with random internet crashes/disconnects. It seems like the router is randomly crashing / rebooting or loosing connection and reconnecting.

I'm on R2 3.3.280 and haven't updated the firmware recently or changed any settings. I've tried using the internet without the router and there were no issues. I've also tried multiple factory resets with no luck.

According to AI it's a memory related crash:

This log shows a memory-related crash in the dpiclass process on your Netduma R2 router, specifically a segmentation fault (SIGSEGV) caused by attempting to read from memory address 0 (null pointer). The router is trying to restart the process but it appears to be happening repeatedly.
The key issues in the log are:
do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000 - This is a null pointer dereference
process '//usr/bin/dpiclass' died restart:true - The router is restarting the crashed process
The crash is occurring in libadpi.so (DPI stands for Deep Packet Inspection)
This is likely a firmware bug in the router's DPI functionality (which analyzes network traffic for QoS or security purposes).
Possible solutions:
Disable DPI/QoS features in the router's settings
Perform a factory reset on the router
Update to the latest firmware if available
Contact Netduma support with these logs for assistance

Here is the system log with the lines related to the crash:
 

Mon Apr 14 18:22:16 2025 user.warn com.netdumasoftware.devicemanag: process '//usr/bin/dpiclass' died restart:true
Mon Apr 14 18:22:16 2025 user.info com.netdumasoftware.neighwatch: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 18:22:16 2025 user.info traceroute-wsapp: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 18:22:16 2025 user.info com.netdumasoftware.devicemanag: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 18:22:16 2025 kern.info kernel: [19125.608919] ra  = 77e412bb in libadpi.so[77e3c000+18000]
Mon Apr 14 18:22:16 2025 kern.info kernel: [19125.598557] epc = 77f88350 in libc.so[77f14000+97000]
Mon Apr 14 18:22:16 2025 kern.info kernel: [19125.580875] do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000

Was there a (stealth) software update on the router / cloud recently or did my router get hacked or is one of it's components dying?

Please help.

/Update

Tried with DPI/QoS off and same problem:
 

Mon Apr 14 20:06:54 2025 user.info com.netdumasoftware.devicemanag: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 20:06:52 2025 user.info com.netdumasoftware.neighwatch: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 20:06:51 2025 user.info traceroute-wsapp: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 20:06:51 2025 user.warn com.netdumasoftware.devicemanag: process '//usr/bin/dpiclass' died restart:true
Mon Apr 14 20:06:51 2025 kern.info kernel: [  869.600521] ra  = 77e172bb in libadpi.so[77e12000+18000]
Mon Apr 14 20:06:51 2025 kern.info kernel: [  869.589949] epc = 77f5e350 in libc.so[77eea000+97000]
Mon Apr 14 20:06:51 2025 kern.info kernel: [  869.572068] do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000

Traffi

[Fairypusher]

You can ask @Netduma Fraser for R2 Beta.
But if you make a reset or update you need a second router for config.

I have a Bug after Update.
I connect the WAN from R2 to Lan1 from my Router with Internet.

If your Dumaos dont start.

Now you can read the log.

192.168.77.1/adv_index.htm

[adro]

3 hours ago, Fairypusher said:

You can ask @Netduma Fraser for R2 Beta.
But if you make a reset or update you need a second router for config.

I have a Bug after Update.
I connect the WAN from R2 to Lan1 from my Router with Internet.

If your Dumaos dont start.

Now you can read the log.

192.168.77.1/adv_index.htm

I'd rather stay on 3.3.280 if it's more stable.

I noticed a cloud update after restarting the router and it's been running without issues for 1h 30min.
 

Mon Apr 14 20:36:03 2025 user.info com.netdumasoftware.config: Downloading files at mirror: http://api.netduma.com/...

Will report back in case it happens again.

RIP

 

RIP!

Mon Apr 14 22:35:51 2025 user.info traceroute-wsapp: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 22:35:51 2025 user.warn com.netdumasoftware.devicemanag: process '//usr/bin/dpiclass' died restart:true
Mon Apr 14 22:35:51 2025 kern.info kernel: [ 7235.366799] ra  = 77d842bb in libadpi.so[77d7f000+18000]
Mon Apr 14 22:35:51 2025 kern.info kernel: [ 7235.356574] epc = 77ecb350 in libc.so[77e57000+97000]
Mon Apr 14 22:35:51 2025 kern.info kernel: [ 7235.338126] do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000
Mon Apr 14 22:35:00 2025 kern.info kernel: [ 7184.613443] ash (24818): drop_caches: 3

...

Mon Apr 14 22:34:48 2025 user.info com.netdumasoftware.devicemanag: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 22:34:46 2025 user.info traceroute-wsapp: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 22:34:46 2025 user.info com.netdumasoftware.neighwatch: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 22:34:46 2025 user.warn com.netdumasoftware.devicemanag: process '//usr/bin/dpiclass' died restart:true
Mon Apr 14 22:34:46 2025 kern.info kernel: [ 7170.623937] ra  = 77e572bb in libadpi.so[77e52000+18000]
Mon Apr 14 22:34:46 2025 kern.info kernel: [ 7170.613747] epc = 77f9e350 in libc.so[77f2a000+97000]
Mon Apr 14 22:34:46 2025 kern.info kernel: [ 7170.596402] do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000
Mon Apr 14 22:34:43 2025 daemon.err cli.lua[3537]: goto outTLS hook failed

 

[NetTech]

I have been having a similar issues for a few days now. No firmware updates were manually applied on my end that would lead to such issues. An added note is that I do have a mesh network connected to my R2 and that happen to have release an update that makes note of "disconnects". However the random disconnects affect my PC that is wired directly to my R2. 

 R2 3.3.280

 

Mesh network TP-Link Deco XE75 Pro v2.0

New Features/Enhancements:
1. Enhance the parental control performance;
2. Improve the system stability and security.

Bug fixes:
1. Fix the bug that the clients frequently disconnect;
2. Fix the bug that the iOS device mistakenly reports "QOS cannot exceed 100mbps".

 

Quote

Mon Apr 14 20:04:07 2025 kern.info kernel: [  709.736990] do_page_fault(): sending SIGSEGV to dpiclass for invalid read access from 00000000
Mon Apr 14 20:04:07 2025 kern.info kernel: [  709.754696] epc = 77f0a350 in libc.so[77e96000+97000]
Mon Apr 14 20:04:07 2025 kern.info kernel: [  709.765122] ra  = 77dc32bb in libadpi.so[77dbe000+18000]
Mon Apr 14 20:04:07 2025 user.warn com.netdumasoftware.devicemanag: process '//usr/bin/dpiclass' died restart:true
Mon Apr 14 20:04:07 2025 user.info com.netdumasoftware.neighwatch: recovering state com.netdumasoftware.dpiclass
Mon Apr 14 20:04:07 2025 user.info traceroute-wsapp: recovering state com.netdumasoftware.dpiclass

 

[adro]

15 minutes ago, NetTech said:

I have been having a similar issues for a few days now. No firmware updates were manually applied on my end that would lead to such issues. An added note is that I do have a mesh network connected to my R2 and that happen to have release an update that makes note of "disconnects". However the random disconnects affect my PC that is wired directly to my R2. 

 R2 3.3.280

 

Mesh network TP-Link Deco XE75 Pro v2.0

New Features/Enhancements:
1. Enhance the parental control performance;
2. Improve the system stability and security.

Bug fixes:
1. Fix the bug that the clients frequently disconnect;
2. Fix the bug that the iOS device mistakenly reports "QOS cannot exceed 100mbps".

 

 

Interesting, that's the same error as mine! Other R2 users have also reported issues recently, but their topics have been moved to the private troubleshooting section and I can't verify if they have the same error as us.

I have disconnected my bricked router for now. Waiting for a response from Netduma.

[Netduma Fraser]

Could you provide the full logs please when it happens, I appreciate you providing the snippets you think are relevant but the full logs would be much more helpful

[adro]

On 4/15/2025 at 10:17 PM, Netduma Fraser said:

Could you provide the full logs please when it happens, I appreciate you providing the snippets you think are relevant but the full logs would be much more helpful

I plugged it in and got the same error twice within 10 minutes. Log attached.

log-1744904863828.txt

[Netduma Fraser]

Couple of things:

1. Are you using HybridVPN?
2. Have you got the router set to the correct time?
3. Your public IP appeared to change - what is the model of the modem/router the R2 is connected to and how have you set that to ensure all traffic flows to the R2? E.g R2 in its DMZ, modem/bridge mode
4. Set devices on the R2 to use a static/reserved IP address and extend the DHCP lease time and see if it improves please

[adro]

2 minutes ago, Netduma Fraser said:

Couple of things:

1. Are you using HybridVPN?
2. Have you got the router set to the correct time?
3. Your public IP appeared to change - what is the model of the modem/router the R2 is connected to and how have you set that to ensure all traffic flows to the R2? E.g R2 in its DMZ, modem/bridge mode
4. Set devices on the R2 to use a static/reserved IP address and extend the DHCP lease time and see if it improves please

Thanks for checking and the support.

1. No, like I said it happens even after factory reset with default settings.
2. Yes, using my time zone and it shows the correct time in the log.
3. Using modem from my isp, it's called ZTE ZXHN, DMZ and bridge mode are not setup currently.
4. I can try, but I think it will crash again since there is something wrong with the software or hardware. Will report back when I have some time for testing.

[Netduma Fraser]

Set the R2 WAN IP as static on the ZTE and put that in the ZTE DMZ also, then the static devices on the R2 and monitor it please

[NetTech]

On 4/15/2025 at 8:17 AM, Netduma Fraser said:

Could you provide the full logs please when it happens, I appreciate you providing the snippets you think are relevant but the full logs would be much more helpful

I have attached my logs. The solution at the moment is to disable QoS and DPI.

Equipment list:
1.ISP Modem: ET2251
2.R2 3.3.280
>Wired PC
3.Deco-XE75Pro (for Wireless) (Firmware Version: 1.2.1.0 Build 20240325 Rel. 57182
>Wired PC at Mesh device

1.Are you using HybridVPN?
No
2. Have you got the router set to the correct time?
Time zone and time are correct.
3. what is the model of the modem/router the R2 is connected to and how have you set that to ensure all traffic flows to the R2? E.g R2 in its DMZ, modem/bridge mode.
Modem only, no setting changes available. Modem model: ET2251

log-1744936655999.txt

[Netduma Fraser]

13 hours ago, NetTech said:

I have attached my logs. The solution at the moment is to disable QoS and DPI.

Equipment list:
1.ISP Modem: ET2251
2.R2 3.3.280
>Wired PC
3.Deco-XE75Pro (for Wireless) (Firmware Version: 1.2.1.0 Build 20240325 Rel. 57182
>Wired PC at Mesh device

1.Are you using HybridVPN?
No
2. Have you got the router set to the correct time?
Time zone and time are correct.
3. what is the model of the modem/router the R2 is connected to and how have you set that to ensure all traffic flows to the R2? E.g R2 in its DMZ, modem/bridge mode.
Modem only, no setting changes available. Modem model: ET2251

log-1744936655999.txt 73.98 kB · 3 downloads

If this is a DPI issue which it sounds like it could be if disabling prevents the issue for you then this has been fixed with the DumaOS 4 Early Access firmware so if you update to that the issue should stop

[NetTech]

On 1/6/2025 at 7:19 AM, Netduma Fraser said:

Nothing as of yet, the team have only just returned from the holidays today

 

2 hours ago, Netduma Fraser said:

If this is a DPI issue which it sounds like it could be if disabling prevents the issue for you then this has been fixed with the DumaOS 4 Early Access firmware so if you update to that the issue should stop

Do you have an up to date list of know issues on the DumaOS 4 Early Access firmware? Last I checked that firmware version produced DNS issues. The symptoms on version 4 actually appear very similar to my current experience on version 3. What can you say about the router having this current problem on version 3 when it had not come up since release. Are the programmers making changes on the cloud that affect router operations?

[Netduma Fraser]

53 minutes ago, NetTech said:

 

Do you have an up to date list of know issues on the DumaOS 4 Early Access firmware? Last I checked that firmware version produced DNS issues. The symptoms on version 4 actually appear very similar to my current experience on version 3. What can you say about the router having this current problem on version 3 when it had not come up since release. Are the programmers making changes on the cloud that affect router operations?

We don't generally provide a public facing list of issues on any router/firmware version. Issues reported on a firmware don't generally apply to everyone using it, have you tried the version? 

I can't really say anything, I'm not a dev but I would imagine that the bug comes to the forefront when triggered by certain settings/traffic, an overload of traffic etc. Not every bug is immediately obvious and these things can take time to appear but I've been told it was fixed so that'll be on DumaOS 4. The cloud just affects the servers on Heatmap/Geo-Filter really, nothing vital to the routers normal functioning.

[NetTech]

2 hours ago, mertocarlo said:

i have a same problem keep disconectin and very slow internet

 

you should post your logs.

[Netduma Fraser]

11 minutes ago, mertocarlo said:

log-1745093991559.txt i have netduma r2 and ist in dmz mode . Google pages open vervy slow and randomly disconecting

5 minutes ago, mertocarlo said:

its happened again log-1745094470115.txt

Please follow my steps from above and provide answers to all the questions asked

[adro]

@Netduma Fraser It can't be a coincidence that we are all getting the same memory related crash (sending SIGSEGV to dpiclass for invalid read access from 00000000). The router was working fine for me since early 2023 and suddenly everybody is getting this error.

cc @mertocarlo@NetTech
 

[Netduma Fraser]

4 hours ago, adro said:

@Netduma Fraser It can't be a coincidence that we are all getting the same memory related crash (sending SIGSEGV to dpiclass for invalid read access from 00000000). The router was working fine for me since early 2023 and suddenly everybody is getting this error.

cc @mertocarlo@NetTech

Out of the 1000's of people that have the router a couple of people receiving the same log entry is common. As mentioned the team knew about a DPI issue and it was fixed on DumaOS 4.