Yealink T57W - random network service reset

[rikels]

Good day,

I hope this post is in the right place, and I apologize if the text is too long.

Structure Overview:​

We operate approximately 1,200 Yealink T57W phones across 25 branch offices.
Each branch has an SBC, and the phones are located in VLAN 160 (our phone VLAN).
Here at the headquarters, around 150 phones are also in this VLAN.
The 3CX system itself is located in a DMZ.

Issue Description:​

In general, the phones work without issues:

• Registration, telephony, and provisioning function flawlessly.

However, some users occasionally report that their phone suddenly loses network connectivity.
Error message: "No Service"
This issue has been occurring more frequently over time.

Previously, with older firmware versions, this problem did not exist.
Affected devices vary; there is no clear pattern or specific set of phones.

Since we use a RADIUS server (Aruba ClearPass), we can trace where the issue occurs.
We see that some phones sporadically re-authenticate. It appears as if the phone is "network-resetting" itself.

Following this, the phones undergo RADIUS authentication again and receive a new IP address from the DHCP server.
During this process, they are unreachable.
The issue becomes especially problematic when it happens during an active call!

Observations:​

• On the switch, we see that the MAC address re-registers on the interface—nothing else unusual.
• The issue occurs irregularly: sometimes 1-2 times, other times up to 10-20 times per device.
• Even rarely used phones (e.g., in the kitchen or server room) are affected.
• It also happens on phones without an SBC, connecting directly to 3CX.
• Some weeks, the issue is more frequent, while in others, it barely occurs.

Error Logs (Folder "Crash"):​

[09 21:29:02][killall:2777] send signo(9) to [cable_dot1x:1908].
[09 21:29:02][killall:2784] send signo(9) to [cable_dot1x:2780].
[09 21:29:12][kill:2932] send signo(15) to [svc_hpd.sh:373].
[09 21:29:12][kill:2932] send signo(15) to [appstart:507].
[09 21:29:12][kill:2932] send signo(15) to [appstart:2125].
[09 21:29:12][kill:2932] send signo(15) to [sh:2927].
[09 21:29:12][kill:2938] send signo(15) to [sh:2933].
[09 01:00:26][killall:1628] send signo(9) to [dot1x.exe:1049].
[09 01:00:35][killall:1641] send signo(9) to [dnsmasq:865].
[22 01:00:21][killall:1359] send signo(9) to [dnsmasq:875].
[22 01:00:21][killall:1392] send signo(9) to [dnsmasq:871].
[23 08:10:14][killall:3381] send signo(9) to [dnsmasq:1602].
[03 01:00:21][killall:1374] send signo(9) to [dnsmasq:867].
[03 21:09:26][killall:2011] send signo(9) to [dnsmasq:1595].
[03 21:32:00][killall:2263] send signo(9) to [dnsmasq:2012].
[03 21:43:14][killall:2514] send signo(9) to [dnsmasq:2264].
[03 21:46:03][killall:2745] send signo(9) to [dnsmasq:2515].

System Specifications:

Phone System:​

• Microsoft Windows Server 2022
• CPU: 16 cores
• RAM: 64 GB
• Storage: 190 GB
• 3CX Version: 20.0 Update 4 (Build 487 Release)

Phones:​

• Quantity: ~1,200 Yealink T57W
• Firmware: 96.86.0.81
• Hardware Version: 97.0.1.0.0.0.0

Maybe someone can find additional clues about this issue and confirm this phenomenon.

Thank you and best regards!



Please excuse me, I had this text translated using ChatGPT.

 

[NicholasP_3CX]

If the phone is receiving a new IP Address it would be logical for the network to reset and disconnect if a call is in progress!

You would need to readjust your DHCP settings to assign any new IPs during times when the phones are not in use.

 

[rikels]

If the phone is receiving a new IP Address it would be logical for the network to reset and disconnect if a call is in progress!

You would need to readjust your DHCP settings to assign any new IPs during times when the phones are not in use.

Hello,

the lease time of the phones has not expired.
And why should the phones do something like this 10-20 times in a row?
Additionally, I would like to know how to configure the DHCP server to assign new IP addresses when the phone is not in use.

 

[SteveITS]

happens on phones without an SBC, connecting directly to 3CX.

Is 3CX on VLAN 160 also?

Are these phones connected via wire, or wireless?

 

[NicholasP_3CX]

And why should the phones do something like this 10-20 times in a row?

I have no way of knowing what your network is doing to cause the phones to disconnect from the network.

Additionally, I would like to know how to configure the DHCP server to assign new IP addresses when the phone is not in use.

I would just configure the DHCP lease to renew at 4AM, when it is usually quite, unless you are a 24/7 operation. But you could still tell the DHCP server to just assign the same IP Address to a phone and not change it all the time.

 

[JohnS_3CX]

Hi,

This sounds exactly like the kind of case you should raise with Yealink Support.

They produce the hardware and firmware and would be in the best position to guide you in troubleshooting, and possibly collecting onboard logs that can reveal more clues.

 

[rikels]

Is 3CX on VLAN 160 also?

Are these phones connected via wire, or wireless?

The 3CX server is not in the same VLAN.
A Palo Alto 1410 is positioned between these two networks.
However, everything is working.
So why do only some phones experience issues, and why is it sporadic?
All phones are connected via cable.

 

[rikels]

Hi,

This sounds exactly like the kind of case you should raise with Yealink Support.

They produce the hardware and firmware and would be in the best position to guide you in troubleshooting, and possibly collecting onboard logs that can reveal more clues.

Yes, I think so too.
It's a shame—I was hoping that someone had already encountered this issue and could provide us with a "simple" solution.

I'll keep you updated on what Yealink Support does.