Before yesterday (10/17/2022), I’ve always had a command prompt window (MS Windows) up running a constant ping at 3 second interval - so I can tell when the network starts to degrade or just stops responding (which has become very frequent in the last few months).As of yesterday morning, both ping and tracert commands consistently fail. As in no longer any response. So it appears the ports used for those commands are now being blocked on the 5g network? I have a 5g phone on Tmobile, and I see the same result. On 5g with hotspot turned on, with computer connected, ping and tracert fail 100%. If I force the phone to use LTE and stay off 5g, ping and tracert start working again. Don’t really understand why Tmobile would block such a basic network analysis command.This is in downtown Scottsdale AZ. As a sidenote, service on the 5g network degrades consistently every day after about 8am, and usually is consistently bad all weekend long. Works great before 8am most days.

Has ping / tracert been blocked on 5g network?

Userlevel 1

inductivesoul
First activity
1 reply
1 year ago
20 October 2022

Yes, as of my post it seems T-Mobile is actively blocking Ping, Tracert, Traceroute and ICMP in addition if they do go through at all, they are deprioritized to the point that they are not effective to use in any capacity.

This sucks really bad...

I use multiple uplinks and depending on the RTT, RTTSD, Loss it switches providers to balance connections across them…

This behavior from the ISP breaks tons of functions as well as my link down failover automations. I hope T-Mobile realizes that ping and tracert/tracroute are 1000% necessary for proper network management and trouble shooting.

Verizon and AT&T don't have this problem of blocking pings, it is a T-Mobile specific issue and really makes the brand look cheap and the network mismanaged.

You can have a secure network without blocking normal and essential network tools that have existed since before I was born.

Userlevel 4

+4

Hi, I too wish T-Mobile would fix this, and that the issue deserves to be newsworthy!

I have been having major issues related to my T-Mobile home internet, primarily devices like home assistants and their accessories that rely on lower latency to fulfil requests. My Nest Mini now takes on average 10 full seconds to begin loading the response. Beyond that, issues with security since my cameras have increased lag making my doorbell camera less useful, and problems with my DNS not able to properly filter the web. I’m not the only one and neither are any of you:

Clearly this is an aggressive rollout of provisioning. To report my locale, I’m in the PNW but most of you are in the South/Midwest states like Texas. Well, location doesn’t seem to matter IMO as I still have noticed many issues recently, to the timeframe given in this forum, in where my web filter has not been able to deny requests properly on one of my home computers. I believe ICMP details noted are related.

Sadly enough, I am JUST learning how to network in college, in my second year and taking classes related specifically to latency and ping... So just wanted to note that I am genuinely affected by this… My own learning material has me doing these types of ICMP ping requests in Windows, and some posts here are lining up with my college material, but while learning from other members here has been fun, it also means to realize that my learning just got a lot harder BECAUSE of my fervent interest in bleeding-edge wireless tech… It doesn’t have to be this way!

When testing using Fast.com and other methods, I’m anywhere between 50ms and 1.2seconds loaded for latency. While that hasn’t been an issue, what’s different is that now never reporting the lower number. One second of latency means I can pretty much throw away my Xcloud membership LMAO I have much better results playing using my phone’s mobile hotspot.

Other tests, like Edge’s in-browser utility, no longer can report latency at all and therefore the program hangs. Just as another user said, it makes it harder for a given person who doesn’t know what’s going on understand what’s happening. This issue is compounded when the customers end up calling staff who have no source of truth for information on ping/latency leading to no progress for either party.

All in all, this is snowballing quickly and the best action really is to roll back these changes or communicate the incoming fixes. At minimum, progress in this would feel like acknowledging that the official stance isn’t that “ping isn’t used anymore” by T-Mobile and instead the stance customers want to see would be closer to working on a solution to the existing latency and ICMP problem for T-Mobile Home Internet users Nationwide.

Thanks to other forum members here, I understand now that if I want to continue to effectively filter my web I now need an addition device, a VPN (a hardware solution akin to a PfSense router) between my modem and personal router… and I would rather configure my own hardware like the ASUS modem I had before, but T-Mobile Home Internet does not support other hardware even if I did try and use a hotspot or 3rd-Party 5G Modem. I really don’t want a bunch of networking equipment at home, I’d wanna leave that for school lol... so it’s horrible to see this provisioning and clearly it’s basically “love it or leave it” unless we outcry. So I’d like to join you all in a big +1 @inductivesoul @iTinkeralot @ShanStewart @Walkabt and ask T-Mobile to sympathize with customers, geeks and nerdy students alike and cool it with the way their locking down these gateways.

Userlevel 7

+8

I am in east TN with the Nokia gateway and I ran a quick test with a Macbook Pro and a Linux client. With the Macbook Pro the pings have a significant delay where several request time out messages were reported prior to an actual response typically 140+ ms between responses. I went to a Linux client and did a forced ipv6 ping based upon a prior test. I get a similar response performing IPv6 and IPv4. Both have extreme delay times and are not very useful Of the forced 23 ipv6 ping packets 5 were received with a ~78% loss. So, yes it appears T-Mobile has jacked up the use of ICMP as a trouble shooting tool.

Based upon the two tests I ran it appears pings with IPv4 fail much worse than IPv6 pings but the use of pings now is pretty much useless.

I fail to see why they feel this would be positive for users. Make it more difficult for users to identify potential issues so they are totally blind. It just makes calls to TMO support less productive when users with such basic knowledge cannot tell support what they do not see.

Userlevel 7

+8

I can understand that. Another user was using the T-Mobile solution as a back up and doing a similar tactic with his negate firewall. In order to insure the failover could take place he had to disable the ping processing. When the Starlink fell it still transitioned to the backup. I don’t mean to lessen the importance of being able to have ICMP packets passed. I have seen the excessive loss and latency of the ICMP responses to 150 ms or more so I get it.

I just don’t think T-Mobile will be invested in resolving it right away. It is hard to tell how extensive the behavior is but it is generating some noise. Maybe with lots more noise they will pay attention.

J

Userlevel 1

jimibak
Learner
8 replies
1 year ago
26 October 2022

Ok, i spoke to T-Mobile tech and they said they no longer support ping so that is interesting.

FFS.

Since we'e picking & choosing cucial communication potocols now, I've decided to no longe suppot the lette 'r'.

Userlevel 7

+8

Ping and Trace Route both leverage ICMP to function so it just makes sense that both fail. I would guess with a VPN open both would work fine. It is hard to say what T-Mobile will do next with their solution. I have been getting more apprehensive since the gateway has now transitioned from the n71 to the n41 frequency. I have seen so many reports of users on the n41 where the throttling T-Mobile does pretty much cripples the cell beyond use. The prioritization for mobile handsets over the home gateways and the extra client loading in the urban locations just seems to be a bad mess. On the bright side my speeds have doubled or better on downloads much of the time.

Choking the ICMP traffic as they have makes no sense to me. Bad move. I used ports 8080 and 443 pinging and it still behaves the same. It looks like they just throttled ICMP to death. For users that do much more than check email, browser the web, or stream a few select services they will eventually jump ship and get another solution that is more reliable and capable.

S

Hi, I too wish T-Mobile would fix this, and that the issue deserves to be newsworthy!

Please call T-MOBILE and have them escalate this issue.

Userlevel 7

+8

So maybe in some locations it works but in others zip. You are among the fortunate. Nice download speed. You must be close in on an n41 frequency. I am on n41 but best I have seen is ~400 down.

ping google.com

PING google.com (108.177.122.138): 56 data bytes

Request timeout for icmp_seq 0

Request timeout for icmp_seq 1

Request timeout for icmp_seq 2

Request timeout for icmp_seq 3

Request timeout for icmp_seq 4

64 bytes from 108.177.122.138: icmp_seq=5 ttl=102 time=146.403 ms

Request timeout for icmp_seq 6

Request timeout for icmp_seq 7

Request timeout for icmp_seq 8

Request timeout for icmp_seq 9

64 bytes from 108.177.122.138: icmp_seq=10 ttl=102 time=76.933 ms

64 bytes from 108.177.122.138: icmp_seq=11 ttl=102 time=114.437 ms

Request timeout for icmp_seq 12

Request timeout for icmp_seq 13

Request timeout for icmp_seq 14

64 bytes from 108.177.122.138: icmp_seq=15 ttl=102 time=129.560 ms

Request timeout for icmp_seq 16

64 bytes from 108.177.122.138: icmp_seq=17 ttl=102 time=104.947 ms

Request timeout for icmp_seq 18

Request timeout for icmp_seq 19

Request timeout for icmp_seq 20

Request timeout for icmp_seq 21

Request timeout for icmp_seq 22…

--- google.com ping statistics ---

49 packets transmitted, 10 packets received, 79.6% packet loss

round-trip min/avg/max/stddev = 76.933/110.130/146.403/22.334 ms

Userlevel 7

+8

That is assuming trace route is working in your location. Recently ping and trace route have been problematic for users in a number of locations over the T-Mobile CGNAT solution.

B

I’m in Gulf shores, AL. Having the same issue as others, not get a ping back.

Spent over an hour with T-Mobile tech support and at one pint was told there is no way to turn off ping blocking on the gateway. My problem started just a few days ago also. Had worked fine for months. Except for this have had zero issues and getting 350+ down and 65+ up.

W

in Phoenix area too. These changes may be the cause of our unpleasant experience recently.

1) We have a Ting Fire device that monitor’s the electrical wiring of our home and it sends information out. It went offline. Couldn’t get it to work, but took the device to a neighbors house and no issues…

2) Had challenges joining a homeowner’s association meeting. The HOA property management firm uses Microsoft Teams. I was able to get on after a couple attempts. Microsoft email seems to have quirks sending out emails now too.

3) Don’t know if it is an Apple iOS/tvOS16.1 issue or T-Mobile change issue, but watching services through Apple TV has had streaming services like Disney, Hulu, Amazon say oops, we can’t connect or your are offline, this is sporadic.

Signal strength and quality haven’t changed. I check the numbers monthly.

If this were a deliberate decision, T-Mobile may be causing issues with the way other company’s products work.

I hope this is a technical error.

J

I live in WV and started noticing the ping issue rear its ugly head on Friday. I first thought it was something on my network going bad or my ad blocker messing up but narrowed it down to T-Mobile. I'm with everyone else, I can't understand why they are blocking such a critical thing. Glad to see that I'm not alone in the issue.

W

I hope more people realized there is an issue, find this post, sign in and post. This has to get to 40 pages in my mind before something happens to make this newsworthy.

Here’s my perspective, I’ve called in on several issues on different products recently, writing to a half dozen companies as a heads up that there are problems and neither issue has been resolved or hit the media.

Apple Airplay2 functionality is busted in iOS/iPadOS 16 on non-Apple devices. It may, or may not, have to do something with Apple Home App and Matter integration standards. I can find a lot more articles on issues for this topic right now on Reddit and other sites but none appear to connect the dots between products - they are focused on one manufacturer’s product, when the issue is Apple AND the manufacturer. From Sonos to Belkin, to receiver manufacturers like Yamaha and Onkyo this is big. This specific issue has been known since July when beta testers were testing the new iOS software and reported it. Airplay2 still works correctly in Mac OS - Ventura. Go to Belkin.com and look up their Soundform Connect and that device’s questions page has customers giving it bad reviews because of the issue blaming Belkin for selling the device that doesn’t work. Belkin’s response has been to please reach out to a special email address to get support. Airplay2 worked, pre iOS 16, then the issues started (the changes in the Home App’s code were being implemented due to upcoming Matter integration). While I am significantly impacted having multiple zones in my home using receivers to send AirPlay information to from my phones or iPads, most people probably just use Apple TVs and Apple Home Pod minis and don’t experience the issue, since Apple’s own devices don’t have the issue.
Amazon has been selling a device for at least two years where each device’s ethernet port of that product is hard coded with the same MAC address making them useless on ethernet if you have more than two devices - you can use over WI-FI... There are threads from the US and India on it, but none are more than four pages long. There are now reviews on Amazon saying the issue, I put a review one out there, it was reviewed by Amazon, but the group that received my call, the group that reviewed my review are not the development team and that’s who needs to see it OR an executive of the company. There are several options to fix, including spoofing a MAC address in the device firmware but it has gone no where. Too niche of an issue and not newsworthy. These are higher end products, sold by Amazon, and since WI-FI works for most that’s the resolution from support.

This issue is too new, and our situation is a little more unique than we may like to admit. It impacts us, and some of us more than others, but the average user probably just thinks it is the “the tower is under construction”. Here’s the headlines I have seen recently that are newsworthy to T-Mobile

everyone loves T-Mobile; https://www.tmonews.com/2022/10/t-mobile-earns-highest-score-in-j-d-power-2022-us-business-wireless-satisfaction-study/
we could be facing an uphill battle - I’ve only seen this report on one site, yet it seems like this is a bigger concern overall, to me. https://thedesk.net/2022/08/t-mobile-home-internet-problems-tower-upgrades/

P

phinster
First activity
2 replies
1 year ago
21 October 2022

My primary beef (the only one actually) is the loss of ping packets.

We will be using these devices as failover internet connections.

What I’ve observed is up to 90% packet loss with ICMP.

At that rate, it will continue to see that as a loss of internet connectivity when it is actually fine.

D

drP
First activity
1 reply
1 year ago
3 November 2022

Yes, as of my post it seems T-Mobile is actively blocking Ping, Tracert, Traceroute and ICMP in addition if they do go through at all, they are deprioritized to the point that they are not effective to use in any capacity.

This sucks really bad...

I use multiple uplinks and depending on the RTT, RTTSD, Loss it switches providers to balance connections across them…

This behavior from the ISP breaks tons of functions as well as my link down failover automations. I hope T-Mobile realizes that ping and tracert/tracroute are 1000% necessary for proper network management and trouble shooting.

Verizon and AT&T don't have this problem of blocking pings, it is a T-Mobile specific issue and really makes the brand look cheap and the network mismanaged.

You can have a secure network without blocking normal and essential network tools that have existed since before I was born.

^^^^ THIS RIGHT HERE!!! ^^^^
T-Mobile has a broken network or is actively breaking their network to the detriment of paying subscribers.

If this was a conscious business decision by T-Mobile, shame on you. Have you updated your terms of service to list the blocking of common network monitoring and management protocols as a feature of your service?

If it is an error - identify the root cause, fix it, and apologize.

Yours Truly,

A formerly happy, now very pissed off customer in North-Central Ohio…

S

This is really bad. This breaks Microsoft Teams among other services I am sure. MS Teams requires ping to verify connectivity to work (pings teams.microsoft.com). If ping fails then Teams will go offline.

Ping fails from my T-Mobile router gateway and from my T-Mobile phone.

J

JMS
Helper Level 1
5 replies
1 year ago
25 October 2022

Here in the S.F. Bay Area I had a similar ping problem for about a week -- about 70-80% lost packets. Starting two days ago, I’ve had no problems. Tried half a dozen sites, all respond with 0% loss. Hopefully the issue gets resolved where you’re at...

R

Randall310
First activity
2 replies
1 year ago
19 October 2022

Same here in Littlerock AR. I am a Citrix Admin (Working remote 100% of the time); I always have the end user ping our NetScaler gateway to see that the response time is when they say their remote session is sluggish.

Now even I cannot get a reliable ping response, but I am still able to work remotely.

Seems like ICMP has been throttled.

Not good.

Might have to go back to SuddenLink.

Cannot use this connection if I cannot even ping a remote site.

On the bright side, a new ISP is coming to my area with fiber, still under construction but fingers crossed they are better than SuddenLink and Century Link

Example:

Pinging google.com [142.250.72.46] with 32 bytes of data:
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Request timed out.
Reply from 142.250.72.46: bytes=32 time=117ms TTL=110
Request timed out.
Request timed out.
Request timed out.

With all the Request time outs, my Internet is fine.

If I RDP into a computer at work and ping a remote site, I get great ping responses.

R

raidz
First activity
1 reply
1 year ago
4 November 2022

Having the same problem here. Started a few weeks ago. Makes my load balancing/failover setup useless. TMHI is supposed to be my backup internet. Useless at this point. Going to cancel if this isn’t fixed soon.

M

I noticed about 4 days ago that most pings now fail. If I do a continuous ping(-t) roughly 1 in 20 gets a response, and all are over 100 ms. Using pinginfoview, I can set the port it uses, and they work.

TMobile needs to fix this. Lots of people use ping to check connectivity, and it’s just going to make them look bad.

Userlevel 7

+8

For some reason it appears T-Mobile is blocking ICMP traffic. Both the ping utility and trace routing use ICMP so it pretty much breaks both. Here in east TN now nothing when pinging. Ten packets sent zero responses. The other day the latency was 180 ms plus and 80% loss. Now nada

Userlevel 7

+8

Testing with a Garuda Linux client with a verbose ping to 8.8.8.8 currently it runs with roughly 100 ms latency but out of 43 packets sent 13 received responses so roughly 70% packet loss. Running a speedtest I am still obtaining 275-300 down and meh… 21 mbps upload so the bandwidth is there but the ICMP traffic is impaired or throttled. I sort of suspect it has something to do with throttling for one reason or another. I suppose it may be related to changes with the CGNAT solution but probably a throttling of the traffic. Testing with a MacBook Pro via a thunderbolt to ethernet adapter the response times are worse than with either of the Linux clients I have used but still bad either way.

traceroute to 8.8.8.8 (8.8.8.8), 64 hops max, 52 byte packets

1 www.webgui.nokiawifi.com (192.168.12.1) 2.145 ms 0.376 ms 0.279 ms

2 192.0.0.1 (192.0.0.1) 0.533 ms 0.698 ms 0.449 ms

3 * 192.0.0.1 (192.0.0.1) 27.721 ms 29.890 ms

4 * * 192.0.0.1 (192.0.0.1) 35.968 ms

5 192.0.0.1 (192.0.0.1) 30.243 ms * 46.172 ms

6 * * *

7 * * *

8 * * *

9 * * *

10 10.164.162.176 (10.164.162.176) 509.603 ms * *

11 * * *

12 * * *

13 * * *

14 * * *

15 * * *

16 *^C

Pretty clear it is useless without using a VPN.

Userlevel 7

+8

Now the Arcadyan and also the Sagemcon gateway both require the T-Mobile home internet mobile application ONLY for administrative management as the go to. It is the go forward T-Mobile seems to have decided upon. Probably due to cost reductions for device code development and etc… It does not feel like an improvement for the end user.

J

Userlevel 1

jimibak
Learner
8 replies
1 year ago
4 November 2022

Aaand it's broken again.

This is making me think tmo doesn't really know how to isp.

Userlevel 7

+8

Doh! Missed one rrrrrr mate!

Since we'e icking & choosing cucial communication potocols now, I've decided to no longe suppot the lette 'r'.

Has ping / tracert been blocked on 5g network?

79 replies

Reply

New badge winners

Reply

New badge winners

Sign up

Social Login

Login to the community

Social Login

Scanning file for viruses.

This file cannot be downloaded