The test does not seem to run anymore after updating RUTX_R_00.07.23.4 and the odometer looks strange :
Hello,
Could you kindly confirm whether the same issue persists on the latest firmware version, 7.23.6 ?
You can download the firmware from our Wiki page here: https://wiki.teltonika-networks.com/view/RUTX50_Firmware_Downloads
Best regards,
The issue persists with 7.23.6
FYI I have 2 SIM cards from different providers; both show the issue. When clicking on the button to start the test it greys out and says ârunningâ and then it returns back to bleu after a while as if the test does run yet nothing is displayed, no speeds appear , the odometer remains as shown.
Thx
Issue persists after router incorrectly upgraded me to 07.23.6
I had a similar experience upgrading to 7.23.6 ; the update was downloaded and executed but the device remained on 7.23.4 , upon the second attempt it did the upgrade
I confirm, speedtest does not work. My configuration is RUTX50 eSIM, 1x classic SIM in position 1. With previous firmware versions, speedtest worked.
7.23.6 Speed Test works on my RUTX50.
Device: RUTX50
Kernel: 6.6.126
Firmware: RUTX_R_00.07.23.6
Build: 4e4c635823
Build date: 2026-06-12 06:33:27
Carrier: DNA Oyj
Interface: MOB1S1A1 - 185.174.xxx.xxx
Server: DNA Welho Oy - 62.197.175.67
Now, just thinking of probable causes. Have you guys checked your router DNS can resolve IP addresses Speed Test need to resolve before it can continue further?
Simple test from CLI. Remember to login root user & with admin password first.
root@rutx50:~# cat /etc/version
RUTX_R_00.07.23.6
root@rutx50:~# nslookup www.google.com
Server: 127.0.0.1
Address: 127.0.0.1#53
Name: www.google.com
Address 1: 142.251.155.119
Address 2: 142.251.150.119
Address 3: 142.251.152.119
Address 4: 142.251.151.119
Address 5: 142.251.154.119
Address 6: 142.251.153.119
Address 7: 142.251.157.119
Address 8: 142.251.156.119
Address 9: 2001:4860:482d:7700::
Address 10: 2001:4860:482c:7700::
Address 11: 2001:4860:4826:7700::
Address 12: 2001:4860:482b:7700::
Address 13: 2001:4860:4828:7700::
Address 14: 2001:4860:482a:7700::
Address 15: 2001:4860:4829:7700::
Address 16: 2001:4860:4827:7700::
root@rutx50:~#
Device: RUTX50
Kernel: 6.6.126
Firmware: RUTX_R_00.07.23.6
Build: 4e4c635823
Build date: 2026-06-12 06:33:27
root@RUTX50:~# cat /etc/version
RUTX_R_00.07.23.6
root@RUTX50:~#
root@RUTX50:~# nslookup www.google.com
Server: 127.0.0.1
Address: 127.0.0.1#53
Name: www.google.com
Address 1: 216.239.38.120
Address 2: 2001:4860:4802:32::78
Could it be related to extra packages that are installed that potentially conflict in some way ? I have email-relay, JSON-RPC support & Tailscale installed
Hi,
I have exactly the same problem. However, I found some speed test servers in the selection list that still work.
I upgraded from firmware version RUTX_R_00.07.09.1 to version RUTX_R_00.07.23.1. At the same time, I also updated the modem firmware from version RG501QEUAAR12A11M4G_04.201.04.201 to version RG501QEUAAR12A11M4G_04.202.04.202.
Since then, I have not only had the problem with the speed test, but also several other issues. I experienced various Wi-Fi connection drops, which made working with Citrix over VPN connections (WireGuard, ZScaler) impossible. The Citrix client kept losing the connection. There was nothing useful in the logs to identify the cause, and the connection to the mobile network did not drop. Using a hotspot via a smartphone, there were no connection drops at all.
Has anyone else experienced any further issues with the current firmware?
Yesterday, I installed the latest firmware. I will now monitor the behaviour.
Best regards,
Marvin
@akwe-xavante Great, it isnât DNS.
I just tested towards YouFibre - 185.43.128.5 (Stockton-on-Tees) Server your picture shows and itâs not working for me either. I tested all their 4 Servers and it appears Manchester and London works, but Stockton-on-Tees and Edinburg are down.
They have may some sites technical issues, any effecting work in progress, or have just limited source addresses these test sites accept testing and thus not answering from where I tried.
How if you try yourselves. Click Change near Server and select some other servers, for example YouFibre London or Manchester if they work better with you too.
By default opened selection screen shows only same country Servers where Iâm from, but using search box above I could find easily others too.
Indeed a couple of servers do work but the ones automatically selected and linked to the respective provider of my 2 simcards donât work. For example my provider Telenet BV/SRL - 213.224.175.169 does not work but when I browse to https://www.speedtest.net/api/js/perform-redirect?server_id=15212 using my desktop browser it does work; unfortunately the IP is not displayed but Iâm pretty sure it is the same server as Telenet only has 1 speedtest server listed
Tried manually selecting different servers to run speed tests the majority âDo Not Workâ.
I did get just one to work correctly after trying approx 7 servers.
@Clipper87 Right, Server / service side issues are not generally unheard of class problem.
I donât know if those speediest service providers aim certain SLA or not. But it isnât terribly hard to build and maintain service side services so that this kind of outages donât make unassuming clients and their equipment fall wrong conclusion that they have problems in their devices.
Out of several options how itâs quite cheap and scaleable all it would require them to make service DNS records (RRâs) updated with a short TTL (time-to-live) value and then just drop out from that pool those that failed. Because speed tests are quite much traffic it would be advisable to delegate that a separate subdomain from main to completely separate DNS-server where that dynamic updates with short TTLâs donât cause operational burden to main DNS. All things considered, not much more than single workday build this kind of thing. And it really doesnât require much of maintenance beyond normal operations. Contrary actually, as now the frees doing updates/upgrades/migrations ordinary work hours without outages as long as at least one of these services is kept running.
From the client side, like Teltonika device in this case, what they could do is to add a simple test to that test page, is check first quickly if the server picked is reachable. If not, then either pick random nearby or disable and switch âRun speed testâ button text to âChoose server firstâ. And then perhaps while generating Choose list test quickly contact each server with âncâ or language generating that page socket binding and gray out those servers which were not reachable. Or something like that which UX designers possibly have good ideas how it should be presented to user.
OK, if you got even one working then fault is not in your router.
Itâs your mobile or WAN side network connectivity issue somewhere upstream network.
A problem known as BGP ârouting screwupâ most likely, between service providers or within your own service provider internal network issue. Itâs not a thing you can fix by yourselves. You either wait until they fix or you may contact and tell them about the issue, which can speed up especially if they havenât for some odd reason noticed that themselves. That kind of things happen and are not uncommon.
if you contact, itâs usually helpful to attach traceroute output. Thatâs a CLI tool that prints what path packet go from your system where you run it towards an IP address you gave to it. It sends packets one hop at time printing as it goes RTT (Round-trip-time) from lowest, average and longest it took. Hopâs it didnât get reply return â*â, and it either means a problem, or just that whoever box that traffic went or is trying past doesnât allow tracing. Either out of ignorance, not understanding it would be a good and beneficial feature core network wide to allow it, or itâs a security policy mandate not allowing for real or imagined reasons and nobody else business they tell even if you ask usually.
If you need you can speed up traceroute a bit using -n option, so that it doesnât resolve addresses while it goes. But if you send output to someone for debug, please include names, it usually is helpful information to see NOC folks which way it came and they donât have to resolve names themselves. You can find contact information of who to report from WHOIS or RDAP services. Itâs bit of work, but bright side is you learn more how internet works if that is your thing you are interested.
Remember staying polite while you report and your changes getting taken seriously are better. Showing evidence kind ways, like
âIt seems from where I run this traceroute, that there is an issue somewhere there ⌠and could you perhaps take a look and see if there is something you could make to get it fixed, please. Then add that raceroute and add what service you were trying to reach.â.
Internet is collection of interconnected individual mostly autonomous networks exchanging traffic either peer-to-peer or purposely built facilities known as IX (internet exchanges) there isnât any single instance or party that would be able to fix things for you. Each party takes care of their own gear.
For an individual non professional network customer best option is to contact is always your own operator and service provider. They know these things how to approach solving routing issues. Speed they act depends usually greatly how important and much paying customer you are. Most important gets fastest turnaround and cheap consumer customers bit less.
Issue began imeadiatly after firmware upgrade and others are experiencing the same issue now after they upgraded. However it could just be a coincidence too.
Yes, could be coincidence or something else. Like if your operator side IP also changed after boot DHCP serivice decided to roll over new IP block retiring old from use.
If and when that happen itâs that new IP perhaps if not same address range it could be that service provider is rearranging addresses. Depending if your outside address is public IP, RFC 1918 or they are short of IPâs and turned on RFC 6598 CGNAT which overloads many customers IPâs same outside public address. And these days providers also quite aggressively make clients IPâs change denying renewing same IP after some period they have set they think is justified some privacy or just technical reason they rather not let anyone hang on some IP for too long. This latter seems to happen me at least once month, both IPv4, IPv6 together IPv6 PD here which is nuisance even though I have DDNS managing it, though independently from Teltonika router.
Whatever it is, if it works even for one Server then fault is not within you device. Because it is very unlikely that router would anyhow differentiate DNS given IPâs and then somehow work differently within that router. No, that is nonsense, there is absolutely zero reason Teltonika or any other vender also do that kind of silliness in feature implementation.
The problem cause and issue need to be fixed is outside your system, operator side either their network or somewhere further away up before reaching those Speed Test servers.
OK, I took look that IP address you picture showed 100.116.231.28 and this is what WHOIS service tells us about that IP block.
$ whois 100.116.231.28
#
# ARIN WHOIS data and services are subject to the Terms of Use
# available at: https://www.arin.net/resources/registry/whois/tou/
#
# If you see inaccuracies in the results, please report at
# https://www.arin.net/resources/registry/whois/inaccuracy_reporting/
#
# Copyright 1997-2026, American Registry for Internet Numbers, Ltd.
#
NetRange: 100.64.0.0 - 100.127.255.255
CIDR: 100.64.0.0/10
NetName: SHARED-ADDRESS-SPACE-RFC6598-IANA-RESERVED
NetHandle: NET-100-64-0-0-1
Parent: NET100 (NET-100-0-0-0-0)
NetType: IANA Special Use
OriginAS:
Organization: Internet Assigned Numbers Authority (IANA)
RegDate: 2012-03-13
Updated: 2024-05-24
Comment: This block is used as Shared Address Space. Traffic from these addresses does not come from IANA. IANA has simply reserved these numbers in its database and does not use or operate them. We are not the source of activity you may see on logs or in e-mail records. Please refer to http://www.iana.org/abuse/
Comment:
Comment: Shared Address Space can only be used in Service Provider networks or on routing equipment that is able to do address translation across router interfaces when addresses are identical on two different interfaces.
Comment:
Comment: This block was assigned by the IETF in the Best Current Practice document,
Comment: RFC 6598 which can be found at:
Comment: http://tools.ietf.org/html/rfc6598
Ref: https://rdap.arin.net/registry/ip/100.64.0.0
OrgName: Internet Assigned Numbers Authority
OrgId: IANA
Address: 12025 Waterfront Drive
Address: Suite 300
City: Los Angeles
StateProv: CA
PostalCode: 90292
Country: US
RegDate:
Updated: 2024-05-24
Ref: https://rdap.arin.net/registry/entity/IANA
OrgTechHandle: IANA-IP-ARIN
OrgTechName: ICANN
OrgTechPhone: +1-310-301-5820
OrgTechEmail: abuse@iana.org
OrgTechRef: https://rdap.arin.net/registry/entity/IANA-IP-ARIN
OrgAbuseHandle: IANA-IP-ARIN
OrgAbuseName: ICANN
OrgAbusePhone: +1-310-301-5820
OrgAbuseEmail: abuse@iana.org
OrgAbuseRef: https://rdap.arin.net/registry/entity/IANA-IP-ARIN
#
# ARIN WHOIS data and services are subject to the Terms of Use
# available at: https://www.arin.net/resources/registry/whois/tou/
#
# If you see inaccuracies in the results, please report at
# https://www.arin.net/resources/registry/whois/inaccuracy_reporting/
#
# Copyright 1997-2026, American Registry for Internet Numbers, Ltd.
#
$ _
If I havenât understood correctly you reside somewhere UK. Your service provider apparently uses this IANA (Internet Assigned Number Authority) IP-block with their permission and advise. From which that itâs not a) recommended but working b) certainly tells you are behind NAT and c) most likely that anti is used CGNAT (Carrier Grade NAT) and reasons we can only guess, but which is quite obvious. Your provider is running short useable IP blockâs and is stretching a bit rules in attempt to keep up services working.
Iâve not experienced myself (luckily) that kind of situation, but I think you could politely give some constructive feedback to your provider while expressing that what you experience now is not fun. Not even a bit, I guess.
Oh, and that CGNAT overloaded IP use perhaps explains also why they are not so keen that customers run Speed Tests there either.
Also it could be that iP you are not being shared externally with some other customer(s) that has had run so much Speed Tests that those providers have put limits or denied for some time that external address as a means to make unwanted traffic towards them to stop. Iâve done this in my former work when it was necessary as there was no other channel to inform overusing parties to make them stop.
Perhaps you could ask if they had another APN which you could change instead Auto, and get bit better connectivity with that.
Also if you have router logs, stored and still there in router or you have forwarded with syslog to another server for longer keeping them. You could check from logs did your address change after upgrade boot or not. And if it changed if old one still was from same CIDR prefix shown above. Here is what that covers
$ sipcalc 100.116.231.28/10
-[ipv4 : 100.116.231.28/10] - 0
[CIDR]
Host address - 100.116.231.28
Host address (decimal) - 1685382940
Host address (hex) - 6474E71C
Network address - 100.64.0.0
Network mask - 255.192.0.0
Network mask (bits) - 10
Network mask (hex) - FFC00000
Broadcast address - 100.127.255.255
Cisco wildcard - 0.63.255.255
Addresses in network - 4194304
Network range - 100.64.0.0 - 100.127.255.255
Usable range - 100.64.0.1 - 100.127.255.254
$
So in short, other things may have changed while that upgrade boot did happen. These providers consumer facing services can surprise occationally. Businesses offered usually more stable and they announce beforehand usually changes expecting being visible to customers.
e: changed bold part
coincidence ? Nope! I have 2 simcards from 2 different providers and the issue occurs with both. When I browse to speedtest.net on my desktop browser that gets itâs internet from the RUTX 5G connection all worksâŚ
My gut feeling is that the integral speed test is in fact running successfully, but the test has lost itâs connection with the UI, so no test progress is being displayed, and the result is not being displayed either.
Thereâs a great deal wrong with firmware 07.23.6.
But then why doesnât it fail with some servers ???