Low-traffic utilization trap. Any way out?

I’m using a RUT240 with a 64kbps IOT plan at a remote location. Experiencing loss of connection from a couple of hours up to 4 months at a time. My thorough investigation uncovered tons of issues with RUT240 but one stands out: LTE chip randomly goes to sleep in low-traffic utilization scenarios. I’m looking at other Teltonika models such as RUT956 but it seems like it suffers the same issues.

In this post Teltonika acknowledged the problem but offered no solution yet. Any updates whatsoever? Are there ANY other models using same or different LTE chipsets that are not having this issue? Please offer any light at the end of this tunnel.

The fact that forum topics close after awhile for inactivity adds to frustration: can’t report my issue to relevant posts such as the one I just linked…

Thanks.

Anyone, please? Any Teltonika router that works well in low traffic utilization, low speed,(64kbps) random packets scenario without LTE chip going to power saving mode?

Hello @Aragorn,

Thank you for reaching out to us.

If you are specifically experiencing packet drops during periods of low network utilization, we recommend testing different MTU values for the mobile interface.

We suggest starting with an MTU of 1500 and, if the issue persists, gradually decreasing it in increments of 20 (e.g., 1500, 1480, 1460, and so on).

Note: Setting the MTU below 1280 on an interface may cause all interfaces sharing the same physical interface to stop supporting IPv4. Therefore, we do not recommend setting the MTU below 1280.

To change the MTU value in the device’s WebUI, follow these steps:

  1. Navigate to Network → WAN.

  2. Locate the active WAN interface and click Edit.

  3. Open Advanced Settings.

  4. Under Override MTU, enter the desired MTU value.

Additionally, could you please provide some more details about the issue?

  • Is the connection completely lost, or does the connection remain active but with reduced performance?

  • Could you share any investigation notes or test results you have collected?

  • How did you determine that the modem goes into a sleep state during periods of low utilization?

Regarding topics being automatically closed after a few weeks of inactivity, we have disabled this feature. Topics will therefore remain open even if there is no activity.

We look forward to your feedback.

Best regards,
P.

Thank you for the response.

>>>Is the connection completely lost, or active with reduced performance?

Complete loss of connection from a few weeks up to 4 months at a time. Internal auto restart stops working during that time. I’ve since installed an external watchdog device, after it forcefully restarts RUT240, connection usually restores until next blackout.

  1. During my recent tests: The interface stays up (the modem remains registered on T-Mobile, the PDP context stays active, DHCP lease renews succeed), but traffic throughput collapses to near zero for a period — pings to 8.8.8.8 time out, and the web UI becomes unreachable. It then recovers on its own after some minutes, or sometimes only after a reboot.

  2. Investigation notes / test results

    Ping-based keepalive logs: ping_reboot.sh runs every 30 s. During good periods it logs “Ping successful” reliably. During the bad periods the pings fail while the interface reports itself as up.

    DHCP renewal: The mob1s1a1_4 lease (7200 s) renews successfully every hour — broadcasting renew is immediately followed by lease of 72.250.80.28 obtained. So the DHCP path is healthy; this is not the lease-loss wedge.

    Traceroute capture: I have a continuous traceroute -m 5 8.8.8.8 loop logging every ~40 s. Over ~108 minutes of healthy operation, hop 1 (T-Mobile gateway, 66.94.23.158) responded in every single block with 0% loss, and the path to hop 5 was continuously reachable. I have not yet captured a block during an actual outage — the loop is still running to catch one. I can share the block for the failure window once it occurs.

    Signal: RSRP ~−98 to −102 dBm, RSRQ ~−9 to −11, SINR ~8–11, on Band 4 only (neighbor scans show no other band visible). So signal is marginal but usable.

    Memory: The router was separately hitting OOM kills of uhttpd; that has been addressed (concurrency limits lowered), but it is a distinct issue from the packet drops.

  3. How did I determine the modem goes into sleep?

I have not proven it does. I raised “sleep” as a hypothesis because the symptom pattern — traffic stops under low utilization, interface stays registered, recovers on its own — matches the low-utilization packet-drop behavior already reported for RUT240/241 in the community thread, and because other users have described the modem appearing to enter a low-power state. But I have no direct capture showing the modem asleep; I only have evidence that the interface stops passing traffic while remaining registered.

I’d like your help to actually confirm or rule it out. Specifically:

Is there a way to check the EC25’s current sleep/power state from the router (e.g. an AT command or a gsmctl query) that would show whether it is in a low-power state during an episode?

Does the EC25 on the RUT240 use AT+QSCLK, and is that configurable/persistent on this firmware? I have sent AT+QSCLK=0 manually (it returned OK), but I don’t know if it survives reboot or whether the hardware gating (USB_VBUS / DTR) overrides it.

Are there known firmware-side power-management settings on the RUT240 that affect the internal modem’s idle behavior?

I will test the MTU stepdown as suggested and report back, but I want to flag that since even small pings fail during the episodes, packet size is unlikely to be the trigger — I’ll confirm with data either way.


I can share logs but I think that must be done privately due to sensitive info there?

Thanks!

I have an update. So the most critical issue we found so far is that the modem chip freezes for months at a time and even schedule reboots and ping reboots don’t restart it. It looks like a freeze starts during brief signal deterioration, but we couldn’t confirm this.

According to provider, the modem remains registered and a small traffic, under 1 MB /day passes through it when it’s hung. It may work for months, then it freezes for up to 4 months and those freezes are caused by modem chip becoming unresponsive. Only a cold restart (cutting power) restores connection.

So it doesn’t look like a MTU mismatch issue.

So looking for a replacement unit, my question is whether this is a widespread issue among all LTE chipsets used in Teltonika which manifests only on conditions like mine (slow connection or perhaps in low-traffic IOT situations, thus not as widespread under normal LTE use), or it’s just a defective unit? If it is a known issue - is it related to low-traffic utilization problem described here, and is there a solution?

Can I get some help please? I described the cure issue in me previous post. If you need logs, please tell me how to send the privately.

I want to add that my Dataworks 64kbps IOT plan (on T-Mobile network) limits connection to a single band. In my case it’s band 4. There’s no bands aggregation.

Greetings,

For troubleshooting purposes, we will require more sensitive information from your end, such as the troubleshooting file, which may contain passwords, public IP addresses, serial numbers, and such. To avoid leaking this information, we have sent you a form to fill out, which you will receive in your e-mail inbox that you have registered your account with in the forums. In the Ticket ID field of the form, please enter the ID of this thread, which is 21846

Best Regards,
Justinas

I filled out the form but there was no way to attach logs. How do I submit logs?

Greetings,

We have sent you an email, you can attach the troubleshoot file and any additional logs you have already collected to the reply to it.

Best Regards,
Justinas