We have two RUTX50 modems which we’ve had a fair bit of stability issues with. They are both running the RUTX_R_00.07.07.1 firmware. The modems are used in a primary/secondary setup on another router, and yesterday we lost connection with the system. After a power cycle, both modems came back online again.
When checking the event log, there are a lot of “IPv4 connectivity started to fail”/“IPv4 connectivity restored” on both the modems. The primary modem seems to have disconnected the mobile modem two days earlier without it being reconnected before the power cycle (at 2024-05-20 16:05) - see the log below.
A side note: Looking at the logs, it also seems like the LAN connection is somewhat unstable. We’ve experienced the speed dropping to 100Mb/s, and also the port state changes without us interacting with the system.
Thank you for the response. We’ve been thinking about setting up auto rebooting to mitigate issues, but it feels a bit wrong without understanding the underlying issue.
Did you fully solve the issue by a power cycle? We’re getting the "IPv4 connectivity started to fail”/“IPv4 connectivity restored” events already 30 minutes after a power cycle.
I don’t know. The rsyslogd runs on another system (an Intel NUC with Linux), here is my /etc/rsyslog.d/02-5grouter.conf;
if $msg contains "mob1s1a1 ([0-9]+): Failed to open device, errno: 2" then /root/usbrelay/usbrelay-off-on.sh
if $msg contains "network: IPv4 connectivity started to fail" then /root/usbrelay/usbrelay-off-on.sh
usbrelay-on-off.sh is a simple script tp power off then on after a 5 sec an external relay. In addition to this syslog monitoring I ping the lan address of the router and 8.8.8.8 from the NUC if one or both fail I drive a reboot via the relay.
We also see this behavior on another RUTX50 modem in another location with a different network provider as well.
We also tried to downgrade the firmware on one of the routers to RUTX_R_00.07.05.4 - we have not seen any such errors on that router after the downgrade.
I have at least one use case (Failed to open device …) where both the ping reboot and scheduled reboot were unable to recover the device I had to come on site and powercycle the device. This was with version 07.05.x if I remember correctly, and occurs since but taken care of now with the external relay.
I have a RUTX50 and i’ve also noticed the IPV4 failing / IPV4 restored messages on a regular basis since the 07.07.01 firmware upgrade. But it hasn’t been an issue for me, however the router is configured to reboot twice a day though.
We got some feedback from Teltonika support that these event logs originate from a new function introduced in the RUTX_R_00.07.07.1 firmware - the “Network → Internet Status” monitor.
This apparently has a too low timeout, and can falsely report that the connection is failing even if this is not the case. They advised us to ignore the messages, or disable the monitor (under the “Network → Internet Status” pane) until this has been fixed in a new firmware.
