Problem after Traccar update: Devices go offline after a few minutes

Hello,

We have updated our Traccar platform from version 4.12 to the latest version.

After the update, we are facing the following issue:

We have about 600 devices connected.

When we restart the server, all devices come back online normally.

However, after a few minutes, all devices automatically go offline for no apparent reason.

The only way to get them back is to restart the server, but the problem repeats itself soon after.

In the log, whenever this happens, several messages like these appear:

2025-04-28 16:35:39  INFO: Event id: xxxx, time: 2025-04-28 16:35:39, type: deviceUnknown, notifications: 0
2025-04-28 16:35:39  INFO: Event id: xxxx, time: 2025-04-28 16:35:39, type: deviceUnknown, notifications: 0
2025-04-28 16:35:46  INFO: Event id: xxxx, time: 2025-04-28 16:35:46, type: deviceUnknown, notifications: 0
2025-04-28 16:35:47  INFO: Event id: xxxx, time: 2025-04-28 16:35:47, type: deviceUnknown, notifications: 0
2025-04-28 16:35:48  INFO: Event id: xxxx, time: 2025-04-28 16:35:48, type: deviceUnknown, notifications: 0
2025-04-28 16:36:22  INFO: Event id: xxxx, time: 2025-04-28 16:36:22, type: deviceUnknown, notifications: 0
2025-04-28 16:36:28  INFO: Event id: xxxx, time: 2025-04-28 16:36:28, type: deviceUnknown, notifications: 0
2025-04-28 16:36:32  INFO: Event id: xxxx, time: 2025-04-28 16:36:32, type: deviceUnknown, notifications: 0
2025-04-28 16:36:44  INFO: Event id: xxxx, time: 2025-04-28 16:36:44, type: deviceUnknown, notifications: 0
2025-04-28 16:36:49  INFO: Event id: xxxx, time: 2025-04-28 16:36:49, type: deviceUnknown, notifications: 0
2025-04-28 16:36:49  INFO: Event id: xxxx, time: 2025-04-28 16:36:49, type: deviceUnknown, notifications: 0

After the devices go offline, if we restart the Traccar server, all devices reconnect and stay online normally for a few minutes, but then the issue repeats.

We are not making any database changes manually, nor any configuration changes after reboot.
The issue started only after updating to the latest version.

Does anyone have any idea what could be causing this behavior?
Is there any cache or connection session setting we need to adjust in the new version?

Thanks in advance.

1. What protocol are you using?
2. How frequently are your devices reporting data?

Thank you very much for your reply, Anton Tananaev.
I truly admire you and your work — you are an incredible person.

I noticed something:
Some protocols like GT06 and GPS103 were affected.
After I disabled some filters, the devices returned to normal operation.

I would make a backup of the database and traccar itself. Create a new traccar instance on another server and then import the database. Point 10 of your devices to this server and check if the issue is solved. In that case you know that the old server itself is the issue instead of traccar server configuration.