Periodic report in Tg bot always shows 0m

[SPerekrestova]

Hi Sam,

Found a bug here. In my TG bot a periodic report always shows 0m active time. On the server in the dashboard uptime is correct.

[SamNet-dev]

Fixed. The bot's get_uptime relied on parsing Docker's StartedAt timestamp with date -d, which can fail silently on some systems (leaving se=0 → uptime = entire epoch → overflows format_duration → 0m).

Now uses the engine's Prometheus metric telemt_uptime_seconds as the primary source (always accurate, no date parsing), with docker inspect as fallback.

Run mtproxymax update && systemctl restart mtproxymax-telegram.

[SPerekrestova]

Thanks 🤝