уровне. Спасибо!

working VNC console. Since September 16 the server has been losing packets. It started at 12-16%, settled at 0.5-2% average per day, and since October 13 it has jumped to 6-10% average per day with individual probes up to 73% loss. I measure automatically every 10 minutes (100 pings) from two independent points in Europe - not from Russia, so Russian censorship is irrelevant here. From the same points, in the same minutes, servers at three other providers lose 0.00-0.1%. The VPS itself is clean: zero interface errors or drops, near-zero load. Traceroutes from both vantage points show the loss appearing on the last segment, inside the provider's network. Support is the real problem. Their SLA promises a first reaction under 15 minutes. In practice every ticket is first answered by a bot with an off-topic template (website TTFB, website migration). In one ticket the first human reply took 50 hours. They sent me a 41-packet traceroute as "proof" of no loss (statistically meaningless at 1-2% loss), a template about Russian censorship although my measurements are European, and in a month they never named the cause. The only offer was a migration to another node with an IP change - with no way to test the new address beforehand. My third ticket, with full measurements and consent to migrate, got only the bot reply; after I answered again asking for an administrator, there was silence for two days. Pros: fair price, good panel and VNC, a non-standard port was opened within days, node migration is offered free of charge. But if your workload depends on the network actually delivering packets, test the location yourself before paying. I will update this review if they ever fix it.