Newly purchased dedicated server is unusable and repeatedly disconnects

Resolved

Ccosiris15DEVELOPER4/5/2026

Hi Zeabur team,

I purchased a dedicated server from Zeabur:

Aliyun Bangkok 2C 4GB

I created a new project and started deploying the very first service, but the server repeatedly became unreachable during deployment.

The system health page showed that 9 out of 9 Zeabur system components were abnormal, including:

vector-aggregator
zeabur-dns
zeabur-kube-watch
zeabur-log-api
nats
cadvisor
fluent-bit
ingress-controller
node-exporter

I also observed:

repeated "Unable to connect" errors
exported runtime logs are empty and only contain timestamp,message
the deployment repeatedly crashes or gets stuck during startup

This appears to be a server/platform-side issue rather than an application issue.

Please help urgently with one of the following:

fully repair and stabilize this server, or
replace/migrate me to a stable server, or
refund this server if it cannot be fixed quickly

This is a server purchased directly from Zeabur, not a self-managed external server.

Thanks.

2 Replies

BCBohan ChenEMPLOYEE4/7/2026

Hi — I just SSH'd into your Bangkok server (8.213.195.240) and the situation looks much better now:

Server uptime: 1 day 22 hours, k3s is active
Disk: 13% used (6G/49G), memory: 1.6G/3.6G — both healthy
All 9 Zeabur system components (cadvisor, fluent-bit, ingress-controller, nats, node-exporter, vector-aggregator, zeabur-dns, zeabur-kube-watch, zeabur-log-api) are now Running
Your 3 deployed services in this environment are all Running

The initial instability you observed during the first deployment was likely caused by the post-install component bootstrapping racing with your deployment — when the system pods were still being pulled/started, the cluster couldn't reliably route ingress, log, and DNS traffic, which is what caused the "Unable to connect" errors and empty log exports.

It looks self-recovered now. Please redeploy your service and see if it stays healthy. If you still hit "Unable to connect" or empty logs, please reply with the timestamp and I'll grab the cluster events for that exact moment to root cause it.

AAnonymous4/11/2026

This post has been inactive for a while. We will be closing it in 2 days if there is no new activity.

This issue has been marked as resolved

New replies are disabled for resolved issues.

Forum

2 Replies