BitVPS

服务 状态

四个数据中心共 16 个组件的实时基础设施健康状况。由我们 ASN 之外的外部探针每 20 秒轮询一次;原始数据公开发布,包括不良数据。

更新时间 02:36 UTC 可用率窗口 90 天 外部探针 · 20 秒间隔
BitVPS network status

计划维护 即将进行

Network · CH-ZRH · Upstream carrier swap — BGP reconvergence inside 90 seconds.

2026-07-19 22:00 UTC 于 7 d 19 h · ~80 分钟

组件与 90 天历史

下方每个柱形代表一天,颜色表示当天最差状态。悬停柱形可查看日期及当天摘要。

Compute · IS-RKV

Iceland
100.00 % 90 天可用率
正常运行

Compute · NL-AMS

Netherlands
100.00 % 90 天可用率
正常运行

Compute · RO-BUH

Romania
99.98 % 90 天可用率
正常运行

Compute · CH-ZRH

Switzerland
99.96 % 90 天可用率
正常运行

Network · IS-RKV

Iceland
100.00 % 90 天可用率
正常运行

Network · NL-AMS

Netherlands
100.00 % 90 天可用率
正常运行

Network · RO-BUH

Romania
99.96 % 90 天可用率
正常运行

Network · CH-ZRH

Switzerland
99.97 % 90 天可用率
正常运行

Block storage · IS-RKV

Iceland
100.00 % 90 天可用率
正常运行

Block storage · NL-AMS

Netherlands
99.98 % 90 天可用率
正常运行

Block storage · RO-BUH

Romania
99.98 % 90 天可用率
正常运行

Block storage · CH-ZRH

Switzerland
100.00 % 90 天可用率
正常运行

Web control panel

100.00 % 90 天可用率
正常运行

Billing & payments

100.00 % 90 天可用率
正常运行

Anycast DDoS shield

99.97 % 90 天可用率
正常运行

Docs & marketing site

100.00 % 90 天可用率
正常运行
2026-04-14 — 2026-07-12
正常运行 性能下降 部分中断 重大中断

近期 事故

所有组件最近 14 次已解决事故——每次页面加载时由探针自动刷新。

Jun 25, 2026 08:47–09:49 UTC

部分中断 Compute · CH-ZRH

IOMMU group re-mapping required a short guest stun — ~3% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

62 分钟 停机时间
Jun 24, 2026 07:45–07:55 UTC

性能下降 Compute · IS-RKV

Hypervisor-level memory pressure investigation — ~9% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

10 分钟 停机时间
Jun 20, 2026 10:15–10:27 UTC

性能下降 Network · RO-BUH

ICMP rate-limit mis-tune surfaced during a probe sweep from SwissIX Zurich. 已完全解决,后续跟进事项内部追踪中。

12 分钟 停机时间
Jun 10, 2026 15:48–16:27 UTC

部分中断 Network · RO-BUH

IPv6 routing anomaly on carrier GTT (AS3257). 已完全解决,后续跟进事项内部追踪中。

39 分钟 停机时间
Jun 2, 2026 07:46–08:31 UTC

部分中断 Compute · RO-BUH

NUMA imbalance briefly elevated p99 boot times on — ~4% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

45 分钟 停机时间
Jun 1, 2026 19:21–20:06 UTC

部分中断 Anycast DDoS shield

L4 scrubbing capacity briefly saturated during a large amplification attack. 已完全解决,后续跟进事项内部追踪中。

45 分钟 停机时间
May 29, 2026 15:14–15:51 UTC

部分中断 Block storage · NL-AMS

Object-store index promotion briefly held the write lock. 已完全解决,后续跟进事项内部追踪中。

37 分钟 停机时间
May 28, 2026 08:45–08:51 UTC

性能下降 Compute · NL-AMS

IOMMU group re-mapping required a short guest stun — ~5% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

6 分钟 停机时间
May 28, 2026 13:09–13:28 UTC

性能下降 Anycast DDoS shield

False-positive mitigation affecting legitimate traffic. 已完全解决,后续跟进事项内部追踪中。

19 分钟 停机时间
May 27, 2026 07:08–07:55 UTC

部分中断 Block storage · RO-BUH

Zoned-namespace firmware upgrade on a subset of drives. 已完全解决,后续跟进事项内部追踪中。

47 分钟 停机时间
May 23, 2026 12:37–12:45 UTC

性能下降 Block storage · IS-RKV

Rebalance of a hot cell into a cooler shelf. 已完全解决,后续跟进事项内部追踪中。

8 分钟 停机时间
May 5, 2026 10:07–10:29 UTC

性能下降 Compute · CH-ZRH

IOMMU group re-mapping required a short guest stun — ~5% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

22 分钟 停机时间
Apr 27, 2026 19:17–19:23 UTC

性能下降 Compute · NL-AMS

Host-agent crash loop after a dependency rollout — ~11% of hosts affected. 已完全解决,后续跟进事项内部追踪中。

6 分钟 停机时间
Apr 19, 2026 12:11–13:05 UTC

部分中断 Network · CH-ZRH

DDoS event absorbed — peak 22 Gbps, scrubbed to edge. 已完全解决,后续跟进事项内部追踪中。

54 分钟 停机时间

月度可用率 汇总

全部 17 个组件按月汇总的服务可用率。低于 99.99 % 的月份将触发 SLA 信用补偿——详见 SLA

Jul 2026
99.995 %
可用率
Jun 2026
99.980 %
可用率
May 2026
99.990 %
可用率
Apr 2026
99.982 %
可用率
Mar 2026
99.999 %
可用率
Feb 2026
99.996 %
可用率

状态与 SLA 常见问题

我们如何衡量可用性、什么算作 Tier-1 事件、在哪里订阅,以及事后复盘如何发布。

How is uptime measured?
We measure uptime per service component (compute, network, storage, control plane) at one-minute granularity from independent probes in five regions. Daily and monthly figures are weighted by incident severity: a major outage counts at 100%, a partial outage at 70%, and degraded performance at 30% of its measured duration.
What counts as a Tier-1 incident?
A Tier-1 incident is a region-wide loss of customer-facing service: control-plane unreachable, all compute in a datacenter offline, or a network partition that prevents customers from reaching healthy nodes. Tier-1 events trigger an immediate banner here, an out-of-band post-mortem within 72 hours and a credit on the next invoice if SLA thresholds are breached.
Where can I subscribe to status updates?
Subscribe via the RSS feed at /status/rss for incident-only notifications, or watch the JSON endpoint at /status/feed.json for machine-readable polling. We do not offer email subscriptions — they require user accounts and we run an account-free status surface deliberately.
Do you publish post-mortems?
Yes. Every Tier-1 and Tier-2 incident receives a written post-mortem within 72 hours: timeline, root cause, customer impact, mitigation and remediation work. Post-mortems are linked from the incident entry below and clearsigned with the same PGP key used on the warrant canary, so authenticity can be verified offline.
What's the difference between Tier-1 and Tier-2 incidents?
Tier-1 is region-wide customer impact (full outage, control-plane loss). Tier-2 is partial impact: a single rack, a single transit provider degraded, or a control-plane subsystem (billing, console) impaired while compute keeps running. Both get post-mortems; only Tier-1 is eligible for SLA credits under our published 99.95% target.