Web Hosting
OperationalManaged hosts running fyzno.com and customer applications. Patched weekly. The only outage on record is the OVH backbone reroute below.
Managed hosts running fyzno.com and customer applications. Patched weekly. The only outage on record is the OVH backbone reroute below.
Outbound SMTP relay with S/MIME keys rotated on the customer domain. Deliverability, signing endpoints, and DMARC alignment.
Authoritative DNS zones plus TLS certificate issuance and rotation. Renewals run on a 14-day buffer; DNSSEC on managed zones.
Public Next.js API routes and worker endpoints. Served behind a single ingress, traced end-to-end, signed where the data warrants it.
Background AI workflows: invoice triage, inbox summaries, report drafting. Every action is signed, logged, and reversible by a human.
Single primary node today. A second operational node and a separate backup node are being implemented; failover design and the migration plan are written down and progressively being rolled out. Encrypted daily backups are retained in Strasbourg, off the primary site. Production workloads run in the EU by default, in line with § 5 of our plain-language terms. The single outage on record was 138 days ago.
OVH themselves rerouted our Gravelines gateway during scheduled backbone maintenance on 22 February 2026. Aggregate downtime was 4 minutes; the failover design kicked in as intended. We phoned every affected customer within 24 hours and confirmed no messages or payments were missed. Root cause was on OVH's side, not ours.
The honest version of all of this lives in § 06 · Operational honesty (SLI) of our plain-language terms. We don't run a multi-region failover cluster; we do run a single primary node with documented failover design plus encrypted backups, and we're actively shipping a second operational node and a separate backup node. When something is wrong, we contact you as soon as we know.
Data note · uptime is shown in parts-per-thousand (‰). 1000‰ over a day means perfect uptime. The only outage on record is the 22 February 2026 OVH backbone maintenance incident above: 4 minutes, all customers contacted, root cause on OVH's side. Outside that day, uptime has been perfect.