Description du projet
TraduitOur website keeps growing, and with the traffic come real problems. Pages load slowly for visitors far from our servers, and over the past few weeks we have been hit by waves of junk requests that take the site down for tens of minutes at a time. We are looking for an engineer who will fix this properly rather than patch a single hole.
The work splits into two large parts. The first is content delivery. We want a CDN connected and static assets cached the right way — images, scripts, styles and fonts — with sensible time-to-live headers and hash-based versioning so that updates reach visitors without a manual purge and stale copies do not linger. Dynamic responses must never be cached by mistake, so a careful, explicit rule set is essential here.
The second part is DDoS protection. We expect a filtering layer at the edge: rate rules, per-IP and per-session limits, reputation and behaviour based screening, a challenge for suspicious clients, and allow-lists for our own services and payment callbacks. Heavy search pages and forms, which are the usual targets, need dedicated shielding.
The third requirement, and it is mandatory, is monitoring. We want clear traffic graphs, a live attack dashboard, alerts on spikes and a log of blocked sources so we can see the picture ourselves instead of guessing. At the end we expect a short guide covering how to change thresholds, how to loosen a filter temporarily and how to read the metrics. Solid experience with nginx, CDN providers and real attacks is required, and examples of past deployments are welcome.
The work splits into two large parts. The first is content delivery. We want a CDN connected and static assets cached the right way — images, scripts, styles and fonts — with sensible time-to-live headers and hash-based versioning so that updates reach visitors without a manual purge and stale copies do not linger. Dynamic responses must never be cached by mistake, so a careful, explicit rule set is essential here.
The second part is DDoS protection. We expect a filtering layer at the edge: rate rules, per-IP and per-session limits, reputation and behaviour based screening, a challenge for suspicious clients, and allow-lists for our own services and payment callbacks. Heavy search pages and forms, which are the usual targets, need dedicated shielding.
The third requirement, and it is mandatory, is monitoring. We want clear traffic graphs, a live attack dashboard, alerts on spikes and a log of blocked sources so we can see the picture ourselves instead of guessing. At the end we expect a short guide covering how to change thresholds, how to loosen a filter temporarily and how to read the metrics. Solid experience with nginx, CDN providers and real attacks is required, and examples of past deployments are welcome.
Administration Espèces, Virement bancaire
