Hello G2 folks, we're looking at load balancers with real-time health checks and automatic failover across redundant servers, so a dead backend never reaches a user. What we're hoping to find:
From the load balancing category:
For teams relying on this, whose health checks and failover actually caught a real outage? And how did you test failover before trusting it in production?
##### Post Metadata - Posted at: 22 days ago - Author title: Marketer - Net upvotes: 1 ## Comments ### Comment 1 The way most teams actually test this, deliberately killing a server to confirm failover triggers, tests a different failure mode than the one that usually causes real damage. A killed server produces an unambiguous signal, no response at all, which a basic health check catches easily. The dangerous case in production is a server that's still technically answering health checks, still returning HTTP 200, but serving degraded or wrong results to real users, a "gray failure" that a simple health check configuration won't catch at all, since it's checking whether the server responds, not whether the response is actually correct. So "confidence it works before you need it" from a deliberate test and "it actually caught the outage that mattered" can be two different claims, and the gap between them is usually the health check's depth, whether it's just checking liveness or actually validating response correctness. ##### Comment Metadata - Posted at: 17 days ago - Author title: Tech Consultant ## Related discussions - [How well does Trello scale into a larger team?](https://www.g2.com/discussions/1-how-well-does-trello-scale-into-a-larger-team) - Posted at: over 13 years ago - Comments: 6 - [Can we please add a new section](https://www.g2.com/discussions/2-can-we-please-add-a-new-section) - Posted at: over 13 years ago - Comments: 0 - [Quantifiable benefits from implementing your CRM](https://www.g2.com/discussions/quantifiable-benefits-from-implementing-your-crm) - Posted at: over 13 years ago - Comments: 4