A redirect configuration that repeatedly sends requests between URLs without reaching a final destination (e.g., A -> B -> A).
A redirect loop repeatedly sends a request through addresses without reaching a final usable destination. A simple example is A redirecting to B while B redirects back to A. More complicated loops can involve several rules or differences in protocol, hostname or path formatting.
The symptom is often a browser error about too many redirects. The underlying cause needs inspection: several layers may be issuing instructions, including the application, server or delivery service. Looking at only one redirect-manager screen may not reveal the entire cycle.
Find the specific rule conflict before changing the site. The goal is a coherent final address and a working response, not simply fewer rules. Disabling all redirects can break legitimate migrations or leave old links stranded, while adding another catch-all rule may make the behavior harder to understand.
Takeaway: Treat redirect loops as routing conflicts that need an exact chain of evidence. Identify the repeating step and responsible layer rather than changing unrelated URLs or disabling the whole redirect system.
Example
A hosting rule sends requests from one hostname to another, while an application setting redirects them back. Neither rule looks like a loop in isolation, but together they create one. The team needs to agree on the intended hostname and correct the conflicting behavior at its source.
What to check
Capture the sequence of status codes and destination addresses, then retest after a scoped correction. Verify that the final page returns the expected content and that unrelated important routes still work. Do not assume clearing a browser’s state repairs a loop that ordinary new requests still encounter.
For more detail, see Redirections in HTTP.