The issue is likely not with Axios itself but rather with the server configuration. Keeping an HTTP connection open for an extended period is generally not a good idea. Here’s what I'll do:

• Implementing a strategy using webhooks or another mechanism to notify the client when the request succeeds or fails. (See Can we reliably keep HTTP/S connection open for a long time?).

If maintaining a persistent HTTP connection is necessary, I'll configure the server settings accordingly. (See nginx keepalive timeout vs ELB idle timeout vs proxy_read_timeout).

As noted in the reference above, client timeout settings are not relevant if the server’s timeout configuration is shorter than the client’s timeout setting.