On Sat, Mar 27, 2021 at 3:19 PM 'Axel Wagner' via golang-nuts < golang-nuts@googlegroups.com> wrote:
> The best way to do it is probably by making an HTTP request and see if it > succeeds. In production, it's always a good idea to have a health check > endpoint anyways. So some service manager can check if it's alive and > restart it if necessary. Or so that a load balancer doesn't send traffic > until the service is healthy. > In addition, if a server relies on other parts/systems to work, you can end up in a situation where those parts fail, but the server doesn't. This means it is hard to have a "barrier" in the program after which something is safe to start, since that assumption might fail later on when the program is running. It's yet another argument why Alex' suggestion of having a health check is beneficial. You can fail and report why you don't provide service right now (I need access to a database, but it's timing out, etc...). Systems which can analyze their own state and report back on known errors are far easier to maintain. -- You received this message because you are subscribed to the Google Groups "golang-nuts" group. To unsubscribe from this group and stop receiving emails from it, send an email to golang-nuts+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/golang-nuts/CAGrdgiWApvJv3RkRR%3DMWxE7-T-Woq6FY0qq%3Dd9TsGSbfec4H2A%40mail.gmail.com.
