If it's imperative they're received in order, you cannot do anything accept wait for at least the first byte of the response. You'll never know if the request was fully received until that point. Even then, only that it was _received_, not validated (e.d. 400 response).
If it is _not_ imperative and you only want to "minimize" the occurrences. You can probably try to use a dynamic timeout value rather than a fixed value. For example, on every request, meausure the time to the first byte of the response. Send the second request when that first byte is received. If the first response took 200ms, send the third request at 100ms (half of the round-trip) after the second. If the first byte of the second response is received in only 100ms, try adjusting the timeout to 50ms. Repeat this process ad infinitum. If you really want to get fancy, you could use a beyesian formula to take into account more than just the most recent response time. Finally, you may be able to get away with less than just .5 * TTFB. However, the lower you make the timeout, the more likely it will be that you send requests out of order. -- 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. For more options, visit https://groups.google.com/d/optout.
