|
From: Oswald B. <osw...@gm...> - 2022-07-01 12:24:59
|
On Fri, Jul 01, 2022 at 01:03:11PM +0200, Yuri D'Elia wrote: >A good distinction could really be if retrying makes sense or not. > yes, but the followup questions are then: retry after how much time, and how often? it's hard to know that if you don't know *exactly* what went wrong. otoh, network errors are always somewhat unreliable, e.g. you might get NXDOMAIN (which is technically a terminal error) just because the network is down and your home router is too stupid to report it correctly (true story). in short, it's a mess. >mbsync can currently get stuck during slow/blocked transfers in >multiple places, most commonly during resolve or during connect. > resolve is indeed synchronous and its timeout outside isync's control. i need to do something about that ... but if you find tangible evidence for isync's Timeout not working in some other scenario, i'm all ears. i suppose the most promising way to debug this would be running with -D (maybe -Dn is sufficient) and saving away the logs where the process got killed by timeout. |