On Sat, Aug 22, 2020 at 09:17:56PM +0100, Ashley Dixon wrote: > On Sat, Aug 22, 2020 at 04:15:38AM +0000, Caveman Al Toraboran wrote: > > just to double check i got you right. due to > > flushing the buffer to disk, this would mean that > > mail's throughput is limited by disk i/o? [...] > When an M.T.A. encounters mail, the content of the mail will first exist in > the > M.T.A.'s local memory, in a buffer. Before sending an "OK" to the > sending > server, it should first make an attempt to write it to disk, through an > fwrite > (stdio) or write (POSIX) call. At that point, it is, in theory, the > kernel's > choice if and when it is _actually_ written to disk, but if one of > the > aforementioned functions return a success code, the M.T.A. has done its bit, > and > can consider the message "safely stored".
true and yes given a sink willing to accept your throughput, an mta is generally disk i/o bound -- Eray