On Thu, 2005-06-16 at 01:23 +0300, Ville Skyttä wrote: > On Wed, 2005-06-15 at 10:45 -0400, Joe Schaefer wrote: > > > Let me ask you guys my original question in another way: > > while monitoring the process, would you say that the process size > > grows gradually during the actual network IO, or does it jump > > suddenly right after the network IO completes? > > I haven't been able to test the latest patch nor can I test the above > right now (but will try tomorrow). But if memory serves me right, the > ballooning happened definitely after, not during the network I/O.
Memory did serve me right, and the answer to the WAG #2 patch is unfortunately nay here too.
