|
From: Philip L. <ph...@ov...> - 2018-10-27 22:32:10
|
On Wed, 17 Oct 2018 10:04:09 +0200 Andrea Vai <and...@un...> wrote: > Hi Phil, > Il giorno mar, 16/10/2018 alle 22.04 -0700, Philip Langdale ha > scritto: > > > > I actually did reply. > Uh, I didn't remember and have overlooked the thread messages sorting, > sorry for that. > > > I said: > > > > > Marcus is likely right. But I'm not sure how we'd go about > > > trying > > to > > > address this. As I understand it, we don't know how many bytes are > > > going to be transferred for a getobjecthandles call so we can't > > detect > > > if we have an exactly 500 byte transfer or not. Would we need to > > only > > > transfer a max of 499 bytes at a time, and do that for every > > single > > > packet? Marcus is the expert here :-) > > > > If Marcus is right, then the problem is that we don't know when > > we'll > > encounter the bug and we can't control anything except the > > individual > > packet size - and we don't know when to vary that. It seems a hard > > problem to solve. > > I understand. Also, I could open an issue in the bug tracker if it's a > better way to have it focused on. You can open an issue if you want, but I don't really understand how to fix it in a reasonable way, based on what we know. Did you try testing with more than the special number of files in the directory? Does that work or does it just keep failing once that number is reached and exceeded? if the diagnosis is correct, it should succeed with more files. --phil |