Hi, I have another idea (and I bet you thought of this too) :-)
Since the reconnect must login again, with a user and password, and it looks like you know all the files because you build the script with file specific names, why not fail completely on the transfer error, then determine which file transfers worked, and rebuild the script minus the good transfers, and then restart. Bill On Fri, Jul 4, 2008 at 4:10 PM, Aethon <[EMAIL PROTECTED]> wrote: > On Fri, Jul 4, 2008 at 2:05 PM, Bill Dorsey <[EMAIL PROTECTED]> wrote: >> Hi, >> >> I have a suggestion. >> >> I think you could handle resolve this by taking a different approach >> to the script. >> >> Use perl to loop thru the files one at a time, in the loop transfer >> the file, check thre return code, look for a returned error, then >> based on the return, do stuff, then start over on the next file. >> >> Bill > > This was my first approach. The server is set up so that if there are > too many connections within a certain time window the IP gets > blacklisted for a few hours. The files I'm uploading range from about > 50 kB to a couple mB, with a few thousand files per script run. > Unfortunately this combined makes a connection per file approach > infeasible. > > Thanks for the suggestion though, > > Aethon >
