RE: V5R4 ftp anomaly

["Joe Pluta" <joepluta@xxxxxxxxxxxxxxxxx>]

I wouldn't have thought to softcode the 250 either, Jim, but as I read the
specification it's pretty clear that there are at least two different
successful completion messages: 250 indicates file transfer complete, 226
indicates file transfer complete AND the data connection has been closed.
By inference, this means that a return code of 250 means that the data
connection is still open.  Whether this would cause a different action on
the part of the client is moot; the fact is that client code should have
been written to support either the 250 or the 226 return code.

In fact, reading the specs carefully, it seems that pretty much anything
starting with a "2" SHOULD be treated as a successful completion.  They try
to pack a lot of meaning into each digit of the reply string, but the state
diagrams pretty much all run off the first digit.

Joe

> From: Jim Franz
> 
> I thought for years IBM has been telling us it is already a RFC compliant
> FTP. Think of all the hardcoding out there. This is not something I would
> think to put in a table. I guess we need to start a list of the changes
> needed to check for to go to V5R4?