Re: HAProxy and proxy protocol support
Nathan Phillip Brink <[email protected]> Thu, 13 Jun 2024 00:16:46 -0400
| Newsgroups | gmane.mail.imap.courier.general |
|---|---|
| Message-ID | <[email protected]> |
On 2024-06-12 23:35, Sam Varshavchik wrote: > Andrew Athan writes: > >> « HTML content follows >> >> »Sam: >> >> >> I've googled but not been able to find a reference to this >> duplication of initial bytes issue in TFO. Can you provide more >> information? > > This is explicitly called out in RFC 7413 section 6.1. > > I also found a layman's description at: > > https://netty.io/wiki/tcp-fast-open.html > > "TFO cannot always be used, however, because it changes how TCP > behave: The receiving end might see this data duplicated due to > retransmission of the SYN packet. For this reason, TFO must only be > used when the data in the initial packet will be processed > idempotently. Whether this is the case, depends on the protocol and > the application." > >> That seems like a catastrophic bug in the TCP contract. I'd like to >> make some colleagues aware of the issue, ideally with a reference to >> some analysis of it. > > This seems like a well known, known issue with this. Doesn’t this just mean that a single connection could be seen by the server as two connections? As long as the initial data isn’t enough to actually enqueue an email, one of those ghost connections should just timeout and the other real connection actually process. Therefore, as long as SMTP itself has a handshake which the client respects, such as the client sending waiting for and validating the response of either EHLO, MAIL, or RCPT before sending DATA, then that client is protected from this issue. Of course, any client which tries to optimize performance by sending blindly without waiting for a server response might end up double-queuing an email, which is a valid concern. I wonder if it could be argued that such a client is non-compliant. I also apologize for any misunderstanding of SMTP as I am not very familiar with it. I would like some confirmation if I am interpreting the concern correctly. I do not think it was mentioned explicitly (unless I missed it). Is the concern the possibility of double-queuing a single email as I purported above? _______________________________________________ courier-users mailing list [email protected] Unsubscribe: https://lists.sourceforge.net/lists/listinfo/courier-users