* pasta: set correct flags in reset message when outgoing connection setup fails
@ 2025-01-25 20:16 Jon Maloy
2025-01-25 22:24 ` Stefano Brivio
0 siblings, 1 reply; 2+ messages in thread
From: Jon Maloy @ 2025-01-25 20:16 UTC (permalink / raw)
To: passt-dev, sbrivio, lvivier, dgibson, jmaloy
Bug #93 states that a connect() will hang if the remote port of a
connection setup doesn't exist. The remote host will respond with
the (RST | SYN | ACK) flags set, but we only relay the RST flag back
to the local user. This causes the situation described above.
We now add a check if the reset happens in the connection state
TAP_SYN_RCVD, in which case we add the missing (SYN | ACK) flags.
This solves the problem.
Signed-off-by: Jon Maloy <jmaloy@redhat.com>
---
tcp.c | 5 +++++
1 file changed, 5 insertions(+)
diff --git a/tcp.c b/tcp.c
index c89f323..46df6bd 100644
--- a/tcp.c
+++ b/tcp.c
@@ -1199,9 +1199,14 @@ static int tcp_send_flag(const struct ctx *c, struct tcp_tap_conn *conn,
*/
void tcp_rst_do(const struct ctx *c, struct tcp_tap_conn *conn)
{
+ int flags = RST;
+
if (conn->events == CLOSED)
return;
+ if (conn->events == TAP_SYN_RCVD)
+ flags |= SYN | ACK;
+
if (!tcp_send_flag(c, conn, RST))
conn_event(c, conn, CLOSED);
}
--
@@ -1199,9 +1199,14 @@ static int tcp_send_flag(const struct ctx *c, struct tcp_tap_conn *conn,
*/
void tcp_rst_do(const struct ctx *c, struct tcp_tap_conn *conn)
{
+ int flags = RST;
+
if (conn->events == CLOSED)
return;
+ if (conn->events == TAP_SYN_RCVD)
+ flags |= SYN | ACK;
+
if (!tcp_send_flag(c, conn, RST))
conn_event(c, conn, CLOSED);
}
--
2.48.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: pasta: set correct flags in reset message when outgoing connection setup fails
2025-01-25 20:16 pasta: set correct flags in reset message when outgoing connection setup fails Jon Maloy
@ 2025-01-25 22:24 ` Stefano Brivio
0 siblings, 0 replies; 2+ messages in thread
From: Stefano Brivio @ 2025-01-25 22:24 UTC (permalink / raw)
To: Jon Maloy; +Cc: passt-dev, lvivier, dgibson
On Sat, 25 Jan 2025 15:16:47 -0500
Jon Maloy <jmaloy@redhat.com> wrote:
> Bug #93 states that a connect() will hang if the remote port of a
> connection setup doesn't exist. The remote host will respond with
> the (RST | SYN | ACK) flags set, but we only relay the RST flag back
> to the local user. This causes the situation described above.
Thanks for looking into this.
> We now add a check if the reset happens in the connection state
> TAP_SYN_RCVD, in which case we add the missing (SYN | ACK) flags.
>
> This solves the problem.
I guess that, actually, this part of the problem (or the whole bug #93?
I'm not sure) was solved by commit db2c91ae86c7 ("tcp: Set ACK flag on
*all* RST segments, even for client in SYN-SENT state").
> Signed-off-by: Jon Maloy <jmaloy@redhat.com>
> ---
> tcp.c | 5 +++++
> 1 file changed, 5 insertions(+)
>
> diff --git a/tcp.c b/tcp.c
> index c89f323..46df6bd 100644
> --- a/tcp.c
> +++ b/tcp.c
> @@ -1199,9 +1199,14 @@ static int tcp_send_flag(const struct ctx *c, struct tcp_tap_conn *conn,
> */
> void tcp_rst_do(const struct ctx *c, struct tcp_tap_conn *conn)
> {
> + int flags = RST;
> +
> if (conn->events == CLOSED)
> return;
>
> + if (conn->events == TAP_SYN_RCVD)
> + flags |= SYN | ACK;
> +
...flags is not used, though. Could it be that you tested this after
rebasing on top of db2c91ae86c7?
> if (!tcp_send_flag(c, conn, RST))
> conn_event(c, conn, CLOSED);
> }
--
Stefano
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-01-25 22:24 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2025-01-25 20:16 pasta: set correct flags in reset message when outgoing connection setup fails Jon Maloy
2025-01-25 22:24 ` Stefano Brivio
Code repositories for project(s) associated with this public inbox
https://passt.top/passt
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for IMAP folder(s).