From: Patrick McHardy <kaber@trash.net>
To: dccp@vger.kernel.org
Subject: Re: DCCP conntrack/NAT
Date: Tue, 08 Apr 2008 10:33:35 +0000 [thread overview]
Message-ID: <47FB49FF.3000806@trash.net> (raw)
In-Reply-To: <47F64C0D.5080700@trash.net>
[-- Attachment #1: Type: text/plain, Size: 255 bytes --]
Patrick McHardy wrote:
> The attached patch contains the changes I've made so far
> based on your review. I'll go through the remaining points
> now.
>
New version attached that doesn't only update the comments
but also the actual state transitions :)
[-- Attachment #2: x --]
[-- Type: text/plain, Size: 3069 bytes --]
diff --git a/net/netfilter/nf_conntrack_proto_dccp.c b/net/netfilter/nf_conntrack_proto_dccp.c
index 44c8aa6..e17bd4f 100644
--- a/net/netfilter/nf_conntrack_proto_dccp.c
+++ b/net/netfilter/nf_conntrack_proto_dccp.c
@@ -70,7 +70,7 @@ static unsigned int dccp_timeout[CT_DCCP_MAX + 1] __read_mostly = {
[CT_DCCP_REQUEST] = 2 * DCCP_MSL,
[CT_DCCP_RESPOND] = 4 * DCCP_MSL,
[CT_DCCP_PARTOPEN] = 4 * DCCP_MSL,
- [CT_DCCP_OPEN] = 5 * 86400 * HZ,
+ [CT_DCCP_OPEN] = 12 * 3600 * HZ,
[CT_DCCP_CLOSEREQ] = 64 * HZ,
[CT_DCCP_CLOSING] = 64 * HZ,
[CT_DCCP_TIMEWAIT] = 2 * DCCP_MSL,
@@ -142,12 +142,12 @@ dccp_state_table[IP_CT_DIR_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
* got lost after we saw it) or reincarnation
* sPO -> sIG Request during PARTOPEN state, server will ignore it
* sOP -> sIG Request during OPEN state: server will ignore it
- * sCR -> sIG MUST respond with Close to CloseReq (8.3.)
+ * sCR -> sIG FIXME MUST respond with Close to CloseReq (8.3.)
* sCG -> sIG
- * sTW -> sIG Time-wait
+ * sTW -> sRQ Reincarnation
*
* sNO, sRQ, sRS, sPO. sOP, sCR, sCG, sTW, */
- sRQ, sRQ, sRS, sIG, sIG, sIG, sIG, sIG,
+ sRQ, sRQ, sRS, sIG, sIG, sIG, sIG, sRQ,
},
[DCCP_PKT_RESPONSE] = {
/*
@@ -188,7 +188,7 @@ dccp_state_table[IP_CT_DIR_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
/*
* sNO -> sIV No connection
* sRQ -> sIV No connection
- * sRS -> sIV No connection
+ * sRS -> sPO Ack for Response, move to PARTOPEN (8.1.5.)
* sPO -> sPO Remain in PARTOPEN state
* sOP -> sOP Regular DataAck packet in OPEN state
* sCR -> sCR DataAck in CLOSEREQ MAY be processed (8.3.)
@@ -196,7 +196,7 @@ dccp_state_table[IP_CT_DIR_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
* sTW -> sIV
*
* sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
- sIV, sIV, sIV, sPO, sOP, sCR, sCG, sIV
+ sIV, sIV, sPO, sPO, sOP, sCR, sCG, sIV
},
[DCCP_PKT_CLOSEREQ] = {
/*
@@ -320,11 +320,11 @@ dccp_state_table[IP_CT_DIR_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
* sPO -> sOP -> sCR Move directly to CLOSEREQ (8.1.5.)
* sOP -> sCR CloseReq in OPEN state
* sCR -> sCR Retransmit
- * sCG -> sIV Already closing
+ * sCG -> sCR Simultaneous close, client sends another Close
* sTW -> sIV Already closed
*
* sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
- sIV, sIV, sIV, sCR, sCR, sCR, sIV, sIV
+ sIV, sIV, sIV, sCR, sCR, sCR, sCR, sIV
},
[DCCP_PKT_CLOSE] = {
/*
@@ -333,12 +333,12 @@ dccp_state_table[IP_CT_DIR_MAX + 1][DCCP_PKT_SYNCACK + 1][CT_DCCP_MAX + 1] = {
* sRS -> sIV No connection
* sPO -> sOP -> sCG Move direcly to CLOSING
* sOP -> sCG Move to CLOSING
- * sCR -> sCG Waiting for close from client
+ * sCR -> sIV Close after CloseReq is invalid
* sCG -> sCG Retransmit
* sTW -> sIV Already closed
*
* sNO, sRQ, sRS, sPO, sOP, sCR, sCG, sTW */
- sIV, sIV, sIV, sCG, sCG, sCG, sCG, sIV
+ sIV, sIV, sIV, sCG, sCG, sIV, sCG, sIV
},
[DCCP_PKT_RESET] = {
/*
next prev parent reply other threads:[~2008-04-08 10:33 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-04-04 15:41 DCCP conntrack/NAT Patrick McHardy
2008-04-04 19:59 ` Jan Engelhardt
2008-04-05 10:09 ` Pablo Neira Ayuso
2008-04-05 10:12 ` Pablo Neira Ayuso
2008-04-06 0:28 ` Patrick McHardy
2008-04-07 21:50 ` Gerrit Renker
2008-04-07 22:45 ` Patrick McHardy
2008-04-08 9:27 ` Gerrit Renker
2008-04-08 10:30 ` Patrick McHardy
2008-04-08 10:33 ` Patrick McHardy [this message]
2008-04-08 11:18 ` Patrick McHardy
2008-04-08 13:38 ` Gerrit Renker
2008-04-08 14:12 ` Patrick McHardy
2008-04-08 14:26 ` Patrick McHardy
2008-04-08 16:21 ` Patrick McHardy
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=47FB49FF.3000806@trash.net \
--to=kaber@trash.net \
--cc=dccp@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox