Amos, Thanks the follow up and for the reminder on SELinux but at this point, I have it off (I don't think I need to relabel after turning off -- any one know?). I'm at a loss too -- starting to add more debugging logic (maybe will even instrument a kernel) to see if I can figure out what's going on. (and going back to school on pbr ;) ) Will keep you posted if I figure out something. /jsb James S. Binder Vice President, Engineering Cyphort Inc., jbinder@xxxxxxxxxxx 408.761.1403 (cell) This information contained in this e-mail message and any attachments thereto, is intended only for the personal and confidential use of the recipient(s) named above. This message may be under the terms of a Mutual Non-Disclosure Agreement communication and/or work product and as such is privileged and confidential. If the reader of this message is not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this document in error and that any review, dissemination, distribution, or copying of this message is strictly prohibited. If you have received this communication in error, please notify use immediately by e-mail and delete this original message. On Mar 15, 2011, at 4:39 PM, Amos Jeffries wrote: > On Tue, 15 Mar 2011 07:41:28 -0700, Jim Binder wrote: >> If I try and add the route, both fail with file exists err >> >> [root@fw01 ~]# ip route add local 0.0.0.0/0 dev eth0 table 100 >> RTNETLINK answers: File exists >> [root@fw01 ~]# ip route add local 0.0.0.0/0 dev eth2 table 100 >> RTNETLINK answers: File exists >> > > Okay, I think that is a good sign. It matches what I see anyway. > > Re-checking the tproxy page I'm reminded about SELinux. > > http://wiki.squid-cache.org/Features/Tproxy4#SELINUX_Policy_tuning > http://wiki.squid-cache.org/Features/Tproxy4#selinux_policy_denials > > If it is not that, you have the (dis)pleasure of hitting something new. > > I find it really weird that Squid is not even getting to the accept() stage after iptables has logged them as approved. IME tracking the packets have been rejected by the FW software either arriving or leaving Squid . Not just disappearing right at the end of the TCP stack like that. > > Amos > >> >> On Mar 15, 2011, at 3:02 AM, Amos Jeffries wrote: >> >>> On 15/03/11 20:22, Jim Binder wrote: >>>> Trying this one more time to see if anyone might know what's wrong in getting my transparent bridging with squid to work. >>>> Config... pings work thought the box (the bridge is working however; the 3129 socket never pops with an HTTP request) >>>> >>>> Admin on Eth1, Internet on eth0 and Inside (client) interface on eth2. Br0 used as the bridge. >>>> >>>> Running Fedora core 14 (but went back as fare as 12 and couldn't get it to work) >>>> >>>> Squid Cache: Version 3.HEAD-20110307 >>>> configure options: '--enable-ecap' '--enable-icap-client' '--enable-linux-netfilter' --enable-ltdl-convenience >>>> >>>> iptables-1.4.9-1.fc14.i686 >>>> kernel-2.6.35.11-83.fc14.i686 >>>> ebtables-2.0.9-5.fc13.i686 >>>> >>>> Went as far to turn on dynamic debug logging and I don't see what's wrong but the connect never seems to get made to the 3129 socket. >>>> >>>> [ 214.914113] TRACE: mangle:PREROUTING:rule:2 IN=eth2 OUT= MAC=00:40:f4:cd:01:70:00:50:56:36:df:78:08:00 SRC=192.168.1.91 DST=192.168.1.88 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=3380 DF PROTO=TCP SPT=48255 DPT=80 SEQ=1363486620 ACK=0 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A02522AA80000000001030306) >>>> [ 214.914155] xt_TPROXY: redirecting: proto 6 c0a80158:80 -> 00000000:3129, mark: 1 >>>> [ 217.920783] TRACE: raw:PREROUTING:policy:3 IN=eth2 OUT= MAC=00:40:f4:cd:01:70:00:50:56:36:df:78:08:00 SRC=192.168.1.91 DST=192.168.1.88 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=3381 DF PROTO=TCP SPT=48255 DPT=80 SEQ=1363486620 ACK=0 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A025236680000000001030306) >>>> [ 217.920846] TRACE: mangle:PREROUTING:rule:2 IN=eth2 OUT= MAC=00:40:f4:cd:01:70:00:50:56:36:df:78:08:00 SRC=192.168.1.91 DST=192.168.1.88 LEN=60 TOS=0x00 PREC=0x00 TTL=64 ID=3381 DF PROTO=TCP SPT=48255 DPT=80 SEQ=1363486620 ACK=0 WINDOW=5840 RES=0x00 SYN URGP=0 OPT (020405B40402080A025236680000000001030306) >>>> [ 217.920891] xt_TPROXY: redirecting: proto 6 c0a80158:80 -> 00000000:3129, mark: 1 >>> <snip> >>>> [root@fw01 ~]# >>>> [root@fw01 ~]# ip route list table all >>>> local default dev lo table 100 scope host >>> >>> Tried with "table 100" created on eth0 and eth2 ? >>> >>> That seems to be needed recently. >>> >>> Everything else looks okay to me. Down to the packets hitting the TPROXY and DIVERT rules. >>> >>> Amos >>> -- >>> Please be using >>> Current Stable Squid 2.7.STABLE9 or 3.1.11 >>> Beta testers wanted for 3.2.0.5 >
