Date: Mon, 05 Nov 2007 12:48:40 +1300 From: Russell Fulton <r.fulton@auckland.ac.nz> To: john.w.court@nokia.com Cc: freebsd-ipfw@freebsd.org, gbell72@rogers.com Subject: Re: IPFW Problem Message-ID: <472E5A58.5090707@auckland.ac.nz> In-Reply-To: <DBA4167E9E1EB44D8476A6F928BE52452B5379@siebe101.NOE.Nokia.com> References: <932971.53959.qm@web88014.mail.re2.yahoo.com> <DBA4167E9E1EB44D8476A6F928BE52452B5379@siebe101.NOE.Nokia.com>
next in thread | previous in thread | raw e-mail | index | archive | help
john.w.court@nokia.com wrote: > Hmm, I may well be missing something very obvious but rule 01000 seems > to be doing exactly what it says it will. Are you sure you meant "deny" > rather than "allow" on rule 01000 ? Note that it is immediately after the check state rule. What the Gardner intended was to drop established tcp traffic that was not part of a session for which there was already state. In fact this rule is redundant since (assuming I've read the rule set correctly) such traffic will get caught by the final deny rule. What is odd about this problem is that it appears to be a timeout problem and thus probably not related to the firewall at all. To me it seems that the initial SYN packet is getting lost and the retry gets through, hence the delay. I suggested to Gardner that he log all dropped packets so he can see if it really is the firewall which is causing the problem. Russell
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?472E5A58.5090707>