At Thu, 19 Apr 2007 23:36:38 -0500, Craig Boston wrote: > > On Fri, Apr 20, 2007 at 08:37:59AM +1000, Peter Jeremy wrote: > > You could try reducing net.inet.tcp.recvspace or disabling > > net.inet.tcp.rfc1323 and see if that helps. (Though RELENG5 should > > also be affected if this is the problem). > > I don't know if it's related, but I've had problems with packets to/from > a recent -CURRENT box not being able to traverse a pf (running on > 6.2-stable) firewall with NAT and state rules. Some of the packets are > dropped unless I disable net.inet.tcp.rfc1323 on the system running > -CURRENT, enough to cause ssh connections to stall and drop before > login. > > 6.x machines behind the same firewall work just fine. I can provide > tcpdump output if someone thinks it will help. Maybe this helps. revision 1.291 date: 2006/02/28 23:05:59; author: andre; state: Exp; lines: +20 -19 Rework TCP window scaling (RFC1323) to properly scale the send window right from the beginning and partly clean up the differences in handling between SYN_SENT and SYN_RCVD (syncache). Further changes to this code to come. This is a first incremental step to a general overhaul and streamlining of the TCP code. PR: kern/15095 PR: kern/92690 (partly) Reviewed by: qingli (and tested with ANVL) Sponsored by: TCP/IP Optimization Fundraise 2005 -- Rui Paulo - rpaulo_at_fnop.netReceived on Fri Apr 20 2007 - 08:18:31 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:39:09 UTC