On 10/13/2008 05:06 PM, Lars Marowsky-Bree wrote: > On 2008-10-13T17:02:36, Florian Haas <[EMAIL PROTECTED]> wrote: > >>> - cutter >>> >>> I think cutter might even work, as the replicated nodes do think that >>> they are "forwarding" the connection. We just need to identify all >>> connections which passed through the failed node. >> Cutter in its current incarnation does not work, because of an overly >> paranoid (?) sanity check. Extract cutter.c from >> http://www.lowth.com/cutter/software/cutter-1.03.tgz and take a look at >> lines 521 through 546. :-) > > I meant in principle. That there are going to be bugs which need fixing > is obvious ;-)
I tried contacting the original author of cutter as to the rationale behind this sanity check a few days ago, but to no avail. Does someone more knowledgeable in TCP connection tracking than myself want to take a look into this and see if it can be safely bypassed? Having something like "cutter --allow-local <ipaddress>" would be really helpful. Thoughts? Cheers, Florian _______________________________________________________ Linux-HA-Dev: [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha-dev Home Page: http://linux-ha.org/
