site_archiver@lists.apple.com Delivered-To: darwin-dev@lists.apple.com Hello all, -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.appl... On Jun 5, 2008, at 7:43 AM, Thorsten Kramp wrote: I've written a TPC socket filter that I'm currently using only to pass all incoming data to the user space and back without any modifications. That is, (1) the sf_data_in function copies the data from the mbufs (via mbuf_copydata) and enqueues them into the control socket, (2) the user-level process reads the data from the control socket and simply writes them back, (3) the ctl_send callback finally injects the data packet in new mbufs back into the TCP stack. This seems to work fine most of the time but fails with some SSL (https) connections (even though it works fine with others). Can anyone imagine why? Any help would be highly appreciated -- thanks a lot in advance! My guess would be a botched packet checksum or a difference in the option bits. The easiest way to find out for sure would be to compare the data you got back with the data you sent up, including all portions of the packet. This email sent to site_archiver@lists.apple.com