On Monday 09 August 2010 10:03 PM, Raj Mathur (राज माथुर) wrote: > On Monday 09 Aug 2010, Rony wrote: > >> On Sunday 08 August 2010 10:31 PM, Raj Mathur (राज माथà¥à¤°) >> wrote: >> >>> On Sunday 08 Aug 2010, jtd wrote: >>> >>>> >>>> Essentially control tells aa to send data to nn. >>>> which is the same as ssh to aa, fire script (which transfers data >>>> to nn) and get out leaving script running. >>>> >>>> So all data flows only from aa to nn. >>>> >>> Cool, just validated that it works that way. Didn't know this >>> earlier, useful piece of information to have. >>> >> Does the flow of data bypass the control machine? Are packets >> directly going from aa to nn? >> > Yes, that's what I confirmed by transferring a large file two remote > servers and watching my traffic. > >
Before you did this experiment, has aa ever established an ssh connection with nn or vice versa? Suppose aa and nn have never communicated with each other, if aa is giving a key to nn, how does nn say 'yes' to accept it, through a third party control machine? I am not doubting your experiment but only clearing my own understanding of ssh so no offense intended. Could you try the same experiment on virgin machines? -- As a proper list etiquette... Please trim your replies. Post your replies below the relevant original text, leaving a line space. Do not re-use old messages to write new ones. Regards, Rony. -- http://mm.glug-bom.org/mailman/listinfo/linuxers
