Talk:OpenSSH/Cookbook/Multiplexing

Thank you for this guide larsnooden,

But... could you be so nice and include "the syntax of the command" that generates the contents of the frames or "the full path to the file" with the frame contents.

Thanks again,

Jringoot (discuss • contribs) 13:37, 25 February 2016 (UTC)
 * The examples work but I've added a few words of clarification. Do they address your question adequately?  Larsnooden (discuss • contribs) 13:56, 25 February 2016 (UTC)

I'll give an example: "table 1" appears to be grepped/annotated netstat output, but it is not clear, you don't mention the netstat command anywhere.

Jringoot (discuss • contribs) 11:09, 26 February 2016 (UTC)


 * Table 1 and Table 2 are annotated netstat output. The unedited output can be grabbed from this:    Should it be mentioned that it is from netstat?  Or should give the exact netstat|awk string and then mention that it was edited?
 * The part about ls needs clarification, I see now. I ran it like 'time ssh -p 22 server.example.org ls' versus 'time ssh -p 22 -S ./somesocket server.example.org ls', with the latter being prepped by having a master session in place.  Part of the reason for the large difference is that the machine acting as a server has a very weak 32-bit processor, specifically a 500MHz AMD Geode so the new connections will be slow. Larsnooden (discuss • contribs) 11:32, 26 February 2016 (UTC)

IMHO It would be a good idea to be completely open about it: include the commands you used.

To improve the learning experience it is good for the reader to try out commands him/herself and compare the output with what is in the manual and also try to find out what is the cause of the differences, if there are any.

Therefore I think it is better not to manually edit output, but rather use tools like sed, awk, troff/groff or others, if the standard output is not suitable for documentation.

Jringoot (discuss • contribs) 08:20, 1 March 2016 (UTC)


 * I understand. I've added pointers to the basic commands.  Those that know can fill in the rest.  I had not included them initially because I do not want to distract from the topic by digressing into shell work and I (mistakenly) thought that they went without saying.  Maybe if the wiki can handle footnotes or sidebars I could include a recipe of methods at the end of the chapter.  That said, Table 1 and Table 2 cannot show anything meaningful unless selected output is aggregated and edited for clarity.  Larsnooden (discuss • contribs) 08:32, 1 March 2016 (UTC)

Larsnooden (discuss • contribs) 10:07, 1 March 2016 (UTC)


 * , I've added a subsection on observing multiplexing which explains how the raw tables and figures can be generated. Larsnooden (discuss • contribs) 08:05, 2 March 2016 (UTC)