> Why fight so hard to run 6 year old software on the public Internet?

I can't speak for anyone else.  But, as someone running NTP even older
than that, my response is, because the expected time investment
required to switch versions is fairly high and the perceived benefit is
quite low.

I run the NTP daemon shipped with the OS version I run.  (Which one
this is varies depending on the machine; there are three OS revs I
run.)  So far, I have seen no reason to think I am vulnerable to any of
the various issues running around, though I did break monlist because
my upstream didn't want to have to special-case me.  (I worked fine as
a bandwidth amplifier for tiny amounts of bandwidth; try to push more
than tiny amounts, though, and you get router-blocked at my border.  My
upstream apparently had something set up that assumed being willing to
bandwidth-amplify for a few packets meant being willing to
bandwidth-amplify for arbitrary levels of traffic.)  And all the other
vulnerabilities I've seen reported have been in facilities I don't use,
mostly crypto stuff.

Until and unless I see reason to think I'm vulnerable to to something
worse than the above, I see no reason to go through the hell of
convincing a more recent ntpd to build on my systems - especially since
(a) the documentation is in such an inconvenient format and (b) it has
drunk the ./configure koolaid, each of which substantially increases
the time and effort involved.

So, I guess, my answer is to ask in return, why fight so hard to create
an unfunded mandate for others to heatseek ntpd versions?

/~\ The ASCII                             Mouse
\ / Ribbon Campaign
 X  Against HTML                [email protected]
/ \ Email!           7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B
_______________________________________________
pool mailing list
[email protected]
http://lists.ntp.org/listinfo/pool

Reply via email to