on the NTP security issues and fixes

[Gary E. Miller]

Yo All!

I pulled git head, running it now on a server in place of chronyd.

Seems to work OK.  I'll keep an eye on it.

A couple things I note right away, consider them non-critical feature
requests.

I really like the chronyd socket interface over the SHM one.  The user is
not playing with magic numbers.

'ntpq -p' is user hostile compared to 'chronyc sources'.  chronyc adds
units to the display, so you do not have to keep referring to them
manual, and it makes it easy to deal with jitter and delay that
varies by orders of magnitude.

So, instead of this:

     remote           refid      st t when poll reach   delay   offset  jitter
==============================================================================
-india.colorado. .NIST.           1 u   33   64  377   79.023  -11.071  10.247
 140.142.16.34   .INIT.          16 u    -  256    0    0.000    0.000   0.000
+clepsydra.labs. .GPS.            1 u   28   64  377   26.473   15.307   4.261
+catbert.rellim. .GPS1.           1 u   31   64  377    0.200   22.061   6.440
-SHM(0)          .GPS.            0 l   45   64  377    0.000  151.500  15.858
*SHM(1)          .GPS1.           0 l   44   64  377    0.000   20.870   7.680
+131.107.13.100  .ACTS.           1 u   25   64  337   14.746   18.469   6.634

I see this:

dagwood ~ # chronyc sources
210 Number of sources = 11
MS Name/IP address         Stratum Poll Reach LastRx Last sample
===============================================================================
#- SHM0                          0   4   377    19   +119ms[ +119ms] +/-  501ms
#- SHM1                          0   4   377    21   +276us[ +276us] +/-   13us
#* SOC2                          0   4   377    22   +260us[ +291us] +/-   24us
#? SOC3                          0   4     0  247d  +7275ms[  +99us] +/-  230ms
^x catbert.rellim.com            1   8   377     0  -1566us[-1566us] +/- 1564us
^- spidey.rellim.com             1   6   377    54    +18ms[  +18ms] +/-   27ms
^- kong.rellim.com               2  10   377   177   -722us[ -735us] +/- 2004us
^- ha2.smatwebdesign.com         2  10   177   492  -2890us[-2905us] +/-  103ms
^- soft-sea-01.servers.octos     2  10   377   775  -5897us[-5843us] +/-   83ms
^- gw-kila-wl0.ipv6.avioc.or     2  10   377   17m   -519us[ -640us] +/-   64ms
^- 66.241.101.63                 2  10   377   449  +7966us[+7973us] +/-   52ms

Neither is ideal, easy pickings for someone to code on.

And last, but not least, ntpd takes way, way, way longer to converge
than chronyd.  Which is why on the fly reconfiguation in ntpd is SO
important.  Last thing you ever want to do is restart ntpd.

Right now, after 10 mins, ntpd has 2,000 times the jitter as chronyd had
when I turned it off.

RGDS
GARY
---------------------------------------------------------------------------
Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
	gem at rellim.com  Tel:+1 541 382 8588
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 473 bytes
Desc: OpenPGP digital signature
URL: <http://lists.ntpsec.org/pipermail/devel/attachments/20160504/f5156d62/attachment.bin>