ntpq sysstat

[Hal Murray]

I'm working on #709, so I've got the sysstat stuff mostly reloaded into my 
cache.

ntpq/sysstat in the current release is buggy.

ntpq> syss
uptime:                 48654
control requests:       43
unexpected vc type 11 for ss_reset, value 1854
packets received:       54
current version:        53
....

It looks like the bug was introduced when you reverted the recent/total 2 
column work.

in do_sysstats
            ("ss_reset", "sysstats reset:       ", NTP_UPTIME),
collect_display doesn't handle NTP_UPTIME
collect_display2 does

I think the fix is either to change that NTP_UPTIME to NTP_INT
or to restore the 2 column mode.

I thought the 2 column mode was working OK.  Do you remember why you reverted 
it?  Note the dates.  It looks like maybe an old merge request that should 
have been dropped got merged instead.

$ git log ntpclients/ntpq.py
commit ccd8756163e1de20ca0b566ac4eb4ded34a3ad37
Author: James Browning <jamesb.fe80 at gmail.com>
Date:   Thu May 27 14:16:24 2021 -0700

    ntpq: functionally partially revert 66ed781a -- `ntpq: Display dual column 
protocol stats ...`

commit cce9e9a3c426a4e3b47b05f55c2ab7b25c10d6b8
Author: James Browning <jamesb.fe80 at gmail.com>
Date:   Sun Feb 7 08:47:11 2021 -0800

    ntpq: -c rv should return associd only when variable(s) not passed.

commit 66ed781a9e5d2df2664368ecb9bf193eef24d803
Author: James Browning <jamesb.fe80 at gmail.com>
Date:   Wed Jun 24 13:56:27 2020 -0700

    ntpq: Display dual column protocol stats ...
    
    This excludes uptime and control requests.
    The control request should be part of control stats (unimplemented)
    Uptime does not relate to when stats were reset.


-- 
These are my opinions.  I hate spam.