[Nagiosplug-devel] [ nagiosplug-Bugs-773588 ] check_ntp vs Cisco & Solaris NTP responses
SourceForge.net
noreply at sourceforge.net
Sun Aug 10 08:32:03 CEST 2003
Bugs item #773588, was opened at 2003-07-18 08:05
Message generated for change (Comment added) made by sghosh
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=397597&aid=773588&group_id=29880
Category: Parsing problem
Group: v1.3.0 beta3
Status: Open
Resolution: None
Priority: 5
Submitted By: Brad Meier (knightorc)
Assigned to: Subhendu Ghosh (sghosh)
Summary: check_ntp vs Cisco & Solaris NTP responses
Initial Comment:
Found some Solaris machines that don't match against
line 263's parsing of the reply from ntpq, they use a
#, not * or o.
Changed it to (\*|o|\#) instead of (\*|o) in the regex
and its happy again, checked against Tardis on windows
and xntpd on Linux and Solaris.
Found another problem, tried against a cisco and it
returned a - where the script expects l,u,m,b. Line
263 again.
Changed ([lumb]+) to ([lumb-]+)
Patch tested against Linux, Solaris, Windows (Tardis)
and Cisco ntp's.
----------------------------------------------------------------------
>Comment By: Subhendu Ghosh (sghosh)
Date: 2003-08-10 11:31
Message:
Logged In: YES
user_id=46572
I am looking at CVS HEAD and it looks like you are referring
to line 272 - please verify.
If adding # in the regex, then the overall ntpq loop needs
to have memory because # implies "selected" but not among
the top 6 peers selected for synchronization in ntpd from
ntp.org.
According to Solaris docs - # implies selected for
synchronization but distance exceeds maximum
This may be something that is not wanted under normal operation.
If we want to match against # then it should only match if *
or o are not present and generate a warning about maximum
distance exceeded.
Are there any links to Tardis docs about the use of tally codes?
-sg
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=397597&aid=773588&group_id=29880
More information about the Devel
mailing list