I'm the maintainer ... not my decision to drop ntpdate True, as maintainer, you are the distro's delegate for watching this package. As such, you're in position to say "we cannot yet drop ntpdate because the functionality of sntp isn't satisfactory". I guess that by now you say "hey, but it is..."; i say no, hope you have
https://bugzilla.novell.com/show_bug.cgi?id=461205 User Ori_Tzoran@bmc.com added comment https://bugzilla.novell.com/show_bug.cgi?id=461205#c5 Ori Tzoran <Ori_Tzoran@bmc.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|CLOSED |REOPENED Platform|x86-64 |All Resolution|INVALID | --- Comment #5 from Ori Tzoran <Ori_Tzoran@bmc.com> 2009-03-12 03:46:53 MST --- hello Peter, i apologize in advance for insisting. pls refer to sntp vs ntpdate output on the bottom. patience to listen. Timekeeping is a fundamental service in production environments and NTP is its natural provider. But NTP, as a project, set of utilities and doc collection is still beyond the reach of many, simply because it's doc is rotten [see bug 483897]. Here's another example from sntp manpage for the -q option: <i>indicates that it should query a daemon save file being maintained by it.</i> i read and reread this phrase. is the term "save file" explained somewhere? no. let alone the repeated 'it'. Even worse, the explanation of sntp's output isn't clear at all. Look at the paragraph starting with "The default is to write the estimated...". Example output '+ 4.567 +/- 0.089 secs' ...". Nowhere does it explain what's the ACTUAL offset! what are those 2 figures? One might guess that offset follows the '+' [e.g. 4.567] and the figure after '+/-' is deviation? [0.089]. But if you look at the outputs below, you'll see that the "offset" on my systems is smaller than this supposed deviation. ntpdate, ntpd and ntpq expressed offsets and delays in msec. sntp breaks from this. why? Last, but not least. i had a problem with timkeeping on vmware. ntpdate was the main tool used to investigate and monitor this issue [that's the reason it was my first finding after installing 11.1]. compare outputs below for the target 'thi-dc-01'; it works flawlessly with ntpdate, sntp fails with a strange and unexpected output. This is unacceptable for a utility you use in a script, where the output is redirected for further processing. = Output = == 11.0 ntpdate == [10:54:06]otzoran@thilnx027[~] $ cat /etc/SuSE-release openSUSE 11.0 (X86-64) VERSION = 11.0 [10:54:15]otzoran@thilnx027[~] $ ntpdate -q 0.vmware.pool.ntp.org server 212.235.102.46, stratum 2, offset -0.000636, delay 0.03609 12 Mar 10:54:26 ntpdate[31739]: adjust time server 212.235.102.46 offset -0.000636 sec [10:54:26]otzoran@thilnx027[~] $ ntpdate -q thi-dc-01 server 137.72.217.10, stratum 3, offset -0.044199, delay 0.04146 12 Mar 10:54:37 ntpdate[31743]: adjust time server 137.72.217.10 offset -0.044199 sec == 11.1 sntp == [10:57:45]otzoran@thi-esx-0360[~] $ cat /etc/SuSE-release openSUSE 11.1 (i586) VERSION = 11.1 [10:57:53]otzoran@thi-esx-0360[~] $ /usr/sbin/sntp 0.vmware.pool.ntp.org 2009 Mar 12 10:58:03.000361 - 0.003831 +/- 0.063270 secs [10:58:03]otzoran@thi-esx-0360[~] $ /usr/sbin/sntp thi-dc-01 sntp: unable to format current local time sntp: Success -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.