I am having problems with xntp on a single processor box (AMD 350 ) but
am basiacally setting up the machine as a prototype networking setup so
I will know what I need to do to get a compaq 300 dual p-333
w/ 1.5 GB ram working onc I get the smartstart disk and rompaqs
installed. I bought two of these machines at an auction and am looking
forward to seeing what Linux can do on them.
I have some concerns about xntp since I have been lurking the list and
have seen several indicators that people are having problems with SMP
boxes and xntp.
My question is this I am having problems with xntp3 on a single proc
box. see below for a copy of a post I posted to the redhat-list and
linux-net with zero responses.
Is this a known problem with 2.2 kernels? is there a work around?
ANY insight will be appreciated. If I can't get it working on a single
proc box what the hell is going to happen when I try the dual?
Thanks for your time and hopefully this is not too far off topic to get
a response or two.
Subject:
Does anybody really know what time it is?
Date:
Wed, 20 Oct 1999 21:14:12 -0500
From:
Bret Hughes <[EMAIL PROTECTED]>
To:
linux-net <[EMAIL PROTECTED]>
Does anybody relly care?
My servers care but they don't know because I can't get xntpd to work
under rh 6.0
a search on redhat.com/support for ntp xntp yields zippo.
Deos anyone have xntpd working on 2.2 kernels?
I have installed xntp3-5.93-12 from the binary rpm and am getting these
messages:
Oct 18 07:35:52 gateway1 xntpd[403]: synchronized to 165.91.52.110,
stratum=3
Oct 18 07:35:56 gateway1 xntpd[403]: time reset (step) 4.362638 s
Oct 18 07:35:56 gateway1 xntpd[403]: synchronisation lost
Oct 18 07:41:16 gateway1 xntpd[403]: synchronized to 165.91.52.110,
stratum=3
Oct 18 07:42:20 gateway1 xntpd[403]: kernel pll status change 89
Oct 18 07:45:32 gateway1 xntpd[403]: synchronized to 129.237.32.1,
stratum=2
Oct 18 07:45:32 gateway1 xntpd[403]: kernel pll status change 89
Oct 18 07:48:43 gateway1 xntpd[403]: synchronized to 165.91.52.110,
stratum=3
Oct 18 07:48:43 gateway1 xntpd[403]: synchronisation lost
The linux notes in the package say the pll is broken and he knows about
it . What can I do to fix it ?
It seems to be keeping the hw clockfrom being updated by the kernel. A
least accoring to the docs for hwclock if the status reported by
adjtimex is <> 0 then the hwclock will not be updated on the 11 minute
cylcle. I don't know how big a deal this is but xntp is definetly
giving up on keeping the time.
a search on freshmeat for xntp shows the package with a comment that the
new versions are called ntp. rpmfind finds a linux/contrib package for
ntp with a 4.x version. will this work? what is the deal with the pll
anyway?
I have been away from the list for a while and appologize if this has
been covered recently.
BTW the output of adjtimex --print :
[root@gateway1 SRPMS]# adjtimex --print
mode: 0
offset: 0
frequency: -4282375
maxerror: 6402956
esterror: 30848
status: 137
time_constant: 2
precision: 1
tolerance: 33554432
tick: 10000
raw time: 940263045s 636128us = 940263045.636128
This is running on an AMD k6 350 with 32MB mem
Let me know if anyone has any ideas
TIA
Bret Hughes
-
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
