[gentoo-user] System Clock Problems

[gentoo-user] System Clock Problems

[Drew Tomlinson]

I have Gentoo 2.6.13-r5 kernel running and have used ntpd in 
broadcastclient mode to keep its time in sync on my home network.  The 
other day, the system suffered and abrupt shutdown due to a power 
outage.  Ever since then, the system clock gains about 10 seconds every 
5 minutes.  Also, I can't get ntpd to sync the clock.  My command line 
is "ntpd -A -b -g -u ntp:ntp".  I've included some output running the 
command with the debug switch below.

I've also tried to gain more info with the ntpdc utility although I 
don't really know what I'm doing.  However it appears that ntpd does see 
my FreeBSD time server even though it's not synced:

ntpdc> peers
     remote           local      st poll reach  delay   offset    disp
=======================================================================
=192.168.1.2     0.0.0.0          2   64    7 0.00038 -9.600170 1.98438

The time server is a FreeBSD 6.0 box on my network.  My other FreeBSD 
box and two Windows boxes get time from it just fine.  Even the Gentoo 
box will set its clock with "ntpd -gq".  I am currently using this brute 
force method via a cron job as a temporary workaround.

Any ideas on what might have caused this recent change in behavior?

Thanks for your help,

Drew


--- Begin debug output ---
ntpd 4.2.0a@1.1191-r Sat Feb 11 19:19:56 PST 2006 (1)
addto_syslog: ntpd 4.2.0a@1.1191-r Sat Feb 11 19:19:56 PST 2006 (1)
addto_syslog: precision = 3.000 usec
create_sockets(123)
addto_syslog: no IPv6 interfaces found
bind() fd 4, family 2, port 123, addr 0.0.0.0, flags=8
addto_syslog: Listening on interface wildcard, 0.0.0.0#123
bind() fd 5, family 2, port 123, addr 127.0.0.1, flags=0
addto_syslog: Listening on interface lo, 127.0.0.1#123
bind() fd 6, family 2, port 123, addr 192.168.1.6, flags=8
addto_syslog: Listening on interface eth1, 192.168.1.6#123
init_io: maxactivefd 6
local_clock: time 0 clock 0.000000 offset 0.000000 freq 0.000 state 0
bind() fd 8, family 2, port 123, addr 192.168.1.255, flags=8
io_setbclient: Opened broadcast client on interface 2, socket: 8
io_setbclient: Opened broadcast clients
addto_syslog: frequency initialized -36.958 PPM from /etc/ntp/ntp.drift
local_clock: time 0 clock 0.000000 offset 0.000000 freq -36.958 state 1
report_event: system event 'event_restart' (0x01) status 'sync_alarm, 
sync_unspec, 1 event, event_unspec' (0xc010)
auth_agekeys: at 1 keys 1 expired 0
timer: refresh ts 0
receive: at 15 192.168.1.6<-192.168.1.2 mode 5 code 5
Finding addr 192.168.1.2 in list of addresses
key_expire: at 15
peer_clear: at 15 assoc ID 33252 refid INIT
newpeer: 192.168.1.6->192.168.1.2 mode 3 vers 4 poll 6 10 flags 0x210 
0x20 ttl 0 key 00000000
receive: at 15 0.0.0.0<-192.168.1.2 mode 5 code 6
peer 192.168.1.2 event 'event_reach' (0x84) status 'unreach, 1 event, 
event_reach' (0x14)
auth_agekeys: at 60 keys 1 expired 0
transmit: at 79 0.0.0.0->192.168.1.2 mode 3
receive: at 79 192.168.1.6<-192.168.1.2 mode 4 code 1
clock_filter: popcorn 7.896193 0.000960
receive: at 82 192.168.1.6<-192.168.1.2 mode 5 code 6
receive: at 82 0.0.0.0<-192.168.1.2 mode 5 code 6
auth_agekeys: at 120 keys 1 expired 0
transmit: at 143 0.0.0.0->192.168.1.2 mode 3
receive: at 143 192.168.1.6<-192.168.1.2 mode 4 code 1
clock_filter: n 2 off -9.600170 del 0.000386 dsp 3.937744 jit 1.703977, 
age 64
receive: at 148 192.168.1.6<-192.168.1.2 mode 5 code 6
receive: at 148 0.0.0.0<-192.168.1.2 mode 5 code 6
auth_agekeys: at 180 keys 1 expired 0
transmit: at 206 0.0.0.0->192.168.1.2 mode 3
receive: at 206 192.168.1.6<-192.168.1.2 mode 4 code 1
clock_filter: n 3 off -9.600170 del 0.000386 dsp 1.938216 jit 1.639332, 
age 127
receive: at 215 192.168.1.6<-192.168.1.2 mode 5 code 6
receive: at 215 0.0.0.0<-192.168.1.2 mode 5 code 6
--- End debug output ---


-- 
gentoo-user@gentoo.org mailing list

[gentoo-user] Re: System Clock Problems

[Harry Putnam]

Drew Tomlinson <drew@mykitchentable.net> writes:

> The time server is a FreeBSD 6.0 box on my network.  My other FreeBSD
> box and two Windows boxes get time from it just fine.  Even the Gentoo
> box will set its clock with "ntpd -gq".  I am currently using this
> brute force method via a cron job as a temporary workaround.
>
> Any ideas on what might have caused this recent change in behavior?

It might be that the clock got off by more than ntp is willing to
adjust.   As I recall there is a threshold above which ntp will not
go.   Hopefully someone more knowledgable might confirm that.

To cure that sort of problem here I run ntp-client at boot.  It sets
the time by any amount I think.  So time gets set right on boot then
ntp will keep it in good order.  Next boot up ntp-client comes in
ahead of ntp and sets the clock before ntp gets to it, so the
too-large discrepancy never occurs.

I seem to recall that the too-large discrepancy is not really that
large.  I remember thinking it seemed kind of small to be a problem. 

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: System Clock Problems

[Drew Tomlinson]

On 2/12/2006 8:01 PM Harry Putnam wrote:
> Drew Tomlinson <drew@mykitchentable.net> writes:
>
>   
>> The time server is a FreeBSD 6.0 box on my network.  My other FreeBSD
>> box and two Windows boxes get time from it just fine.  Even the Gentoo
>> box will set its clock with "ntpd -gq".  I am currently using this
>> brute force method via a cron job as a temporary workaround.
>>
>> Any ideas on what might have caused this recent change in behavior?
>>     
>
> It might be that the clock got off by more than ntp is willing to
> adjust.   As I recall there is a threshold above which ntp will not
> go.   Hopefully someone more knowledgable might confirm that.
>   

Thanks for your reply.  If I understand correctly, using the '-g' option 
will allow a one time clock adjustment of any size.

> To cure that sort of problem here I run ntp-client at boot.  It sets
> the time by any amount I think.  So time gets set right on boot then
> ntp will keep it in good order.  Next boot up ntp-client comes in
> ahead of ntp and sets the clock before ntp gets to it, so the
> too-large discrepancy never occurs.
>   

I have used both 'ntpupdate' (the utility used by ntp-client) and 'ntpd 
-gq' (ntpd's way to mimic the behavior of ntpupdate) with successful 
results.  However I still can not seem to get ntpd to sync when run in 
daemon mode.

> I seem to recall that the too-large discrepancy is not really that
> large.  I remember thinking it seemed kind of small to be a problem. 
>   
1000 seconds according to my understanding of the man page.

Anyway, I really appreciate your thoughts.  Because this box runs 
MythTV, time is *VERY* important.  Imagine my surprise when I went to 
watch the first day of the Olympics on to find out that my recordings 
were off by over an hour and half.  :)

Thanks,

Drew


-- 
Visit The Alchemist's Warehouse
Magic Tricks, DVDs, Videos, Books, & More!

http://www.alchemistswarehouse.com

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: System Clock Problems

[Mark Knecht]

On 2/13/06, Drew Tomlinson <drew@mykitchentable.net> wrote:

> Anyway, I really appreciate your thoughts.  Because this box runs
> MythTV, time is *VERY* important.  Imagine my surprise when I went to
> watch the first day of the Olympics on to find out that my recordings
> were off by over an hour and half.  :)
>
> Thanks,
>
> Drew

Drew,
   Please excuse me jumping in here. I have seen no other emails in
this thread so maybe I'm wasting time. I hope not.

   I have a number of Gentoo machines. For some reason in the last
week or 10 days my AMD64 machine stopped keeping time. I'd boot it in
the morning and even the day would be off. It was horrible.

   With a little help from others I found that, at least in my case, I
needed to start running ntp-client in the default run level. I never
ran this before but there appears to have been change recently that
has made it more important. Either that or I was just lucky before.

   I've since added ntp-client on all the machines and things are now
dead on as far as I can tell.

   One other thing that was recommended to me was to remove the
/etc/adjust file when doing this. I did that also.

   Again, if all of this has already been covered I apologize for
taking up too much time.

Good luck,
Mark

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: System Clock Problems

[Jeff]

Hey Mark.

I'd also like to pitch in here.

Since about two weeks ago, my machines are no longer keeping time as
well. I'm getting errors when my systems boot during sysinit - the clock
script fails miserably, and some systems it forces me into console to
try and correct the problem.

I'm not sure that sysinit scripts have much to do with the system's
time, being that the kernel takes care of time - but again, since about
two weeks back, I have at least 4 systems that will not keep time,
regardless of hwclock or ntpdate commands.

The following message is recurring on these 4 machines:

select() to /dev/rtc to wait for clock tick timed out

Anyone know what's going on here?

Mark Knecht wrote:
> On 2/13/06, Drew Tomlinson <drew@mykitchentable.net> wrote:
> 
> 
>>Anyway, I really appreciate your thoughts.  Because this box runs
>>MythTV, time is *VERY* important.  Imagine my surprise when I went to
>>watch the first day of the Olympics on to find out that my recordings
>>were off by over an hour and half.  :)
>>
>>Thanks,
>>
>>Drew
> 
> 
> Drew,
>    Please excuse me jumping in here. I have seen no other emails in
> this thread so maybe I'm wasting time. I hope not.
> 
>    I have a number of Gentoo machines. For some reason in the last
> week or 10 days my AMD64 machine stopped keeping time. I'd boot it in
> the morning and even the day would be off. It was horrible.
> 
>    With a little help from others I found that, at least in my case, I
> needed to start running ntp-client in the default run level. I never
> ran this before but there appears to have been change recently that
> has made it more important. Either that or I was just lucky before.
> 
>    I've since added ntp-client on all the machines and things are now
> dead on as far as I can tell.
> 
>    One other thing that was recommended to me was to remove the
> /etc/adjust file when doing this. I did that also.
> 
>    Again, if all of this has already been covered I apologize for
> taking up too much time.
> 
> Good luck,
> Mark
> 

-- 
C-3PO:
	We seem to be made to suffer. It's our lot in life.

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: System Clock Problems

[Mark Knecht]

Answer is at the bottom

On 2/13/06, Jeff <jmg_071769@comcast.net> wrote:
> Hey Mark.
>
> I'd also like to pitch in here.
>
> Since about two weeks ago, my machines are no longer keeping time as
> well. I'm getting errors when my systems boot during sysinit - the clock
> script fails miserably, and some systems it forces me into console to
> try and correct the problem.
>
> I'm not sure that sysinit scripts have much to do with the system's
> time, being that the kernel takes care of time - but again, since about
> two weeks back, I have at least 4 systems that will not keep time,
> regardless of hwclock or ntpdate commands.
>
> The following message is recurring on these 4 machines:
>
> select() to /dev/rtc to wait for clock tick timed out
>
> Anyone know what's going on here?
>
> Mark Knecht wrote:
> > On 2/13/06, Drew Tomlinson <drew@mykitchentable.net> wrote:
> >
> >
> >>Anyway, I really appreciate your thoughts.  Because this box runs
> >>MythTV, time is *VERY* important.  Imagine my surprise when I went to
> >>watch the first day of the Olympics on to find out that my recordings
> >>were off by over an hour and half.  :)
> >>
> >>Thanks,
> >>
> >>Drew
> >
> >
> > Drew,
> >    Please excuse me jumping in here. I have seen no other emails in
> > this thread so maybe I'm wasting time. I hope not.
> >
> >    I have a number of Gentoo machines. For some reason in the last
> > week or 10 days my AMD64 machine stopped keeping time. I'd boot it in
> > the morning and even the day would be off. It was horrible.
> >
> >    With a little help from others I found that, at least in my case, I
> > needed to start running ntp-client in the default run level. I never
> > ran this before but there appears to have been change recently that
> > has made it more important. Either that or I was just lucky before.
> >
> >    I've since added ntp-client on all the machines and things are now
> > dead on as far as I can tell.
> >
> >    One other thing that was recommended to me was to remove the
> > /etc/adjust file when doing this. I did that also.
> >
> >    Again, if all of this has already been covered I apologize for
> > taking up too much time.
> >
> > Good luck,
> > Mark
> >

Jeff,
   I realized over lunch I didn't give Drew all the ammo he might have
wanted to go try out the fix for himself. (Not that he needs ammo from
me.) If your problems are at all similar to the ones I was suffering
through then you can test the fix by hand doing the following
commands:

/etc/init/ntpd stop
/etcinit.d ntp-client start
/etc/init.d ntpd start

/etc/init.d/ntpd zap after the stop to clear up stuff left behind by
the scripts or by ntpd crashing or stopping sometime unexpectedly. I
did.

After doing that my clocks were immediately back on time and I haven't
hd trouble since then, although it;s only been a few days since the
guys on the gentoo-amd64 list gave me the info. You might also have to
optionally do


I understand that the /etc/adjust file tells the system something
about how much to adjust time and that it's better to allow that to be
recreated or just let the system run without it. In my case I ran
without on for a few days and magically I now have a new one created
for me automatically:

mark@lightning ~ $ ls -la /etc/adjtime
-rw-r--r--  1 root root 44 Feb 13 08:35 /etc/adjtime
mark@lightning ~ $


Hope this helps,
Mark

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] Re: System Clock Problems

[Drew Tomlinson]

On 2/13/2006 12:22 PM Mark Knecht wrote:
> On 2/13/06, Drew Tomlinson <drew@mykitchentable.net> wrote:
>
>   
>> Anyway, I really appreciate your thoughts.  Because this box runs
>> MythTV, time is *VERY* important.  Imagine my surprise when I went to
>> watch the first day of the Olympics on to find out that my recordings
>> were off by over an hour and half.  :)
>>
>> Thanks,
>>
>> Drew
>>     
>
> Drew,
>    Please excuse me jumping in here. I have seen no other emails in
> this thread so maybe I'm wasting time. I hope not.
>   

Thanks for jumping in.  I'll try any suggestions at this point.  :)

> I have a number of Gentoo machines. For some reason in the last
> week or 10 days my AMD64 machine stopped keeping time. I'd boot it in
> the morning and even the day would be off. It was horrible.
>
>    With a little help from others I found that, at least in my case, I
> needed to start running ntp-client in the default run level. I never
> ran this before but there appears to have been change recently that
> has made it more important. Either that or I was just lucky before.
>   

This leads me to believe that your issues are related to the initial 
step being beyond the 1000 second limit by default.  I thought about 
this and thought the '-g' switch would override any such limit.  I have 
also tried your suggestion in my testing.  Not specifically by enabling 
ntp-client but by running ntpdate from the command line and then 
immediately running ntpd -A -d from the command line.  Still, no sync.

>    I've since added ntp-client on all the machines and things are now
> dead on as far as I can tell.
>
>    One other thing that was recommended to me was to remove the
> /etc/adjust file when doing this. I did that also.
>   

I don't seem to have one of these.

>    Again, if all of this has already been covered I apologize for
> taking up too much time.

No, this hasn't been covered and they're good suggestions.  I have a bit 
of an update.  In experimenting, I tried both "broadcastclient" and 
"server 192.168.1.2" (my time server machine) in ntp.conf.  Neither 
worked.  However, I read the man page for ntpd and came across the 
options of 'burst" and "iburst".  I added both of these options so now 
my ntp.conf contains the line "server 192.168.1.2 burst iburst".  This 
seems to work but I don't understand why.  One thing I noticed is that 
'ntpd -d' output shows a "burst" of many communications between my 
Gentoo box and my time server each time it communicates.  This results 
in a sync most of the time.  However my ntp.log file shows entries such 
as this:

13 Feb 19:11:20 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:12:10 ntpd[8830]: no servers reachable
13 Feb 19:12:24 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:13:13 ntpd[8830]: no servers reachable
13 Feb 19:13:27 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:14:16 ntpd[8830]: no servers reachable
13 Feb 19:14:32 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:15:22 ntpd[8830]: no servers reachable
13 Feb 19:15:36 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:16:27 ntpd[8830]: no servers reachable
13 Feb 19:16:41 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:17:31 ntpd[8830]: no servers reachable
13 Feb 19:17:45 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:18:34 ntpd[8830]: no servers reachable
13 Feb 19:18:48 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:19:39 ntpd[8830]: no servers reachable
13 Feb 19:19:53 ntpd[8830]: synchronized to 192.168.1.2, stratum 2
13 Feb 19:19:01 ntpd[8830]: time reset -52.104173 s

The "no servers reachable" bothers me a bit.  Is this indicative of a 
network issue and thus, the cause of my trouble?  Both my Gentoo box and 
my time server are on a 100 mbps LAN and so there should not be a 
"reachable" issue.

Just trying to understand...

Thanks,

Drew

-- 
Visit The Alchemist's Warehouse
Magic Tricks, DVDs, Videos, Books, & More!

http://www.alchemistswarehouse.com

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] System Clock Problems

[Nick Rout]

If your system stopped suddenly the ntp.drfit file may have become corrupted. As I understand it this file has a value in it that tells the system how much drift there is in the system clock, and uses the figutre to compensate. If the figure is way  out then the compensation will be way out.

Take a look at /var/lib/ntp/ntp.drift, if its seems like a large figure, then zero it and start again from scratch. ie stop ntpd, set the clock with ntpdate, then start the ntpd service again (preferably not while recording with mythtv LOL).


On Sun, 12 Feb 2006 12:10:20 -0800
Drew Tomlinson wrote:

> I have Gentoo 2.6.13-r5 kernel running and have used ntpd in 
> broadcastclient mode to keep its time in sync on my home network.  The 
> other day, the system suffered and abrupt shutdown due to a power 
> outage.  Ever since then, the system clock gains about 10 seconds every 
> 5 minutes.  Also, I can't get ntpd to sync the clock.  My command line 
> is "ntpd -A -b -g -u ntp:ntp".  I've included some output running the 
> command with the debug switch below.
> 
> I've also tried to gain more info with the ntpdc utility although I 
> don't really know what I'm doing.  However it appears that ntpd does see 
> my FreeBSD time server even though it's not synced:
> 
> ntpdc> peers
>      remote           local      st poll reach  delay   offset    disp
> =======================================================================
> =192.168.1.2     0.0.0.0          2   64    7 0.00038 -9.600170 1.98438
> 
> The time server is a FreeBSD 6.0 box on my network.  My other FreeBSD 
> box and two Windows boxes get time from it just fine.  Even the Gentoo 
> box will set its clock with "ntpd -gq".  I am currently using this brute 
> force method via a cron job as a temporary workaround.
> 
> Any ideas on what might have caused this recent change in behavior?
> 
> Thanks for your help,
> 
> Drew
> 
> 
> --- Begin debug output ---
> ntpd 4.2.0a@1.1191-r Sat Feb 11 19:19:56 PST 2006 (1)
> addto_syslog: ntpd 4.2.0a@1.1191-r Sat Feb 11 19:19:56 PST 2006 (1)
> addto_syslog: precision = 3.000 usec
> create_sockets(123)
> addto_syslog: no IPv6 interfaces found
> bind() fd 4, family 2, port 123, addr 0.0.0.0, flags=8
> addto_syslog: Listening on interface wildcard, 0.0.0.0#123
> bind() fd 5, family 2, port 123, addr 127.0.0.1, flags=0
> addto_syslog: Listening on interface lo, 127.0.0.1#123
> bind() fd 6, family 2, port 123, addr 192.168.1.6, flags=8
> addto_syslog: Listening on interface eth1, 192.168.1.6#123
> init_io: maxactivefd 6
> local_clock: time 0 clock 0.000000 offset 0.000000 freq 0.000 state 0
> bind() fd 8, family 2, port 123, addr 192.168.1.255, flags=8
> io_setbclient: Opened broadcast client on interface 2, socket: 8
> io_setbclient: Opened broadcast clients
> addto_syslog: frequency initialized -36.958 PPM from /etc/ntp/ntp.drift
> local_clock: time 0 clock 0.000000 offset 0.000000 freq -36.958 state 1
> report_event: system event 'event_restart' (0x01) status 'sync_alarm, 
> sync_unspec, 1 event, event_unspec' (0xc010)
> auth_agekeys: at 1 keys 1 expired 0
> timer: refresh ts 0
> receive: at 15 192.168.1.6<-192.168.1.2 mode 5 code 5
> Finding addr 192.168.1.2 in list of addresses
> key_expire: at 15
> peer_clear: at 15 assoc ID 33252 refid INIT
> newpeer: 192.168.1.6->192.168.1.2 mode 3 vers 4 poll 6 10 flags 0x210 
> 0x20 ttl 0 key 00000000
> receive: at 15 0.0.0.0<-192.168.1.2 mode 5 code 6
> peer 192.168.1.2 event 'event_reach' (0x84) status 'unreach, 1 event, 
> event_reach' (0x14)
> auth_agekeys: at 60 keys 1 expired 0
> transmit: at 79 0.0.0.0->192.168.1.2 mode 3
> receive: at 79 192.168.1.6<-192.168.1.2 mode 4 code 1
> clock_filter: popcorn 7.896193 0.000960
> receive: at 82 192.168.1.6<-192.168.1.2 mode 5 code 6
> receive: at 82 0.0.0.0<-192.168.1.2 mode 5 code 6
> auth_agekeys: at 120 keys 1 expired 0
> transmit: at 143 0.0.0.0->192.168.1.2 mode 3
> receive: at 143 192.168.1.6<-192.168.1.2 mode 4 code 1
> clock_filter: n 2 off -9.600170 del 0.000386 dsp 3.937744 jit 1.703977, 
> age 64
> receive: at 148 192.168.1.6<-192.168.1.2 mode 5 code 6
> receive: at 148 0.0.0.0<-192.168.1.2 mode 5 code 6
> auth_agekeys: at 180 keys 1 expired 0
> transmit: at 206 0.0.0.0->192.168.1.2 mode 3
> receive: at 206 192.168.1.6<-192.168.1.2 mode 4 code 1
> clock_filter: n 3 off -9.600170 del 0.000386 dsp 1.938216 jit 1.639332, 
> age 127
> receive: at 215 192.168.1.6<-192.168.1.2 mode 5 code 6
> receive: at 215 0.0.0.0<-192.168.1.2 mode 5 code 6
> --- End debug output ---
> 
> 
> -- 
> gentoo-user@gentoo.org mailing list

-- 
Nick Rout <nick@rout.co.nz>

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] System Clock Problems

[Drew Tomlinson]

On 2/13/2006 1:11 PM Nick Rout wrote:
> If your system stopped suddenly the ntp.drfit file may have become corrupted. As I understand it this file has a value in it that tells the system how much drift there is in the system clock, and uses the figutre to compensate. If the figure is way  out then the compensation will be way out.
>
> Take a look at /var/lib/ntp/ntp.drift, if its seems like a large figure, then zero it and start again from scratch. ie stop ntpd, set the clock with ntpdate, then start the ntpd service again (preferably not while recording with mythtv LOL).
>   
I've tried both zeroing this file and removing it altogether.  Zeroing 
doesn't work and if I delete it, it never gets recreated.  I've even 
tried "touch"ing it but it remains empty.  After touching it, I chmod it 
to ntp:ntp but still remains empty.  I'm by no means an expert but I 
don't get it.

Thanks for your reply,

Drew

-- 
Visit The Alchemist's Warehouse
Magic Tricks, DVDs, Videos, Books, & More!

http://www.alchemistswarehouse.com

-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] System Clock Problems

[Benno Schulenberg]

Drew Tomlinson wrote:
> I've tried both zeroing this file and removing it altogether. 
> Zeroing doesn't work and if I delete it, it never gets recreated.
>  I've even tried "touch"ing it but it remains empty.  After
> touching it, I chmod it to ntp:ntp but still remains empty.

>From 'man ntpd': "...  After one hour the frequency file is created 
and the current frequency offset written to it."

Benno
-- 
gentoo-user@gentoo.org mailing list

Re: [gentoo-user] System Clock Problems

[Drew Tomlinson]

On 2/14/2006 6:32 AM Benno Schulenberg wrote:
> Drew Tomlinson wrote:
>   
>> I've tried both zeroing this file and removing it altogether. 
>> Zeroing doesn't work and if I delete it, it never gets recreated.
>>  I've even tried "touch"ing it but it remains empty.  After
>> touching it, I chmod it to ntp:ntp but still remains empty.
>>     
>
> From 'man ntpd': "...  After one hour the frequency file is created 
> and the current frequency offset written to it."
>   

Right.  I should have been more clear.  Even after 2 hours, there was no 
file.  I suspected file permissions which is I used chown to set the 
file to ntp:ntp.

Thanks,

Drew

-- 
Visit The Alchemist's Warehouse
Magic Tricks, DVDs, Videos, Books, & More!

http://www.alchemistswarehouse.com

-- 
gentoo-user@gentoo.org mailing list