Philip Prindeville wrote:
Joy?
Or no joy?
Ron Byer Jr. wrote:
Sorry. I don't know how that came out to be 2041. Twice, no less. Intended
to say 2136. I think the latest was 2140 when I pulled it, and I scrambled
that to 2041.
Ron Byer Jr.
NetWeave Integrated Solutions, Inc.
+1.732.786.8830 x120
-----Original Message-----
From: Philip Prindeville [mailto:[EMAIL PROTECTED]
Sent: Friday, December 05, 2008 3:05 PM
To: AstLinux Users Mailing List
Subject: Re: [Astlinux-users] NTP Mixup :: the sequel
Why 2041?
Ron Byer Jr. wrote:
#1)
I'm building 2041, which will take a while.
#2)
Regarding modprobe -l mis-use:
Possibly so, but lsmod makes a mess on the console during boot time. It
still shows ipv6 loaded upon boot.
Status at present:
- With IPV6 commented out in rc.conf, lsmod | grep ipv6 shows ipv6 as a
loaded module, and ntpd bails out.
- For grins, I renamed ipv6.ko in the kernel modules path. Then lsmod |
grep
ipv6 shows no ipv6 loaded, and ntpd starts and works.
- I will try #2041 with ipv6 renamed back when it is cooked.
- I'm still missing something obviously, but I've got a hack-workaround at
present that will keep me quiet and off the list <gr>.
rb
-----Original Message-----
From: Philip Prindeville [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 04, 2008 6:55 PM
To: AstLinux Users Mailing List
Subject: Re: [Astlinux-users] NTP Mixup :: the sequel
We'll talk about 2131 off-line... :-)
Pick up #2133. This might fix your issue.
-Philip
Darrick Hartman wrote:
Ron,
I've confirmed two things. If ipv6 is NOT enabled, ntpd runs fine.
(restart did not work--fixed in SVN 2131).
You would also have to have IPV6=YES uncommented in rc.conf (it's not
used by default). If you enable IPV6, ntpd does indeed fail.
Darrick
Ron Byer Jr. wrote:
My log entries seem to make it clear that the failure to bind is the
reason. I had read the earlier NTP mixup posts and had decided to post
this when it appeared to be different. It doesn't appear to get that
far. (to comparing time differences).
/var/log/messages
Mar 3 14:28:03 sk3 daemon.info ntpd[1518]: Listening on interface #0
wildcard, 0.0.0.0#123 Disabled
Mar 3 14:28:03 sk3 daemon.err ntpd[1518]: unable to bind to wildcard
socket address :: - another process may be running - EXITING
Mar 3 14:28:04 sk3 daemon.notice ntpd[1520]: ntpd [EMAIL PROTECTED] Sat
Nov 15 06:25:14 UTC 2008 (1)
Mar 3 14:28:04 sk3 daemon.debug ntpd[1522]: signal_no_reset: signal 13
had flags 4000000
Mar 3 14:28:04 sk3 daemon.info ntpd[1522]: precision = 1.686 usec
Mar 3 14:28:04 sk3 daemon.debug ntpd[1522]: ntp_io: estimated max
descriptors: 1024, initial socket boundary: 16
Mar 3 14:28:04 sk3 daemon.info ntpd[1522]: Listening on interface #0
wildcard, 0.0.0.0#123 Disabled
Mar 3 14:28:04 sk3 daemon.err ntpd[1522]: unable to bind to wildcard
socket address :: - another process may be running - EXITING
The end result is a couple of ntpd zombies and a date back in March
1980.
sk3 log # ps -w | grep ntpd
1521 root Z [ntpd]
1522 root Z [ntpd]
I modified the ntpd startup script to strace ntpd and found the
following relevant system calls and status:
if nslookup $first >/dev/null; then
# Set the clock (large change) and exit
strace -f ntpd -g -q -c /etc/ntpd.conf
sleep 1
# Maintain the clock (small changes)
strace -f ntpd -c /etc/ntpd.conf
The results were as follows. Note the two bind calls IPV4 - works, and
IPV6 - which fails with the EADDRINUSE
setsockopt(16, SOL_SOCKET, SO_REUSEADDR, [0], 4) = 0
bind(16, {sa_family=AF_INET, sin_port=htons(123),
sin_addr=inet_addr("0.0.0.0")}
, 16) = 0
setsockopt(16, SOL_SOCKET, SO_TIMESTAMP, [1], 4) = 0
fcntl(16, F_SETFL, O_RDONLY|O_NONBLOCK) = 0
rt_sigaction(SIGPIPE, {0xb7f49762, [], SA_RESTORER, 0xb7f4db48},
{SIG_IGN}, 8)
=
0
time([320960202]) = 320960202
open("/etc/TZ", O_RDONLY) = 4
read(4, "EST5EDT\n", 68) = 8
read(4, "", 60) = 0
close(4) = 0
getpid() = 1631
write(3, "<30>Mar 3 14:36:42 ntpd[1631]: "..., 89) = 89
rt_sigaction(SIGPIPE, {SIG_IGN}, NULL, 8) = 0
socket(PF_INET, SOCK_STREAM, IPPROTO_IP) = 4
close(4) = 0
socket(PF_INET6, SOCK_STREAM, IPPROTO_IP) = 4
getsockname(4, {sa_family=AF_INET6, sin6_port=htons(0),
inet_pton(AF_INET6, "::"
,
&sin6_addr), sin6_flowinfo=0, sin6_scope_id=0}, [28]) = 0
close(4) = 0
socket(PF_INET6, SOCK_DGRAM, IPPROTO_IP) = 4
fcntl(4, F_DUPFD, 16) = 17
close(4) = 0
setsockopt(17, SOL_SOCKET, SO_REUSEADDR, [0], 4) = 0
bind(17, {sa_family=AF_INET6, sin6_port=htons(123), inet_pton(AF_INET6,
"::", &s
in6_addr),
sin6_flowinfo=0, sin6_scope_id=0}, 28) = -1 EADDRINUSE (Address
alrea
dy in use)
close(17) = 0
rt_sigaction(SIGPIPE, {0xb7f49762, [], SA_RESTORER, 0xb7f4db48},
{SIG_IGN}, 8) =
0
time([320960202]) = 320960202
open("/etc/TZ", O_RDONLY) = 4
read(4, "EST5EDT\n", 68) = 8
read(4, "", 60) = 0
close(4) = 0
getpid() = 1631
write(3, "<27>Mar 3 14:36:42 ntpd[1631]: "..., 120) = 120
rt_sigaction(SIGPIPE, {SIG_IGN}, NULL, 8) = 0
_exit(1) = ?
Process 1631 detached
Ron Byer Jr.
NetWeave Integrated Solutions, Inc.
+1.732.786.8830 x120
-----Original Message-----
From: Tod Fitch [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 04, 2008 3:47 PM
To: AstLinux Users Mailing List
Subject: Re: [Astlinux-users] NTP Mixup :: the sequel
I don't have access to my net5501 AstLinux box from here. I will get
around to reconfiguring my VPN sometime, I promise. :)
So this is from memory...
I believe that my ntpd was also unable to bind to ipv6, but that is
not what was causing it to quit. What was causing it to quit was that
the hardware clock time was more than 1000 seconds different from the
correct time. The work around was to stop zaptel, start rtc, set the
date to something close the correct time, use hwclock to write the
good time into the hardware clock then start up ntpd and zaptel.
Do your log entries show that failing to bind to the ipv6 port caused
the daemon to exit or did it exit later with a time error too large
type of error?
--Tod
On Dec 4, 2008, at 12:30 PM, Ron Byer Jr. wrote:
No change. Still fails on the ipv6 bind.
Ron Byer Jr.
NetWeave Integrated Solutions, Inc.
+1.732.786.8830 x120
-----Original Message-----
From: Lonnie Abelbeck [mailto:[EMAIL PROTECTED]
Sent: Thursday, December 04, 2008 2:11 PM
To: AstLinux Users Mailing List
Subject: Re: [Astlinux-users] NTP Mixup :: the sequel
On Dec 4, 2008, at 12:38 PM, Ron Byer Jr. wrote:
Lonnie, et al:
Do you have the NTP startup issue ? I'm stumped by why this doesn't
have
more pervasive impact. I'm always ready to find out that it's a
cockpit
error on my part.
No problem here, but I run my own NTP server with only IPv4 DNS.
Try a single server like "ntp3.cs.wisc.edu" and see if that works for
you.
Lo
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Astlinux-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/astlinux-users
Donations to support AstLinux are graciously accepted via PayPal to [EMAIL
PROTECTED]
Delayed joy.
the build came to grief on iptables. lots of unresolved symbols in link.
going to sort through the rubble in the morning.
rb
------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you. Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/
_______________________________________________
Astlinux-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/astlinux-users
Donations to support AstLinux are graciously accepted via PayPal to [EMAIL
PROTECTED]
