I spent a lot of time chasing what I thought was a missing syscall in the
seccomp list, but that was a wild goose chase.
For me, it currently breaks even without seccomp.
I'm running code that was built on the 24th. That was probably shortly after
a git pull, so the change that broke things i
> Hmm. Works just fine for meâ¦
I think the problem is that gdb is broken on ARM. Does it work for you?
I get the same results on another system and when running code that is known
to work when not run from gdb. It crashes before it gets to a breakpoint at
main.
--
These are my opinions.
strom...@nexgo.de said:
> I avoid the use of debuggers whenever possible, so I can't comment (that's a
> habit I got into when maintaining some large-scale numerical simulation
> codes and it stuckâ¦).
It seems like the right tool for this problem. What I need is a stack trace
from a crash.
I
strom...@nexgo.de said:
> I don't see any option for ntpd to not fork into a daemon process, so you'll
> at least have to setup gdb to follow the process into the child.
-n on the command line avoids the fork. Works great for things like this.
--
These are my opinions. I hate spam.
_
I collected 3724 samples each for IPv4 and IPv6.
3323 IPv4 samples had names.
2919 IPv6 samples had names.
There were 300 unique IPv4 addresses and 275 unique IPv4 names.
There were 275 unique IPv6 addresses and 119 unique IPv6 names.
I didn't check to see if the forward DNS matched.
The longest
The mru list is getting printed out oldest first. I thought you fixed that.
The print-what-you-have after ^C doesn't work if the output is a pipe.
(Classic doesn't work either.)
[murray@second ~]$ ntpq -nc mru | tee foo.log
Ctrl-C will stop MRU retrieval and display partial results.
^Cclose fai
[murray@glypnod play]$ ntpq -c "rv 0"
status=0115 leap_none, sync_pps, 1 event, clock_sync,
/bin/stty: when specifying an output style, modes may not be set
/bin/stty: when specifying an output style, modes may not be set
/bin/stty: when specifying an output style, modes may not be set
/bin/stty: w
The current peers command is broken. It's printing a big bunch of blanks
instead of the host name for each slot. It way overflows an 80 character
line.
My mail stuff will probably break these long lines.
[murray@glypnod play]$ ntpq -p cent2
remote refid st t when poll rea
I just did a fresh clone (chasing another bug).
I did ./waf configure build
Then I did cd ntpq and ./ntpq
It doesn't work. I have to run ./waf check to build some of the python ntp
libraries.
[murray@hgm ntpq]$ ./ntpq
ntpq: can't find Python NTP library -- check PYTHONPATH.
No module named ntp
It doesn't have a ^C catcher for the printout part of the mru command. I was
expecting ^C to stop printing stuff and go back to the main read-a-command
loop.
-
>> The mru list is getting printed out oldest first. I thought you fixed
that.
> I thought I did too. Testing...
It still
Waf: Leaving directory `/home/murray/ntpsec/play/hgm/main'
Wrote test log to: /home/murray/ntpsec/play/hgm/main/test.log
execution summary
tests that pass 2/3
/home/murray/ntpsec/play/hgm/main/tests/test_libntp
/home/murray/ntpsec/play/hgm/main/tests/test_ntpd
tests that fail 1/3
e...@thyrsus.com said:
> Very nice - except that (a) I failed to notice that the TIOCGWINSZ call
> returned (columns, rows) rather than (rows, columns), and (b) I forgot to
> 'waf build' after changing pylib/util.py, so I ran the stale version of
> pylib/util.py when testing, so I didn't see your b
e...@thyrsus.com said:
> No, that's not it. I know how the waf build logic works, I've had to modify
> it quite often. I know you think you've found a bug there, but I think
> you're misinterpreting something. I'll figure it out.
I think the test case that started this thread indicates that s
> Here's the inside view from looking at the code: --disable-kernel-PLL turns
> off the use of ntp_adjtime() to slew time, leaving adjustments to be done at
> much coarser granularity by the old-style adjtime(2) call.
How about we rename it to --disable-ntp_adjtime?
I assume we will get the sam
> Done, and for good measure I've similarly wrapped the peers display.
Should it be around the whole command loop? Do we want ^C when reading a
command to start over reading a command or exit? I vote for try again. I'll
use ^D if I want to exit.
Is DNS lookup called by any other commands?
[murray@hgm play]$ ntpq -c "rv 0 version"
version="ntpd 0.9.6-1eea2ce-hgm Dec 2 2016 05:18:33"
[murray@hgm play]$ ntpq -c "rv 0 version" 2>&1
version="ntpd 0.9.6-1eea2ce-hgm Dec 2 2016 05:18:33"
[murray@hgm play]$ ntpq -c "rv 0 version" 2>&1 | cat
Bailing out...
[murray@hgm play]$ ntpq -c "rv 0
> I'm stuck trying to get Ubuntu to install libpython2.7.a
Who needs it? What for?
Is this a waf bug - stuff in main that should be in host?
--
These are my opinions. I hate spam.
___
devel mailing list
devel@ntpsec.org
http://lists.ntpsec.org/m
g...@rellim.com said:
> Lot's of people say they do it, plus I'm getting complaints about NTPsec
> build failing on ARM.
Building directly works for me on Raspbian, NetBSD, and FreeBSD.
> I'm not saying there is not a waf bug, but my immediate problem is that for
> waf to compile the test.c pro
[moved from users@]
e...@thyrsus.com said:
> That is correct. Broadcast client mode has been removed because it is
> impossible to secure.
We really should have a single top-level file that covers all the simple
differences from ntp-classic.
We should probably patch the parser to emit an err
I tried it, and it barfed with:
[murray@hgm ntpclients]$ ./ntpmon
Traceback (most recent call last):
File "./ntpmon", line 91, in
variables, peer.associd))
File "/usr/local/lib/python2.7/site-packages/ntp/util.py", line 224, in
summary
if hmode == ntp.magic.MODE_BCLIENT:
AttributeE
e...@thyrsus.com said:
> Should be dot, sys.path, and PYTHONPATH in that order. Best guess is you
> weren't running in the directory you thought you were.
I'm pretty sure I was in the right place. I remember doing an ls and there
was the ntp link I expected.
> That's probably due to an over
gha...@gmail.com said:
> Does this mean that "ephermal" mode, "Ephemeral associations are mobilized
> upon arrival of broadcast or multicast server packets and demobilized by
> timeout", no longer exists? Or is there another use case for that mode?
They are used by the pool mode.
> As NTPSec
e...@thyrsus.com said:
> It now traps attempts to draw past end of screen. And takes a hostname
> argument. -n and -w options parallel to those of ntpq have been added.
Thanks. It's off the ground. I get a screen full of ntpq printout when
pointed at a system with lots of servers but nothing
I think you are automagically generating pylib/magic.py
I commented out MODE_BCLIENT in ntp.h as it's no longer used in ntpd or the c
libraries after I removed a layer of code no longer used by broadcastclient
and multicastclient.
You added some code in util.py that references it to print out
e...@thyrsus.com said:
> Remove it. I'm in some doubt that operators even notice the 'b' code, let
> alone ever act on it.
I figured out another approach. Leave the old one commented out, but add a
new symbol, say MODE_BCLIENTx with the same value (and a comment) and patch
ntpq.py to use it.
e...@thyrsus.com said:
> I don't know how to read the damned output. I have barely
> any idea what those columns mean. I can't tell descending
> from ascending from middle-outwards!
It's what you want if you are trying to figure out who is using your server.
A couple of examples from a handy
e...@thyrsus.com said:
>> It's what you want if you are trying to figure out who is using your
server.
> *blink*
...
> /me adds that sentence to the mrulist
I didn't see where you added it.
It's probably appropriate to change the "using" to "using or abusing".
An example might be interesting
e...@thyrsus.com said:
> I'd like to hack on this, but my priorities now have to be (a) ripping out
> the interface-scanning hair, and (b) simplifying and fixing some broken
> stuff in the waf recipe. I can take some time out for ntpmon because it's
> so trivial to modify, but I need to focus.
> Note it said "Python NTP library". That is the one it can not find.
A module name would be helpful.
> What is your PYTHONPATH? And where are your 'site-packages'?
My PYTHONPATH and site-packages are the same as they were yesterday when
things worked.
I think the problem is tangled up with n
It works on 7.0.2
-bash-4.3$ ntpq -p
ntpq: can't find Python NTP library -- check PYTHONPATH.
Shared object "libpython2.7.so.1.0" not found
-bash-4.3$ uname -a
NetBSD bob.example.com 6.1.5 NetBSD 6.1.5 (GENERIC) i386
-bash-4.3$ locate libpython2.7.so.1.0
/usr/pkg/lib/libpython2.7.so.1.0
-bash-4.3$
> I like it, and learn towards saying yes. Let's see what Hal and others say.
I'm happy to change the default, but I think we need some text that explains
why.
The old noquery was there to fix a DDoS amplification attack using the old
monlist. That's not possible any more, but there
It's commenting a line I expect to get translated.
Input:
#define MODE_CONTROL6 /* control mode */
#define MODE_PRIVATE7 /* Dead: private mode, ntpdc */
/*
* This is a madeup mode for broadcast client. No longer used.
*/
/* #define MODE_BCLIENT6 ## broadcast
Disclaimer: I could easily get something backwards. I've worked with both
and can easily switch even when I shouldn't.
> +#define LAST32MASK 0xUL
> +#define FIRST32MASK 0xUL
FIRST and LAST seem like poor choices for words. Do you mean first on the
wire? Ne
waf distclean doesn't get rid of the stuff added to pylib/
Should that stuff really go into build/main/pylib/?
waf disclean doesn't remove the ntp link from ntpclients
The link from build/main/pylib/ntpc.so is absolute rather than relative
--
These are my opinions. I hate spam.
___
It works if I change the x to X
Old:
>From ntp.h
#define MODE_BCLIENTx 6 /* for pylib/util.py */
In magic.py
#define MODE_BCLIENTx 6 # for pylib/util.py
New:
#define MODE_BCLIENTX 6 /* for pylib/util.py */
MODE_BCLIENTX = 6 # for pylib/util.py
--
These are my
e...@thyrsus.com said:
> The simplest way to get to a consistent UI design was to remove all
> command-line options. Which is a good idea anyway - less to remember, and
> the command keystrokes cover the functionality.
I think you are rationalizing a bad decision. By that logic, I could improv
This is on a 32 bit Fedora.
[142/166] Compiling tests/libntp/vi64ops.c
In file included from ../../include/ntp_syslog.h:11:0,
from ../../include/ntp_stdlib.h:17,
from ../../tests/libntp/vi64ops.c:2:
../../tests/libntp/vi64ops.c: In function âTEST_vi64ops_HiLoVUI
e...@thyrsus.com said:
> Some or all of these are fixed now; I responded earlier today to a report
> from a 32-bit MIPS build.
A git pull didn't find anything new.
--
These are my opinions. I hate spam.
___
devel mailing list
devel@ntpsec.org
ht
> Hm. OK. Please reproduce those warnings and post them as a tracker issue.
There is already one there that looks close enough. See #203
--
These are my opinions. I hate spam.
___
devel mailing list
devel@ntpsec.org
http://lists.ntpsec.org/mai
> Right, I think I fixed those. "mips4 port fix", Tue Dec 13 08:53:16 2016
> -0500.
That didn't fix the errors I was seeing.
I pushed a fix that works for me. I can't test mips or other-endian.
What sort of buildbot do we have? I thought there was some sort of setup
that ran proposed commit
I have a crappy wifi-setup. Great for testing things like this. :)
[murray@hgm ~]$ ntpmon ntp-wifi
***Request timed out
[murray@hgm ~]$
Is there any simple way I can collect more info?
--
These are my opinions. I hate spam.
___
devel mailing l
For some of my examples, it takes a long long long time to get started. If I
hadn't been through the DNS stuff before, I'd assume that it was hung. The
screen is blank, no hint of progress.
I could get off the ground quickly if I could put the "n" on the command line.
I can't find the targe
e...@thyrsus.com said:
> I've never seen it before, but that's pretty much what would be expected if
> openhost() fails. Not much that can be done about it either, except by not
> using a crappy betwork.
No, it runs long enough to give me a screen so openhost() seems to be working.
> Transact
> ntpq: standard-mode lookup of -D failed, Name or service not known ntpq: ndp
> lookup failed, Servname not supported for ai_socktype
>From pylib/packet.py:
return hinted_lookup(port="ndp", hints=0)
My /etc/services doesn't have any slots for ndp. Is that a typo?
--
Was: Subject: Re: ntpmon now has some keystroke commands
gha...@gmail.com said:
> Unfortunately, I am unable to play with this new toy, as all my servers that
> run NTPsec are in the pool, and ntpmon just goes to sleep on them (see issue
> #206).
There are two problems in this area. (at least t
e...@thyrsus.com said:
> That's a puzzler. Naively setting up debugging on a TUI won't work. Do you
> get analogous behavior from ntpq?
I'm willing to hack the source code.
I don't remember any troubles from peers on ntpq. With ntpq, mrulist hangs
rather than exiting.
--
These are my opin
e...@thyrsus.com said:
>> From pylib/packet.py:
>> return hinted_lookup(port="ndp", hints=0)
>> My /etc/services doesn't have any slots for ndp. Is that a typo?
> If so, it's copied from the C version. I didn't understand it, so I didn't
> screw with it.
grep ndp -w do
Context is mrulist not working on ntpq when there are lots of slots.]
> I think you have misdiagnosed this. My testing on your flaky wifi link
> suggested strongly that lots of clients isn't the issue - lots of packet
> dropouts is the issue.
The new ntpq times out when run via localhost.
The
>> I don't remember any troubles from peers on ntpq.
> Wait, are you having peers troubles with npmon? That's new.
I don't remember any troubles from peers from ntpq.
ntpmon bails without telling me which section it was working on. All it says
is:
***Request timed out
so I don't know if that's
I think I figured out one contribution to my confusion.
When printing out a backtrace, it uses the text from the installed version
but it's actually running the local code.
[murray@hgm ntpclients]$ ./ntpmon ntp-wifi
Traceback (most recent call last):
File "./ntpmon", line 164, in
variabl
> That doesn't sound possible to me. Think about how an interpreter works.
Do you have a better suggestion?
It said:
> File "/usr/local/lib/python2.7/site-packages/ntp/util.py", line 273, in
> summary
> sys.stderr.write("DNS lookup ends.\n")
> NameError: global name 'logfp' is not defined
The traffic on the pool took a big jump recently. There were a couple of
comments on the pool list, but nothing past "lots of traffic".
Understanding what's going on is the top of my list.
A couple of interesting counters were lost in the recent (month or 2 ago)
work on the packet processing.
Recent NTP pool traffic increase
http://mailman.nanog.org/pipermail/nanog/2016-December/089525.html
Not much interesting so far.
--
These are my opinions. I hate spam.
___
devel mailing list
devel@ntpsec.org
http://lists.ntpsec.org/mailman/listinf
[was broadcastclient]
e...@thyrsus.com said:
>> Should that sort of trap fail to start? Will users get surprised
>> when they miss the error message or annoyed when they know
>> what to expect but they have to work harder to try it?
> I'm a big fan of failing noisily and early in circumstances l
e...@thyrsus.com said:
>> I took a quick look at the code. There is quite a bit of broadcastclient
>> cruft that didn't get removed. Any reason not to nuke it?
> No. I think I've removed it all since you wrote it.
You got most of what I noticed. I think I pushed something that got a few
mor
[was broadcastclient]
e...@thyrsus.com said:
>> We really should have a single top-level file that covers all
>> the simple differences from ntp-classic.
> It's a section on docs/index.txt. I'm careful to keep it up to date.
Thanks. That looks good. I wonder how people are going to find it. H
Thanks. That's a big help.
[mru retransmissions]
> Each span request except the first is supposed to include identifications of
> late MRU entries from the previous span. If the daemon can't match those
> from the MRU records it's holding in core, that means some of the records
> that existed a
fr...@nicholasfamilycentral.com said:
> Most modern retail routers/WiFi access points will distribute NTP via DHCP.
Right. But who sets that up?
What I was trying to say is that most retail ISPs don't have their own NTP
servers so they don't include any NTP server info in their DHCP responses.
g...@rellim.com said:
>> I assume that's a typo. The counter has to be in ntpq
>> rather than ntpd. ntpd doesn't store any state for clients.
> My experiments with the lastest ntpmon contradict your statement.
> ntpd does seem to be keeping the number of requests in the mrulist between
> ntpmon r
e...@thyrsus.com said:
> Correct. As you note, the documentation used ti caution agains using the
> IANA multicat address. I don't know why.
Because multicast wasn't well thought out and never widely implemented in a
way that worked well.
The fundamental problem is how much of the net do you
gha...@gmail.com said:
> As a broadcast client, I listen to broadcast messages. I know what happens
> when I get one, I mobilise an (ephermal) association. What if I get two?
You mobilize another ephemeral association. There is probably some sanity
check for a total limit.
> As a broadcast
> I have since fixed this. Mode 6 monitoring statistics are back.
Thanks.
--
These are my opinions. I hate spam.
___
devel mailing list
devel@ntpsec.org
http://lists.ntpsec.org/mailman/listinfo/devel
I've found an interesting problem.
A successful batch returns a new nonce. ntpq asks for more using the old
one. Nothing comes back.
That seems broken-by-design. If the response is lost, the server will have
switched to the new nonce but the client only knows the old one. It might
work if
I added + and - as keyboard commands to ntpmon to bump the debug level.
The first + opens ntpmon.log in append mode
ntpq's debug commands now open ntpq.log
I haven't sorted out what gets printed at each level.
I bumped the hex packet printout to level 5.
Level 4 just prints out the text which wo
There is a showall parameter to the worker printout code for the peers
display in ntpq and ntpmon.
ntpmon has the s keyboard command to toggle it. ntpq has lpeers and
lassociations.
Short version: I vote we nuke it - general cruft reduction.
Long version:
I've never found a good use for it.
e...@thyrsus.com said:
> Wait, then I have failed to undersrtaand your bug report. This can happen
> in a different, less odd way than the nonce update getting lost by packet
> drop?
Yes. The no-packet-lost case is broken if it needs a second batch.
Each batch gets a new nonce. The code does
e...@thyrsus.com said:
> The Python client library now requests nonces every 16 seconds during an
> MRU-request sequence, rather than every 4 requests.
I don't think we are on the same wavelength yet.
You may have fixed a bug (thanks), but ntpd still sends back a new nonce that
is getting igno
Lots of work still to do, but we are off the ground.
The second batch never worked. In hindsight, it's actually easy to test the
multiple batch case. Just set frags=1 That gets about 4 slots, so a server
with a handful of clients is plenty.
I fixed it to take frags=xx and limit=xx on the com
More on NANOG yesterday.
Starts roughly here:
http://mailman.nanog.org/pipermail/nanog/2016-December/089588.html
>> I found devices doing lookups for all of these at the same time
>> {0,0.uk,0.us,asia,europe,north-america,south-america,oceania,africa,europe}
.pool.ntp.org
> Confirmed - starti
>> Note that it is using the old nonce.
> I had to stare at this for quite a while before I figured it out.
> When I switched from the old C code's buggy use of a counter to trigger the
> new nonce request to a timeout, I also needed to invert the order of
> operation of the nonce request and the
I implemented a direct mode. It writes out each batch of slots as soon as it
gets them. Any sort options are ignored. There will be duplicates of any
slots that get updated after they are retrieved. I think the filtering stuff
should still work but I didn't try it.
The code and UI need mo
Is there any way to set a flag and continue, or does python always do the
equivalent of a longjmp to the Except catcher?
I don't think so, but it seems worth checking before I go uglify some code.
Context is rare crashes when ntpq is trying to print out a mru list slot that
is only half setup.
> I can't find any commit that looks like it involvs a 'direct' flag. Did you
> push this, or is it a private set of changes? If the latter, I'd like to
> see the patch and play with it.
Sigh. I forgot to push last night. I just fixed that.
My cron jobs worked last night.
e...@thyrsus.com
> Hal, I couldn't think up a better UI than yours for direct mode. I don't
> like it much, but I think under our constraints you did as well as you
> could.
It might be simpler to describe if "direct" called mru rather than setting a
flag. It could also turn off hostnames since they don't make
I implemented the mru sort=addr
I fixed the ^C during collection. Only a few lines.
(Plan B was to mask the signal. There doesn't seem to be any way to do that
from Python.)
Minor quirks:
The frags= and limit= on the mru command are only used for the first batch.
I'd like them to stick.
T
e...@thyrsus.com said:
>> My claim is that the interpreter is doing the right thing
>> and finds the local util.pyc but the debugger isn't following
>> the same search rules and finds the wrong util.py
>> Or something like that.
> Do you still have this problem? Did you ever figure out what was
[Context is T_Auth not really being needed.]
> I think the intent here is to enable you to type ":config reset auth" in
> ntpq and reset the counter tracking authorization lookups.
Ahh. I guess that makes sense. I was thinking the whole batch got reset on
the hourly print stats run or someth
e...@thyrsus.com said:
>> The frags= and limit= on the mru command are only used
>> for the first batch. I'd like them to stick.
> There's a computation of those for second and later span requests that I
> transcribed from the C, down around line 1287 in packet.py. I'm very
> reluctant to mess
[context is config_reset_counters]
>> Does it get called with the whole list or the right slot from the list?
> I think the intent is that you can reset multiple flag types with one
> command.
I hacked in some logging.
It gets called once for the config file with a list of all the things to
re
I want to add a file to /etc/logrotate.d/ to tell logrotate how to process
ntpd's log files.
I get email telling me:
/etc/cron.daily/logrotate:
error: failed to open config file ntpd: Permission denied
error: found error in file ntpd, skipping
google tells me that's because SELinux is blo
strom...@nexgo.de said:
> From what you've been showing I think the config file needs to be in the
> system_u context in order for logrotate to not pick up any files that may
> have been dropped into the directory maliciously.
Thanks. That sounds right, but what do I type to make it happen?
e...@thyrsus.com said:
> Hal, I was reviewing some recent changes, and it occurred to me that your
> commit changing the default of showall in ntpq may have not been a good
> idea.
> I don't think the showall functional change in either ntpmon or ntpq was a
> bad thing in itself. But backwards-i
> It does not exist on any system I have access to, and ntp runs fine.
It's the API to the kernel PPS code. It's not needed unless you have
something like a GPS that also provides a PPS signal, usually on DCD, pin 1,
of a serial port.
--
These are my opinions. I hate spam.
__
> ... and if you are not using gpsd ?
If you have PPS hardware, there are 3 approaches.
You can use GPSD. It will use the kernel PPS capture if the kernel supports
it. If not, it will try user mode timing. I think that is only available on
Linux.
You can use the PPS driver. That requires t
gha...@gmail.com said:
> (If this is documented so clearly somewhere, please point me there. Else I
> will summarise and add to the gpsd and ntpsec docs later).
There is a chunk about timepps in INSTALL.
I don't think there is any overview about PPS processing. There should be.
I was going
ja...@azze.org said:
> To change the context of the file, try:
> chcon -t system_u ntp.conf
Thanks. That's the part I was missing.
It doesn't work when SELinux is disabled, or something like that.
I ended up having better luck with --ref rather than -t.
--
I've pushed logrotate-ntpd
I added a few lines of code to the main receive routine. They were copied
from old-old code and restore bumping sys_newversion and sys_oldversion.
There is also a bail case that I don't fully understand. Would you please
check. It seems to be working, but there might be something interestin
I have an cleaned up version of the allocator mostly working.
My current problem is that I need to fix ntpq to not give up if it asks for a
slot that that ntpd doesn't support. Can you either fix it, or give me some
guidance on where the fix should go?
You can test with "rv 0 leap,foo" and/or
[how to document a flowchart]
> I would do this kind of thing in pic, which is why I wrote the pic2graph
> tool in the groff suite.
Thanks. I think I'll use text. It's just a simple linear graph.
> You are way, way ahead of me in understanding that code. Do what you think
> is needed. Try
File "/usr/local/lib/python2.7/site-packages/ntp/packet.py", line 1429, in
mrulist
del parms['recent']
TypeError: 'str' object does not support item deletion
The default parameters get 100 or so in one batch. You can force a second
batch something like this:
mru recent=100 frags=1
--
> If you you any reason not to tag and release 0.9.6 in the next few days,
> please let us know.
I think it needs a lot of testing and bug fixing and polishing rather than
new features that end up breaking something.
A few days is New Years. I don't think it will get enough testing and or
fi
e...@thyrsus.com said:
> This is -- , where is the commit count
> since the last tag.
The tick-this-version idea might be more obvious if you change the - between
version and tick to a +
--
These are my opinions. I hate spam.
___
devel mailin
The rawstats table in the man page for ntp.conf is just a bit too wide for my
windows so the right border line wraps around. My window is 80 characters
wide.
My attempts at fixing it didn't work.
The code is in docs/includes/mon-commands.txt
Several of the --- lines that are 81 characters
> I can't reproduce this. I saw it a few times when you first reported it,
> pushed a fix, and haven't seen it since.
Here is my recipe.
git clone foo
cd foo
./waf configure build
cd ntpclients
./ntpq
I get:
[murray@hgm ntpclients]$ ./ntpq
ntpq: can't find Python NTP library -- check PYTHONP
I made the tarball in my working directory so things may have been confused
by stuff that distclean didn't clean.
1290 ./waf dist
1291 ll noname-1.0.tar.bz2
1292 mkdir ../foo
1293 mv noname-1.0.tar.bz2 ../foo/
1294 cd ../foo/
1295 ll
1296 tar -xvf noname-1.0.tar.bz2
1297 cd nona
We are dependent on waf and autorevision. Any others?
They are interesting in that we provide the copy we want rather than
expecting the environment to provide it.
How do we get new versions and/or contact the maintainers if you get run over
by a truck? Can we find them with a quick google o
Thanks.
>> Do we have a recipe for testing new versions?
> We don't. Fortunately, in both cases, if the tool broke due to a flag day it
> would probably become unignorably obvious pretty quickly.
For a start, my suggestion would be to make a tarball and unpack and build
and run on another syst
The details of how it works are in the ntpq man page. Search for maxage or
minage.
A server will end up operating in one of several modes, depending on the
ratio of traffic and MRU table size.
If the server has only a few clients, they each get a slot and they will
stick around until ntpd is
It does work on 7.0.2
The problem is that it can't find
Shared object "libpython2.7.so.1.0" not found
That comes from ntpc.so
-bash-4.3$ ldd /usr/local/lib/python2.7/site-packages/ntp/ntpc.so
/usr/local/lib/python2.7/site-packages/ntp/ntpc.so:
-lpython2.7.1.0 => not found
-lutil
More info on this quirk...
The problem is that this line from ntpd/version.h doesn't compile because
True isn't defined.
> const int VCS_WC_MODIFIED = True;
We call autorevision.sh twice. They make slightly different versions of the
cache file. The python version has True. The c/header v
301 - 400 of 2105 matches
Mail list logo
