Re: ✘Python 2.7 broken
Richard Laager
rlaager at wiktel.com
Sun Dec 13 02:04:45 UTC 2020
On 12/12/20 7:53 PM, Gary E. Miller via devel wrote:
> On Sat, 12 Dec 2020 19:46:22 -0600
> Richard Laager via devel <devel at ntpsec.org> wrote:
>
>> On 12/12/20 7:07 PM, Gary E. Miller via devel wrote:
>>> NTPsec git head broke Python 2.7, badly.
>>
>> I believe James B is looking at test failures under certain
>> conditions, but I've never been able to reproduce them. I wonder if
>> this is that issue. Do the installed binaries break in the same way
>> when run normally, or are they only broken at the test step?
>
> Installed python files run fine. The pure python module runs fine
> installed.
So only broken during the test? That sounds like what James B was
looking into. I can't speak intelligently to that issue without more work.
I probably can help, at least somewhat, with this:
> ntpc.so still getting
> installed in the wrong place (/usr/local/lib64/python2.7/site-packages)
I agree that's not right. It should be something like /usr/local/lib64.
I just pushed a tiny change to master:
commit eee4607b256a5fb4263bd83b75f311e36be3a9ad
Author: Richard Laager <rlaager at wiktel.com>
Date: Sat Dec 12 20:02:10 2020 -0600
configure: Print LIBDIR
This may be helpful for debugging install issues.
Signed-off-by: Richard Laager <rlaager at wiktel.com>
diff --git a/wscript b/wscript
index 3b4eb869c..2db41c581 100644
--- a/wscript
+++ b/wscript
@@ -886,6 +886,7 @@ int main(int argc, char **argv) {
msg_setting("LDFLAGS", " ".join(ctx.env.LDFLAGS))
msg_setting("LINKFLAGS_NTPD", " ".join(ctx.env.LINKFLAGS_NTPD))
msg_setting("PREFIX", ctx.env.PREFIX)
+ msg_setting("LIBDIR", ctx.env.LIBDIR)
msg_setting("Droproot Support", droproot_type)
msg_setting("Debug Support", yesno(ctx.options.enable_debug))
msg_setting("Refclocks", ", ".join(sorted(ctx.env.REFCLOCK_LIST)))
Can you run that and see what LIBDIR is set to for you at configure time?
> Since the file is not in the right place ldconfig does not
> know to look there. Nor should it.
Agreed.
--
Richard
More information about the devel
mailing list