On Tue, Mar 05, 2024 at 12:35:33PM +0900, Tatsuki Makino wrote: > I recently saw a similar problem and thought I had looked into something, but > this was it. > https://lists.freebsd.org/archives/freebsd-ports/2024-February/005546.html > > David Wolfskill wrote on 2024/03/05 10:27: > > albert(14.0-S)[14] pdf2ps pdf/bike.pdf >/tmp/bike.ps > > ld-elf.so.1: /usr/local/lib/libidn.so.12: version LIBIDN_1.0 required by > > /usr/local/lib/libgs.so.10 not defined > .... > It seems that not only fstrm but also libidn contained the same problem.
Indeed. And (as far as I know), perhaps others.
> This would seem to indicate that there has been a change on the part of the
> OS that reverses the availability of version-script in your environment.
> However, since the package version has not changed, the actual package used
> would not have been reinstalled.
> And since a version-script availability was reversed at the same time as some
> dependent package was rebuilt, all packages created in poudriere were assumed
> to be so.
>
> I guess that's how it is :)
>
> For its part, configure seems to output the following line
>
> checking if LD -Wl,--version-script works... yes
>
> If you are still keeping a large amount of poudriere logs, can you check to
> see if there is a mix of yes and no on the lines?
>
> Regards.
> ....
I am, yeah. :-}
And running:
grep -r 'checking if LD -Wl,--version-script works' 20*
in the parent directory of the logs (from 2023-11-26_15h39m37s -
2024-03-03_11h45m10s) shows "yes" responses consistently through:
2023-12-31_11h44m48s/logs/ImageMagick6-nox11-6.9.12.77_7,1.log:checking if LD
-Wl,--version-script works... yes
the next entry is:
2024-01-21_13h15m44s/logs/gnutls-3.7.10.log:checking if LD -Wl,--version-script
works... no
and all of the subsequent entries are "no".
Summarizing:
33 yes
43 no
76 total
For the last "yes" entry, the build machine was running:
FreeBSD 14.0-STABLE #33 stable/14-n266091-1e9df419f14c: Sat Dec 30 11:51:06 UTC
2023
[email protected]:/common/S1/obj/usr/src/amd64.amd64/sys/GENERIC
amd64 1400503 1400503
For the first "no" entry, it was running:
FreeBSD 14.0-STABLE #34 stable/14-n266092-ab71e3480921: Mon Jan 1 11:34:00 UTC
2024
[email protected]:/common/S1/obj/usr/src/amd64.amd64/sys/GENERIC
amd64 1400503 1400503
(I was away from home from 01 Jan - 20 Jan; I did not attempt to update
anything during that time.)
The next record of a "uname string" on the build machine is:
FreeBSD 14.0-STABLE #36 stable/14-n266466-9d29be87d86c: Mon Jan 22 11:52:14 UTC
2024
[email protected]:/common/S1/obj/usr/src/amd64.amd64/sys/GENERIC
amd64 1400506 1400506
Peace,
david
--
David H. Wolfskill [email protected]
Alexey Navalny was a courageous man; Putin has made him a martyr.
See https://www.catwhisker.org/~david/publickey.gpg for my public key.
signature.asc
Description: PGP signature
