:: Re: [DNG] Why Debian 8 Pinning is (…
Page principale
Supprimer ce message
Répondre à ce message
Auteur: Rick Moen
Date:  
À: dng
Sujet: Re: [DNG] Why Debian 8 Pinning is (or isn't) pointless
Quoting Rainer Weikusat (rweikusat@???):

> The purpose of libsystemd0 is to enable packages whose code has been
> 'enhanced' with spurious systemd depedencies to work on systemd-less
> systems. That's absolutely not harmless.


Your implied concept of 'purpose' is IMO a bit problematically abstract
and teleological for my taste. What I find more connected to the real
world, clearer in reference to my experience, is the concept of
_function_, i.e., what things do.

As has been abundantly documented, without systemd itself present,
/lib/[$ARCH]/libsystemd.so.0 does basically nothing at all, as it cannot
do anything. (I'd rather have it gone, and have suggested some ways to
do so, which should be tested before use on production systems.)

If you insist on talking, by contrast, about 'purpose', I'd say that the
purpose of every piece of software on my server is to support my and my
users' computing. The codebases are of course severely imperfect at
that, because code is imperfect. If you disagree and think they have
different purposes, I'll just stoutly assert that it's because you hew
to a different religion, while I profess Me-ism. ;->

Be careful where attribution of 'purpose' to code takes you:

[rick@linuxmafia]
~ $ ldd $(which ssh)
        linux-gate.so.1 =>  (0xb77b0000)
        libresolv.so.2 => /lib/i686/cmov/libresolv.so.2 (0xb7736000)
        libcrypto.so.0.9.8 => /usr/lib/i686/cmov/libcrypto.so.0.9.8 (0xb75de000)
        libdl.so.2 => /lib/i686/cmov/libdl.so.2 (0xb75da000)
        libz.so.1 => /usr/lib/libz.so.1 (0xb75c6000)
        libgssapi_krb5.so.2 => /usr/lib/libgssapi_krb5.so.2 (0xb7597000)
        libc.so.6 => /lib/i686/cmov/libc.so.6 (0xb7450000)
        /lib/ld-linux.so.2 (0xb77b1000)
        libkrb5.so.3 => /usr/lib/libkrb5.so.3 (0xb739e000)
        libk5crypto.so.3 => /usr/lib/libk5crypto.so.3 (0xb737b000)
        libcom_err.so.2 => /lib/libcom_err.so.2 (0xb7378000)
        libkrb5support.so.0 => /usr/lib/libkrb5support.so.0 (0xb7371000)
        libkeyutils.so.1 => /lib/libkeyutils.so.1 (0xb736d000)
        libpthread.so.0 => /lib/i686/cmov/libpthread.so.0 (0xb7354000)
[rick@linuxmafia]
~ $ 


Is the 'purpose' of /usr/bin/ssh being dependent on sonames
libgssapi_krb5.so.2 and libkrb5.so.3 on my Kerberos-less server to
enable packages whose code has been 'enhanced' with spurious Kerberos
dependencies to work on Kerberos-less systems? Is the 'purpose' of
/usr/bin/ssh being dependent on soname the enabling of packages whose
code has been 'enhanced' with spurious kernel keyutils key-retention
functions to work on keyutils-less systems?

It's all bit high-flown for me, and a bit too much like theology.
Personally, I would rather just be concerned with what things _do_.

And, of course, with process: In my experience, packagers often build
into dynamic binaries calls to libraries that somebody, somewhere
_might_ use. E.g., I doubt that the packager of /usr/bin/ssh is a
secret agent attempting to sneak Kerberos into everyone's computing.

But you knew all that, right?