The debian-private mailing list leak, part 1. Volunteers have complained about Blackmail. Lynchings. Character assassination. Defamation. Cyberbullying. Volunteers who gave many years of their lives are picked out at random for cruel social experiments. The former DPL's girlfriend Molly de Blanc is given volunteers to experiment on for her crazy talks. These volunteers never consented to be used like lab rats. We don't either. debian-private can no longer be a safe space for the cabal. Let these monsters have nowhere to hide. Volunteers are not disposable. We stand with the victims.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: DPKG SHOWSTOPPER! (need confirmation and additional analysys)

Michael Alan Dorman <mdorman@calder.med.miami.edu> writes:

> I think I have a solution, etc., but I would like some additional eyes
> on this.

Your solution is a little incorrect.

> void assertpredep(const char *const *argv) {
>   static struct versionrevision predepversion = {~0UL,0,0};
>   struct pkginfo *pkg;
>   
>   if (*argv) badusage("--assert-support-predepends does not take any arguments");
> 
>   modstatdb_init(admindir,msdbrw_readonly);
>   pkg= findpackage("dpkg");
>   if (!predepversion.epoch == ~0UL) {
>     predepversion.epoch= 0;
>     predepversion.version= nfstrsave("1.1.0");
>     predepversion.revision= 0;
>   }

[...]

> So, it seems to me (and testing seems to confirm this) that the proper
> solution is to change the line:
> 
>   if (!predepversion.epoch == ~0UL) {
> 
> to:
> 
>   if (!predepversion.epoch == 0UL) {

No, it should be
    if (predepversion.epoch == ~0UL) {

The function initializes the static predepversion with bogus data, so
the first time through the function it wants to set the data elements
correctly.  The test is checking for the bogus epoch, so it should be
what I've listed above.  It doesn't really matter; your solution
happens to work also.

Anyway, I'm putting things into rex-updates for a 1.2.4 release on
tomorrow's cron run, so I will go ahead and build a dpkg 1.4.0.7 with
the following patch and your corrected preinst.


Guy





>  which will then result in the info on the version supporting
> pre-depends getting set correctly.  So, could anyone with some free
> time try and verify my suppositions?  Anyone who has time to try and
> figure out whether the change to versionsatisfied3 that enable
> epochs is going to break anything else would be a godsend.  I'm
> going to be building a dpkg-1.4.0.7 and making it available at:
> ftp://debian.med.miami.edu/pub/mdorman/ It will have the bug fix in
> it, as well as a slightly modified preinst which will bypass the
> --assert-supports-predepends test, which should allow it to install
> on systems with 1.4.0.6.  I'd appreciate any adventurous testers
> having a go with this.  Cheers, Mike.  -- Please respect the
> confidentiality of material on the debian-private list.  TO
> UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
> debian-private-REQUEST@lists.debian.org . Trouble? e-mail to
> Bruce@Pixar.com


--
Please respect the confidentiality of material on the debian-private list.
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-private-REQUEST@lists.debian.org . Trouble? e-mail to Bruce@Pixar.com