Re: Backwards-incompatible RPM format change in Fedora 34?

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Pavel Raiskup <praiskup@xxxxxxxxxx> writes:

> On Monday, January 25, 2021 7:50:08 PM CET Dan Čermák wrote:
>> Pavel Raiskup <praiskup@xxxxxxxxxx> writes:
>> 
>> > On Monday, January 25, 2021 8:18:33 AM CET Panu Matilainen wrote:
>> >> On 1/22/21 8:19 PM, Adam Williamson wrote:
>> >> > On Fri, 2021-01-22 at 09:57 +0200, Panu Matilainen wrote:
>> >> >> On 1/21/21 8:37 PM, Adam Williamson wrote:
>> >> >>> On Thu, 2021-01-21 at 10:53 +0100, Kevin Kofler via devel wrote:
>> >> >>>> Florian Weimer wrote:
>> >> >>>>> With rpm-4.15.1-3.fc32.1.x86_64, I get this error:
>> >> >>>>>
>> >> >>>>> $ rpm -qip
>> >> >>>>> https://dl.fedoraproject.org/pub/fedora/linux/development/rawhide/Everything/aarch64/debug/tree/Packages/m/ModemManager-debugsource-1.14.10-1.fc34.aarch64.rpm
>> >> >>>>> error: /var/tmp/rpm-tmp.6iU66n: signature hdr data: BAD, no. of
>> >> >>>>> bytes(88084) out of range error:
>> >> >>>>> https://dl.fedoraproject.org/pub/fedora/linux/development/rawhide/Everything/aarch64/debug/tree/Packages/m/ModemManager-debugsource-1.14.10-1.fc34.aarch64.rpm:
>> >> >>>>> not an rpm package (or package manifest)
>> >> >>>>>
>> >> >>>>> Is this expected?
>> >> >>>>>
>> >> >>>>> It seems that rpm-4.16.1.2-1.fc33.x86_64 can parse the RPM just fine.
>> >> >>>>> But rpm-4.14.3-4.el8.x86_64 does not like it, either.
>> >> >>>>
>> >> >>>> Considering that direct upgrades from F32 to F34 (n to n+2) are supposed to
>> >> >>>> be supported, this sounds like a blocker to me.
>> >> >>>
>> >> >>> openQA N+2 upgrade tests have indeed been running into this for a few
>> >> >>> days:
>> >> >>>
>> >> >>> https://openqa.fedoraproject.org/tests/759545#step/upgrade_run/20
>> >> >>>
>> >> >>> I had been meaning to dig into it a bit more before filing a bug.
>> >> >>>
>> >> >>
>> >> >> Folks, when rpm starts spitting errors like that, don't think, just file
>> >> >> a bug. It's very, very very very unlikely that it's "ok" in any
>> >> >> imaginable meaning.
>> >> > 
>> >> > It's not that I thought it was "OK", it's just that these days I tend
>> >> > to like filing a bug report with detailed cause analysis and stuff all
>> >> > wrapped up :)
>> >> 
>> >> And that is certainly appreciated!
>> >> 
>> >> But if there's even a wiff of a package generational bug, it's better to 
>> >> act first and think later because those things are not entirely unlike a 
>> >> virus outbreak, those buggers spread fast on every sneeze and stopping 
>> >> it early is the key to damage control :D
>> >
>> > I'm curious how are we going to fix this?  Mock started to complain now that it
>> > is not even able to install rawhide bootstrap chroot on F32:
>> > error: /var/lib/mock/fedora-rawhide-x86_64-bootstrap-1611584804.803151/root/var/cache/dnf/fedora-2d95c80a1fa0a67d/packages/python3-libs-3.9.1-3.fc34.x86_64.rpm: signature hdr data: BAD, no. of bytes(384156) out of range
>> >
>> > Does it imply rebuild of all affected packages, including Python3.9?
>> 
>> No, I've been seeing this issue for the past week and thanks to Fabio
>> I've been able to solve it by running
>> mock -r /etc/mock/fedora-rawhide-x86_64.cfg --clean --scrub=all
>
> Cleaning caches would make sense if there was newer variant of that package in
> the repository.  But in this case it wasn't.  Rebuild was really needed to fix
> the bootstrap installation.
>
> Fortunately, according to Miro Hrončok, this was (Miro already rebuilt the
> package, and it is going to be added in the next compose) the only "broken"
> package affecting bootstrap installation so nothing else needs an immediate
> rebuild now.
>
> As I was informed, there is upcoming mass rebuild, so any residual packages would
> be fixed automatically very soon (in case anyone wants to have mock bootstrap
> disabled for any reason).
>
> What is IMO guaranteed to work-around this issue is to use
> mock --use-bootstrap-image feature.

Interesting,

because I had *exactly* the same issue yesterday evening and using
bootstrap image did not fix the issue. However, cleaning up every *did*,
albeit I was on a Fedora 33 host and not on 32.


Cheers,

Dan

Attachment: signature.asc
Description: PGP signature

_______________________________________________
devel mailing list -- devel@xxxxxxxxxxxxxxxxxxxxxxx
To unsubscribe send an email to devel-leave@xxxxxxxxxxxxxxxxxxxxxxx
Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
List Archives: https://lists.fedoraproject.org/archives/list/devel@xxxxxxxxxxxxxxxxxxxxxxx

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Index of Archives]     [Fedora Announce]     [Fedora Users]     [Fedora Kernel]     [Fedora Testing]     [Fedora Formulas]     [Fedora PHP Devel]     [Kernel Development]     [Fedora Legacy]     [Fedora Maintainers]     [Fedora Desktop]     [PAM]     [Red Hat Development]     [Gimp]     [Yosemite News]

  Powered by Linux