Corrupt RPM package in Fedora 31 compose: perf-debuginfo-5.5.9-200.fc31.x86_64

[Florian Weimer]

Installation fails like this:

Running transaction
  Preparing        :                                                       1/1 
  Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 
Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
  Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 

The RPM headers look okay, but the payload seems to be missing
completely.

perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
this might be a deterministic issue. 8-(

Any suggestions how to debug this further?

Thanks,
Florian

[Petr Pisar]

On Mon, Apr 06, 2020 at 12:14:23PM +0200, Florian Weimer wrote:
>  Installation fails like this:
>  
>  Running transaction
>    Preparing        :                                                       1/1 
>    Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 
>  Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
>    Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 
>  
>  The RPM headers look okay, but the payload seems to be missing
>  completely.
>  
>  perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
>  this might be a deterministic issue. 8-(
>  
>  Any suggestions how to debug this further?
>   
Downloading the package from Koji
<https://kojipkgs.fedoraproject.org//packages/kernel-tools/5.5.15/200.fc31...
hangs. It looks like a storage failure and a compose process just did not
verify that the download finished successfully and copied the incomplete
package to the repository.

I recommend you reporting it to the Fedora Infrastracture.

-- Petr

[Florian Weimer]

* Petr Pisar:

>  On Mon, Apr 06, 2020 at 12:14:23PM +0200, Florian Weimer wrote:
> > Installation fails like this:
> > 
> > Running transaction
> >   Preparing        :                                                       1/1 
> >   Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 
> > Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
> >   Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1 
> > 
> > The RPM headers look okay, but the payload seems to be missing
> > completely.
> > 
> > perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
> > this might be a deterministic issue. 8-(
> > 
> > Any suggestions how to debug this further?
> > 
>  Downloading the package from Koji
>
> <https://kojipkgs.fedoraproject.org//packages/kernel-tools/5.5.15/200.fc31...
>  hangs. It looks like a storage failure and a compose process just did not
>  verify that the download finished successfully and copied the incomplete
>  package to the repository.
>
>  I recommend you reporting it to the Fedora Infrastracture. 
Done: https://pagure.io/fedora-infrastructure/issue/8816

Thanks,
Florian

[Panu Matilainen]

On 4/6/20 2:06 PM, Petr Pisar wrote:
>  On Mon, Apr 06, 2020 at 12:14:23PM +0200, Florian Weimer wrote:
> > Installation fails like this:
> >
> > Running transaction
> >    Preparing        :                                                       1/1
> >    Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1
> > Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
> >    Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1
> >
> > The RPM headers look okay, but the payload seems to be missing
> > completely.
> >
> > perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
> > this might be a deterministic issue. 8-(
> >
> > Any suggestions how to debug this further?
> >
>  Downloading the package from Koji
>
> <https://kojipkgs.fedoraproject.org//packages/kernel-tools/5.5.15/200.fc31...
>  hangs. It looks like a storage failure and a compose process just did not
>  verify that the download finished successfully and copied the incomplete
>  package to the repository.
>  
>  I recommend you reporting it to the Fedora Infrastracture. 
It's not just an interrupted download because the header+payload md5sum 
is calculated on this very payload, the above install attempt would've 
stopped at pre-transaction verification if that wasn't the case.

To double-check:

[pmatilai🎩︎lumikko Downloads]$ rpm -Kv 
perf-debuginfo-5.5.15-200.fc31.x86_64.rpmperf-debuginfo-5.5.15-200.fc31.x86_64.rpm:
     Header SHA256 digest: OK
     Header SHA1 digest: OK
     Payload SHA256 digest: OK
     MD5 digest: OK

So whatever happened, the corruption took place at the time of rpmbuild 
already.

	- Panu -

[Stephen John Smoogen]

On Mon, 6 Apr 2020 at 07:07, Petr Pisar <ppisar(a)redhat.com&gt; wrote:

>  On Mon, Apr 06, 2020 at 12:14:23PM +0200, Florian Weimer wrote:
>  > Installation fails like this:
>  >
>  > Running transaction
>  >   Preparing        :
>     1/1
>  >   Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64
>     1/1
>  > Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
>  >   Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64
>     1/1
>  >
>  > The RPM headers look okay, but the payload seems to be missing
>  > completely.
>  >
>  > perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
>  > this might be a deterministic issue. 8-(
>  >
>  > Any suggestions how to debug this further?
>  >
>  Downloading the package from Koji
>  <
>
> https://kojipkgs.fedoraproject.org//packages/kernel-tools/5.5.15/200.fc31...
>  >
>  hangs. It looks like a storage failure and a compose process just did not
>  verify that the download finished successfully and copied the incomplete
>  package to the repository.
>
>  I recommend you reporting it to the Fedora Infrastracture.
>
>  
I have checked all rpms in the directory and they are reporting OK

[root@koji01 x86_64][PROD]# rpm -Kv *perf*
perf-5.5.15-200.fc31.x86_64.rpm:
    Header SHA1 digest: OK (35eb95ffadc8e5458ce31659a9e2d6c4c366fc41)
    MD5 digest: OK (c49edfefeb54e48492e2b060a343dd1f)
perf-debuginfo-5.5.15-200.fc31.x86_64.rpm:
    Header SHA1 digest: OK (1d4f176b5fda5d5dd5bc72640f99d06d16bf6996)
    MD5 digest: OK (1ed09e9c13d646498bc3d69930468f6b)
python3-perf-5.5.15-200.fc31.x86_64.rpm:
    Header SHA1 digest: OK (d2e9d2a7937ac900a0d257c4413bbd334e27d0a9)
    MD5 digest: OK (6b22d1d9074e17126832b5b1e97510fa)
python3-perf-debuginfo-5.5.15-200.fc31.x86_64.rpm:
    Header SHA1 digest: OK (12e6aca80fcc96e5239f1eecdca4484c56f9206a)
    MD5 digest: OK (ccba53cc44adf1aca7170a681d5f69a1)
[root@koji01 x86_64][PROD]# rpm -q rpm
rpm-4.11.3-43.el7.x86_64

I do not know why kojipkgs is screwed up at the moment.. but I am working
up the layers




>  -- Petr
>  _______________________________________________
>  devel mailing list -- devel(a)lists.fedoraproject.org
>  To unsubscribe send an email to devel-leave(a)lists.fedoraproject.org
>  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/[e-mail]
>  

-- 
Stephen J Smoogen.

[Panu Matilainen]

On 4/6/20 1:14 PM, Florian Weimer wrote:
>  Installation fails like this:
>  
>  Running transaction
>     Preparing        :                                                       1/1
>     Installing       : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1
>  Error unpacking rpm package perf-debuginfo-5.5.15-200.fc31.x86_64
>     Verifying        : perf-debuginfo-5.5.15-200.fc31.x86_64                 1/1
>  
>  The RPM headers look okay, but the payload seems to be missing
>  completely.
>  
>  perf-debuginfo-5.5.9-200.fc31.x86_64 seems to be affected as well, so
>  this might be a deterministic issue. 8-( 
As corrupt rpm packages tend to get people edgy, here's the short 
summary of what this was about:

kernel-tools.spec had inherited a manual perf-debuginfo sub-package from 
the kernel.spec, but unlike the kernel itself it has automatic debuginfo 
generation enabled. These silently clash and cause said debuginfo 
package tobe written twice, the result being a timing dependent coin-toss.

So the rpm bug is that it doesn't cleanly error out on this situation, 
but it's nothing more serious than a funny little corner-case with 
packages doing their own debuginfo sub-packaging.

	- Panu -