Hello Bruno!
A few days ago on Fedora devel@ list, the ancient and stalled Review Request
of mindi-busybox has been pointed at. Together with the hint that the
submitter is still waiting to be sponsored since five years. That's also
the reason I'm posting a copy of this message to devel@ list.
When I've had a first look in bugzilla, I've discovered that the ticket
has not been displayed properly in the review queue, because it was assigned
to a reviewer via the "fedora-review" flag. The review request tracker
displays such tickets on a separate page only,
http://fedoraproject.org/PackageReviewStatus/
and potential reviewers as well as sponsors assume somebody else is
working on the review. There are various reasons why a review may have
stalled. Only occasionally somebody makes extra effort to recheck aging
tickets. Here, there has been no progress in that ticket at all until
recently when it has been discovered.
As a next step, I've located a few related review requests, since there
are inter-dependencies. I've found:
mindi : https://bugzilla.redhat.com/187317
mondo : https://bugzilla.redhat.com/187318
mindi-busybox : https://bugzilla.redhat.com/476234
buffer : https://bugzilla.redhat.com/759818
There is some confusion, and it's not easy to track down the details:
1) Package "buffer" is a re-review, because it has been retired in 2011
after it FTBFS. The re-review had been assigned to the "buffer" component,
so it has not been displayed as a Review Request anywhere.
2) It has been questioned why it had been retired. Dist git claims:
http://pkgs.fedoraproject.org/cgit/buffer.git/plain/dead.package
| This package was retired on 2011-07-25 due to it being unable to
| build this package for multiple releases (FTBFS).
https://lists.fedoraproject.org/pipermail/scm-commits/2011-July/639234.html
But koji shows a successful build on 2011-02-08 as done by Dennis Gilmore
for the Fedora 15 mass-rebuild:
http://koji.fedoraproject.org/koji/packageinfo?packageID=7297
In July 2011, it has been listed as "orphan",
[ACTION REQUIRED] Retiring packages in F-16 (final warning)
https://lists.fedoraproject.org/pipermail/devel/2011-July/154622.html
and I could not find out its status at that time. Pkgdb activity gets
logged to scm-commits@ list, so perhaps somebody else has more luck.
3) In koji, dist git and pkgdb, I've found that user "bcornec" has been
an approved packager before *and* even the maintainer of that package:
[pkgdb] buffer was added for bcornec
https://lists.fedoraproject.org/pipermail/scm-commits/2008-December/231912.…
an older FTBFS ticket for that maintainer:
https://bugzilla.redhat.com/511589
In the Fedora Account System, user "bcornec" is not a member of the
"packager" group anymore. What has happened there?
4) In the review queue, the review requests have been submitted using three
different e-mail addresses.
What's the status/plan with regard to these package review requests?
Regards,
--
Michael Schwendt (mschwendt)
Fedora release 20 (Heisenbug) - Linux 3.12.0-0.rc6.git1.2.fc21.x86_64
loadavg: 0.04 0.12 0.17
why does Fedora's openssl not list "aesni" compared to RHEL?
the values below are showing it is in fact supported by both, on the
other hand "SSLCryptoDevice aesni" for Apache does not work on Fedora
and i am unsure if it is used automatically by default in that case
in doubt this may waste factor 8 in performance on modern hardware
http://httpd.apache.org/docs/2.4/mod/mod_ssl.html#sslcryptodevice
____________________________________________
cat /etc/redhat-release; openssl engine; cat /proc/cpuinfo | grep "model name"
CentOS release 6.4 (Final)
(aesni) Intel AES-NI engine
(dynamic) Dynamic engine loading support
model name : Intel(R) Xeon(R) CPU E5640 @ 2.67GHz
openssl speed aes-256-cbc
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
aes-256 cbc 58003.13k 62866.47k 63308.37k 135122.94k 135858.86k
openssl speed -evp aes-256-cbc
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
aes-256-cbc 460493.45k 508006.53k 517264.90k 519457.45k 520594.75k
____________________________________________
cat /etc/redhat-release; openssl engine; cat /proc/cpuinfo | grep "model name"
Fedora release 18 (Spherical Cow)
(dynamic) Dynamic engine loading support
model name : Intel(R) Xeon(R) CPU E5640 @ 2.67GHz
openssl speed aes-256-cbc
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
aes-256 cbc 58347.04k 63168.39k 64110.99k 135981.36k 137257.23k
openssl speed -evp aes-256-cbc
type 16 bytes 64 bytes 256 bytes 1024 bytes 8192 bytes
aes-256-cbc 462203.14k 512258.69k 519790.56k 522836.19k 523612.23k