Unannounced soname bump (Rawhide): net-snmp (libnetsnmp.so.30 ->
libnetsnmp.so.35)
by Adam Williamson
net-snmp was updated from 5.7.3-41 to 5.8-1 in Rawhide on 2018-07-20.
This update bumped the soname of libnetsnmp (in the net-snmp-libs
subpackage) from libnetsnmp.so.30 to libnetsnmp.so.35.
This soname bump was not announced, as it is supposed to be.
This library has many dependencies, and from a quick survey, few or
none of them appears to have been rebuilt. Here's the list of all
packages depending on libnetsnmp.so.30 from the current compose:
OpenIPMI-0:2.0.25-6.fc29.x86_64
collectd-snmp-0:5.8.0-14.fc29.x86_64
corosync-0:2.99.3-1.fc29.x86_64
cyrus-imapd-0:3.0.7-8.fc29.x86_64
foghorn-0:0.1.6-20.fc29.x86_64
ifstat-0:1.1-27.fc29.x86_64
keepalived-0:2.0.5-2.fc29.x86_64
lldpd-0:1.0.1-2.fc29.x86_64
mbrowse-0:0.4.3-17.fc29.x86_64
nagios-plugins-snmp-disk-proc-0:1.3.1-7.fc29.x86_64
ntop-0:5.0.1-15.fc28.x86_64
nut-0:2.7.4-18.fc29.x86_64
openhpi-0:3.8.0-3.fc29.x86_64
openhpi-libs-0:3.8.0-3.fc29.x86_64
openhpi-subagent-0:2.3.4-35.fc29.x86_64
opensips-snmpstats-0:2.3.4-2.fc29.x86_64
php-snmp-0:7.2.8-1.fc29.x86_64
ptpd-0:2.3.1-12.235e9b4.fc29.x86_64
quagga-0:1.2.2-4.fc29.x86_64
rsyslog-snmp-0:8.36.0-1.fc29.x86_64
tog-pegasus-libs-2:2.14.1-44.fc29.x86_64
tog-pegasus-test-2:2.14.1-44.fc29.x86_64
zabbix-proxy-mysql-0:3.0.16-1.fc29.x86_64
zabbix-proxy-pgsql-0:3.0.16-1.fc29.x86_64
zabbix-proxy-sqlite3-0:3.0.16-1.fc29.x86_64
zabbix-server-mysql-0:3.0.16-1.fc29.x86_64
zabbix-server-pgsql-0:3.0.16-1.fc29.x86_64
These and their dependencies are now all broken.
Once again, folks, *please* announce your soname bumps, and co-ordinate
rebuilds.
--
Adam Williamson
Fedora QA Community Monkey
IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net
http://www.happyassassin.net
5 years, 10 months
Intent to retire denyhosts
by Jason L Tibbitts III
Denyhosts is a daemon which watches for failed ssh login attempts and
blocks them. I have long maintained denyhosts in Fedora and while I did
not originally agree with it being branched to EPEL, I have done light
maintenance on EPEL7 as well. The EPEL6 branch is, however, very old.
The upstream project is mostly moribund. The software does not support
reading from the systemd journal; I spent some effort trying to make
that work but the core logic is very poorly suited to doing it properly.
It did sort of work but was never in an upstreamable state.
The software also primarily works by modifying /etc/hosts.deny, which
was rendered useless when tcp_wrappers support was removed from our
openssh packaging. It can support iptables, but doesn't properly
support firewalld and in any case requires manual configuration to set
this up.
In general, fail2ban is simply a far better choice. The primary feature
it appears to lack is the ability to synchronize lists of blocked hosts
between machines. (And I could be quite wrong about that.)
My intent is to retire denyhosts in rawhide, EPEL7 and EPEL6 in a week,
but I will happily hand it over to someone who wishes to maintain it
properly moving forward. I may even be willing to stay on as a
comaintainer in that case, at least for a bit. If you do wish to take
over maintenance, there are a few open bugs and an update to 3.0 or
3.1beta (released in 2015) would probably be needed, as well as a switch
to python3.
- J<
5 years, 10 months
Correcting licenses in several packages I maintain
by Zdenek Dohnal
Hi,
I reviewed licenses in files for my packages, so I added several
corrections of license tags:
cups:
- from GPLv2
- to GPLv2+ and LGPLv2+ with exceptions and AML
cups-filters:
- from GPLv2 and GPLv2+ and GPLv3 and GPLv3+ and LGPLv2+ and MIT
- to GPLv2 and GPLv2+ and GPLv3 and GPLv3+ and LGPLv2+ and MIT and BSD
with advertising
hplip:
- from GPLv2+ and MIT and BSD
- to GPLv2+ and MIT and BSD and IJG and Public Domain and GPLv2+ with
exceptions and ISC
qpdf:
- from Artistic 2.0
- to (Artistic 2.0 or ASL 2.0) and MIT
sane-backends:
- from GPLv2+ and GPLv2+ with exceptions and Public Domain
- to GPLv2+ and GPLv2+ with exceptions and Public Domain and IJG and
LGPLv2+ and MIT
sane-frontends:
- from GPLv2+
- to GPLv2+ and LGPLv2+ and GPLv2+ with exceptions
vim:
- from Vim
- to Vim and MIT
--
Zdenek Dohnal
Associate Software Engineer
Red Hat Czech - Brno TPB-C
5 years, 10 months
build failed on s390x, other archs building okay
by Kaleb S. KEITHLEY
Is this a spurious failure or ???
...
Task info: https://koji.fedoraproject.org/koji/taskinfo?taskID=28564187
...
buildArch (glusterfs-4.1.2-1.fc29.1.src.rpm, s390x): free -> FAILED:
Fault: <Fault 1: 'Traceback (most recent call last):\n File
"/usr/lib/python2.7/site-packages/koji/daemon.py", line 1244, in
runTask\n response = (handler.run(),)\n File
"/usr/lib/python2.7/site-packages/koji/tasks.py", line 307, in run\n
return koji.util.call_with_argcheck(self.handler, self.params,
self.opts)\n File "/usr/lib/python2.7/site-packages/koji/util.py", line
209, in call_with_argcheck\n return func(*args, **kwargs)\n File
"/usr/sbin/kojid", line 1195, in handler\n fn =
self.localPath("work/%s" % pkg)\n File
"/usr/lib/python2.7/site-packages/koji/tasks.py", line 477, in
localPath\n fsrc = six.moves.urllib.request.urlopen(url)\n File
"/usr/lib64/python2.7/urllib2.py", line 154, in urlopen\n return
opener.open(url, data, timeout)\n File
"/usr/lib64/python2.7/urllib2.py", line 435, in open\n response =
meth(req, response)\n File "/usr/lib64/python2.7/urllib2.py", line 548,
in http_response\n \'http\', request, response, code, msg, hdrs)\n
File "/usr/lib64/python2.7/urllib2.py", line 473, in error\n return
self._call_chain(*args)\n File "/usr/lib64/python2.7/urllib2.py", line
407, in _call_chain\n result = func(*args)\n File
"/usr/lib64/python2.7/urllib2.py", line 556, in http_error_default\n
raise HTTPError(req.get_full_url(), code, msg, hdrs, fp)\nHTTPError:
HTTP Error 503: Backend fetch failed\n'>
thanks,
--
Kaleb
5 years, 10 months
RFC: Pass --auto-features=enabled in meson
by Igor Gnatenko
Hello,
meson 0.47.0 has new option type - feature
<https://mesonbuild.com/Release-notes-for-0-47-0.html#new-type-of-build-op...>
(tri-state option - enabled/disabled/auto). It is quite common that in
autofoo and less in cmake worlds people rely on auto-detection of
dependencies and enabling features based on those.
I believe that for distribution we should make sure that all default
features are enabled and if not, packager should explicitly disable
feature. For instance when I was testing some RPM patches for new
compression types, it was disabling some default feature because the
configure script was written in a wrong way.
It would be nice if we could have some standardized way of specifying
options for buildsystems which would convert into autofoo/cmake/meson way
of specifying parameters but this is topic for another discussion.
Thoughts? Objections?
Unless there is strong opposition, I'm going to switch this parameter in
%meson macro and you would be able to change its value by redefining
%__meson_auto_features.
--
-Igor Gnatenko
5 years, 10 months
In the OpenShift Origin/CRI-O/Kubernetes effort we have a dilemma.
by Daniel Walsh
Users of OpenSHift Origin require CRI-O 1.10 right now. But Kubernetes
users want to try out the latest packages for kubernetes 1.11 which
would require CRI-O 1.11. Origin might not be ready to move to
Kubernetes 1.11 for a while.
Bottom line we want to be able to ship CRI-0 1.10.* and CRI-O 1.11.*
releases in the same Fedora 28.
I believe this is what Modularity was designed to fix.
Can I do this with Modularity? If I can how do I use fedpkg to make
this happen?
Dan
5 years, 10 months
