Showing posts with label yum. Show all posts
Showing posts with label yum. Show all posts

Monday, July 8, 2013

yum error

I just got the following error while trying to update the system:
# yum update
Loaded plugins: langpacks, refresh-packagekit
adobe-linux-x86_64                             |  951 B  00:00:00
fedora/19/x86_64/metalink                      |  25 kB  00:00:00
Could not parse metalink https://mirrors.fedoraproject.org/metalink?repo=fedora-19&arch=x86_64 error was
No repomd file
Error: File /var/cache/yum/x86_64/19/fedora/metalink.xml does not exist
As seen from the output, the problem is with some Fedora mirror. After googling for a solution and trying different fixes, none of which worked for me, I finally edited /etc/yum.repos.d/fedora.repo, commented out the line mirrorlist and uncommented baseurl line. After that, everything worked.

To conclude this post, the problem is with some mirror picked by yum.

Monday, November 12, 2012

Do you want to connect to IPv6 Internet in a minute or so?

Well, I just learned a very quick way to connect to IPv6 Internet that really works! That is, if you have Fedora 17, but probably for other distributions it is equally easy. Here are two commands to execute that will enable IPv6 network connectivity to you personal computer:
yum -y install gogoc
systemctl start gogoc.service
First command installs package gogoc, while the second one starts it. Next time you'll need only start command. After the start command apparently nothing will happen but in a minute or so you'll have working IPv6 connection. Check it out:
# ping6 www.google.com
PING www.google.com(muc03s02-in-x13.1e100.net) 56 data bytes
64 bytes from muc03s02-in-x13.1e100.net: icmp_seq=1 ttl=54 time=54.0 ms
64 bytes from muc03s02-in-x13.1e100.net: icmp_seq=2 ttl=54 time=55.0 ms
^C
--- www.google.com ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1000ms
rtt min/avg/max/mdev = 54.023/54.551/55.080/0.577 ms
As you can see, Google is reachable on IPv6 addresses. You can also try traceroute6:
# traceroute6 www.google.com
traceroute to www.google.com (2a00:1450:4016:801::1011), 30 hops max, 80 byte packets
1 2001:5c0:1400:a::722 (2001:5c0:1400:a::722) 40.097 ms 42.346 ms 45.937 ms
2 ve8.ipv6.colo-rx4.eweka.nl (2001:4de0:1000:a22::1) 47.548 ms 49.498 ms 51.760 ms
3 9-1.ipv6.r2.am.hwng.net (2001:4de0:a::1) 55.613 ms 56.808 ms 60.062 ms
4 2-1.ipv6.r3.am.hwng.net (2001:4de0:1000:34::1) 62.570 ms 65.224 ms 66.864 ms
5 1-3.ipv6.r5.am.hwng.net (2001:4de0:1000:38::2) 72.339 ms 74.596 ms 77.970 ms
6 amsix-router.google.com (2001:7f8:1::a501:5169:1) 80.598 ms 38.902 ms 39.548 ms
7 2001:4860::1:0:4b3 (2001:4860::1:0:4b3) 41.833 ms 2001:4860::1:0:8 (2001:4860::1:0:8) 46.500 ms 2001:4860::1:0:4b3 (2001:4860::1:0:4b3) 48.142 ms
8 2001:4860::8:0:2db0 (2001:4860::8:0:2db0) 51.250 ms 54.204 ms 57.569 ms
9 2001:4860::8:0:3016 (2001:4860::8:0:3016) 64.727 ms 67.339 ms 69.540 ms
10 2001:4860::1:0:336d (2001:4860::1:0:336d) 80.203 ms 82.302 ms 85.290 ms
11 2001:4860:0:1::537 (2001:4860:0:1::537) 87.769 ms 91.180 ms 92.931 ms
12 2a00:1450:8000:1f::c (2a00:1450:8000:1f::c) 61.213 ms 54.156 ms 55.931 ms
It simply can not be easier that that. Using ip command you can check address you were given:
# ip -6 addr sh
1: lo: mtu 16436
    inet6 ::1/128 scope host
       valid_lft forever preferred_lft forever
3: wlan0: mtu 1500 qlen 1000
    inet6 fe80::f27b:cbff:fe9f:a33b/64 scope link
       valid_lft forever preferred_lft forever
5: tun: mtu 1280 qlen 500
    inet6 2001:5c0:1400:a::723/128 scope global
       valid_lft forever preferred_lft forever
And also routes:
# ip -6 ro sh
2001:5c0:1400:a::722 via 2001:5c0:1400:a::722 dev tun  metric 0
    cache
2001:5c0:1400:a::723 dev tun  proto kernel  metric 256  mtu 1280
2a00:1450:4008:c01::bf via 2a00:1450:4008:c01::bf dev tun  metric 0
    cache
2a00:1450:400d:803::1005 via 2a00:1450:400d:803::1005 dev tun  metric 0
    cache
2a00:1450:4013:c00::78 via 2a00:1450:4013:c00::78 dev tun  metric 0
    cache
2a03:2880:2110:cf01:face:b00c:: via 2a03:2880:2110:cf01:face:b00c:: dev tun  metric 0
    cache
2000::/3 dev tun  metric 1
unreachable fe80::/64 dev lo  proto kernel  metric 256  error -101
fe80::/64 dev vmnet1  proto kernel  metric 256
fe80::/64 dev vmnet8  proto kernel  metric 256
fe80::/64 dev wlan0  proto kernel  metric 256
fe80::/64 dev tun  proto kernel  metric 256
default dev tun  metric 1
Probably I don't have to mention that if you open Google in a Web browser you'll be using IPv6. :) In case you don't believe me, try using tcpdump (or wireshark) on tun interface.
You can stop IPv6 network by issuing the following command:
systemctl stop gogoc.service
If you try ping6 and traceroute6 commands after that, you'll receive Network unreachable messages, meaning Google servers can not be reached via their IPv6 address.

Monday, October 29, 2012

yum and fastestmirror plugin...

Few hours ago I lost my nerves because when I started yum to update my system, download was painfully slow, somewhere around 20kB/s. It is outrageous because I was using 100 Mbps link and that is probably the slowest link in the chain that ends up somewhere in GEANT. Thus, things have to be much faster than that! The best speed that can be achieved is somewhere around 50Mbps and what I was getting wasn't even remotely close to it! This wasn't something I was prepared to accept as is, so I decided to see what's happening.

Yum has a plugin, fastestmirror. The purpose of that plugin is to determine the fastest available mirror and makes yum download from it, not some random one. Usually, this plugin works very well, but this time it didn't. I tried to reset everything with
yum clean all
and than again
yum update
But it didn't help. Googling around I quickly determined that the first command didn't remove fastestmirror's data. What is necessary is to remove cache file stored in /var/cache/yum/x86_64/17/timedhosts.txt (this is location on 64-bit Fedora 17). Well, guess what, this didn't help either. Namely, fastestmirror plugin determines which mirror is the best one based on measuring how much time is necessary to establish connection with a mirror, and then it immediately disconnects. This is all OK, until mirror starts to apply some throttling effectively capping maximum speed. And this was exactly what happened to me.

It used to be possible to send SIGINT signal to yum (pressing Ctrl+C) on which yum would switch to another mirror. But this doesn't work any more. When you press Ctrl+C yum exits. Now, this is expected behavior, but the previous one was actually useful! So, there should be some way to tell yum to switch to next mirror.

In the end I solved this by looking which mirror(s) yum was using. This is printed when yum starts, e.g.:

Loading mirror speeds from cached hostfile
 * fedora: gd.tuwien.ac.at
 * fedora-debuginfo: fedora.inode.at
 * rpmfusion-free: mirrors.coreix.net
 * rpmfusion-free-debuginfo: mirrors.coreix.net
 * rpmfusion-free-updates: mirrors.coreix.net
 * rpmfusion-free-updates-debuginfo: mirrors.coreix.net
 * rpmfusion-nonfree: mirrors.coreix.net
 * rpmfusion-nonfree-debuginfo: mirrors.coreix.net
 * rpmfusion-nonfree-updates: mirrors.coreix.net
 * rpmfusion-nonfree-updates-debuginfo: mirrors.coreix.net
 * rpmfusion-nonfree-updates-testing: mirrors.coreix.net
 * rpmfusion-nonfree-updates-testing-debuginfo: rpmfusion.blizoo.mk
 * updates: gd.tuwien.ac.at
 * updates-debuginfo: fedora.intergenia.de
The problem was Fedora's main repository, which was downloaded from gd.tuwien.ac.at. So, I edited fastestmirror's configuration file /etc/yum/pluginconf.d/fastestmirror.conf and added the following line:
exclude=.at
That excluded a bit more mirrors than I intended, but it definitely solved my problem.

Wednesday, August 1, 2012

Google Chrome yum repository...

Yesterday I noticed that yum repository on Google servers with packages for Chrome doesn't work. I see this in yum output when I run update process:
google-chrome                                            |  951 B     00:00 !!!
Note that three exclamation marks at the end! Also, once I received one  other error message, but I didnt' write it down, so I don't remember exactly what it was. Anyway, checking manually URL of the repository shows that it is not available:
# lftp http://dl.google.com/linux/chrome/rpm/stable/x86_64
cd: Access failed: 404 Not Found (/linux/chrome/rpm/stable/x86_64)
But, it turns out that there is no linux repository at all:
# lftp http://dl.google.com/linux/
cd: Access failed: 404 Not Found (/linux)
I did some Googling and what puzzles me is that there is nothing about this change. I found several posts in the past that the repository isn't available, but that was over a year ago, and in the mean time repository was restored. Also, there are some more recent posts, but again, those were solved!

I checked my update logs and found that the last update was on July, 27th, so is it possible that the URL will be restored as it was in the past?

Wednesday, June 6, 2012

Fedora 17, XeLaTeX and Beamer...

I decided to write this post about XeLaTeX and my attempts to process one presentation written using Beamer class into PDF file. The reason is that I certainly lost few hours trying to resolve the set of errors, and maybe someone finds this information/experience helpful.

The first problem I had after starting xelatex was the following error:
! Package pgfsys Error: Driver file ``pgfsys-xetex.def'' not found..

See the pgfsys package documentation for explanation.
Type  H   for immediate help.
 ...                                             
                                                 
l.847 ...ver file ``\pgfsysdriver'' not found.}{}}
                                                 
? X
No pages of output.
So, I tried to find out which package provides this file using yum. But, no luck there. Then, I continued with google-fu. I found many links, but nothing that could be applied to my case, until I stumbled on this solution which basically tells me to copy one existing pgfsys file into the missing file, without any changes. What this post doesn't tell you explicitely is that you have to run texhash after that so that tex's file database is rebuilt.

When I again started xelatex, this time it seemed as if it will work! It was processing files and constantly giving some output. But no! At one point it stopped and I got the following error:
(/usr/share/texmf/tex/latex/beamer/themes/font/beamerfontthemeserif.sty)
! Font \zf@basefont="TeX Gyre Cursor" at 10.0pt not loadable: Metric (TFM) file
 or installed font not found.
\zf@fontspec ...ntname \zf@suffix " at \f@size pt
                                                  \unless \ifzf@icu \zf@set@...
l.80 ...d,NoCommon,NoContextual}]{TeX Gyre Cursor}
                                                 
?
Back to google-fu! Still, for this one it was easier to identify what's the problem. Namely, the font is missing (Metric (TFM) file or installed font not found). Again, using yum I found out that this font isn't packaged with Fedora 17 but it can be obtained from here. I downloaded zip archive, unpacked it, and moved all the font files into /usr/share/fonts/otf directory (before that I created otf subdirectory as it didn't exist). Afterwards, everything worked and presentation was successfully created. Note that I initially tried to place those fonts into /usr/share/texmf/fonts/opentype/public/tex-gyre/ directory. But that didn't work for unknown reason!

And guess what?! Now the source tex file was successfully transformed into PDF file.

While googling to find out solutions to problems I had, I realized that TeXLive distribution in Fedora 17 is 2007, while the newest version is 2011 with 2012 in development. Furthermore, I found out that  there is a plan for Fedora 18 to update TeXLive to at least version 2012. Since beta (or alpha) packages are readily available for F17, and I felt bold enough, I decided to give it a try.  So I installed repository per instructions on a given page and run 'yum update', and after a long wait, I tried again to run xelatex on the beamer presentation. This time I was completely out of luck:
kpathsea: Running mktexfmt xelatex.fmt
I can't find the format file `xelatex.fmt'!
For some reason, there was no xelatex.fmt file. As usual, first thing was to run 'yum whatprovides' to find out which package provides that file. But, none does! It seems to be a 'bug' in packages, so I had to revert texlive distribution. This was also a kind of a trial and error process, but to cut the story short what I did was: disabled texlive repository with experimental packages. Then, I used rpm to force removal of existing texlive packages, and then using yum I reinstalled the old ones (note that yum downgrade didn't work!). Finally, I used 'yum check/install' to fix any unresolved dependencies.

Tuesday, September 27, 2011

CentOS... something is happening!

I just noticed that RPM packages from RHEL 6.1 appeared on mirrors. Actually, they announced it few days earlier but that was sooner that I was expecting. :)

There is a small catch. In order for yum to be able to catch those packages you'll have to add new repository. Namely, CentOS team decided to go with a mechanism they call Continuous updates. In that way they'll try to be faster, but, time will tell if it will work or not.

The quickest way to do that is to run the following command:

rpm -ivh ftp://ftp.funet.fi/pub/mirrors/centos.org/6/cr/i386/RPMS/centos-release-cr-6-0.el6.centos.i686.rpm

or for 64-bit systems:

rpm -ivh ftp://ftp.funet.fi/pub/mirrors/centos.org/6/cr/x86_64/RPMS/centos-release-cr-6-0.el6.centos.x86_64.rpm

This will install necessary data for yum. Then, just run 'yum update' and that's it!

Still, we'll have to wait a bit more for 6.1, and especially for FreeIPA 2 that I'm waiting for!

About Me

scientist, consultant, security specialist, networking guy, system administrator, philosopher ;)

Blog Archive