Take the 2-minute tour ×
Unix & Linux Stack Exchange is a question and answer site for users of Linux, FreeBSD and other Un*x-like operating systems.. It's 100% free, no registration required.

I have been struggling to get yum to work when trying to install emacs in my recent installation of Fedora 20, and posted the question, on this site. "How can I get yum and the internet browser to work after fresh Fedora 20 installation", which inspired several useful answers (many thanks to the kind posters), but none of them, unfortunately, solved it. I don't expect any more to come from that question, but I have tried an approach that might work, but it still fails with the error in this question.

I came across the suggestion

[root@localhost Harry]# setenforce 0
[root@localhost Harry]# yum clean expire-cache
[root@localhost Harry]# yum update selinux-policy\*
[root@localhost Harry]# setenforce 1

And tried it out, not getting very far:

[root@localhost Harry]# setenforce 0
[root@localhost Harry]# yum clean expire-cache
Loaded plugins: langpacks, refresh-packagekit
Cleaning repos: fedora updates
0 metadata files removed
[root@localhost Harry]# yum update selinux-policy\*
Loaded plugins: langpacks, refresh-packagekit
fedora/20/i386/metalink                                  |  33 kB     00:00     
http://ftp.up.pt/fedora/releases/20/Everything/i386/os/repodata/repomd.xml: [Errno 14] curl#6 - "Could not resolve host: ftp.up.pt"
Trying other mirror.

and so it went on for ages, the same error, [Errno 14] curl#6 - "Could not resolve host: www.mirrorservice.org" I get whenever I try to use yum.

I followed the lead I had and tried this, which looked promising, but again it failed with an error that I cannot fathom out:

[root@localhost Harry]# yum-config-manager --add-repo=http://download.fedoraproject.org/pub/fedora/linux/updates/testing/20/i386/
Loaded plugins: langpacks, refresh-packagekit
adding repo from: http://download.fedoraproject.org/pub/fedora/linux/updates/testing/20/i386/

name=added from: http://download.fedoraproject.org/pub/fedora/linux/updates/testing/20/i386/

[root@localhost Harry]# yum install emacs
Loaded plugins: langpacks, refresh-packagekit

You have enabled checking of packages via GPG keys. This is a good thing. 
However, you do not have any GPG public keys installed. You need to download
the keys for packages you wish to install and install them.
You can do that by running the command:
    rpm --import public.gpg.key

Alternatively you can specify the url to the key you would like to use
for a repository in the 'gpgkey' option in a repository section and yum 
will install it for you.

For more information contact your distribution or package provider.

Problem repository: download.fedoraproject.org_pub_fedora_linux_updates_testing_20_i386_

But again a failure, I get it whenever I try import:

[root@localhost Harry]# rpm --import public.gpg.key
error: public.gpg.key: import read failed(2).

So my question is: is this is a useful approach, and, if so, how do I get round the import failed(2) error?

In answer to Leiaz's comment, thank you, this is what I get from ifconfig. It's a wired connection to a BT Homehub 3 router, May I add that I have no problem with the internet in Windows XP installed on the same dual boot machine, nor on the two laptops with a wireless connection to the same router. I also have no trouble with the Firefox browser in this Linux, which is the one with the yum trouble, and the one I am using to post this question: in fact it is faster than the one on my laptop.

[root@localhost Harry]# ifconfig
lo: flags=73<UP,LOOPBACK,RUNNING>  mtu 65536
        inet  netmask
        inet6 ::1  prefixlen 128  scopeid 0x10<host>
        loop  txqueuelen 0  (Local Loopback)
        RX packets 8  bytes 668 (668.0 B)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 8  bytes 668 (668.0 B)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

p20p1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet  netmask  broadcast
        inet6 fe80::76d4:35ff:fe5c:f999  prefixlen 64  scopeid 0x20<link>
        ether 74:d4:35:5c:f9:99  txqueuelen 1000  (Ethernet)
        RX packets 62356  bytes 42267371 (40.3 MiB)
        RX errors 0  dropped 0  overruns 0  frame 0
        TX packets 56754  bytes 11151168 (10.6 MiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

Thanks vonbrand and again Leiaz. This is what I get:

[root@localhost Harry]# cat /etc/resolv.conf 
# Generated by NetworkManager
domain home
search home
[root@localhost Harry]#  route -n
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface         UG    1024   0        0 p20p1   U     0      0        0 p20p1
[root@localhost Harry]# 

[root@localhost Harry]# ping
PING ( 56(84) bytes of data.
64 bytes from icmp_seq=1 ttl=64 time=0.701 ms
64 bytes from icmp_seq=2 ttl=64 time=0.678 ms
64 bytes 
                                         ...    time=0.683 ms
64 bytes from icmp_seq=10 ttl=64 time=0.668 ms
64 bytes from icmp_seq=11 ttl=64 time=0.665 ms
[1]+  Stopped                 ping
[root@localhost Harry

Having no idea how to deal with ipv6, and unable to find the answer on the Internet, I thought I would try a different Linux, so I have installed openSUSE, but it appears to have the same problem. As with Fedora, I am using Firefox from within Linux, to post this, and the 'net is fully working in the other windows partition.

Please, please can someone help me to solve this, I am completely stumped. Without emacs and a C compiler, well Linux is no longer any use to me, after happily using it for more than ten years.

Here is the relevant openSUSE result:

harry@linux-18x3:~> gcc
If 'gcc' is not a typo you can use command-not-found to lookup the package that contains it, like this:
    cnf gcc
harry@linux-18x3:~> cnf gcc

The program 'gcc' can be found in following packages:
  * gcc [ path: /usr/bin/gcc, repository: zypp (openSUSE-13.1-1.10) ]
  * gcc [ path: /usr/bin/gcc, repository: zypp (repo-oss) ]

Try installing with:
    sudo zypper install gcc

harry@linux-18x3:~> sudo zypper install gcc
root's password:
Download (curl) error for 'http://download.opensuse.org/distribution/13.1/repo/non-oss/content':
Error code: Connection failed
Error message: Could not resolve host: download.opensuse.org

Abort, retry, ignore? [a/r/i/? shows all options] (a):
share|improve this question
You should deal with "Could not resolve host" first. Your real problem is your internet connection. How are you connected to the internet ? Does the browser work now ? Does ifconfig shows an active interface ? –  Leiaz Mar 17 '14 at 17:17
I think Fedora uses Gnome's network-manager. You should check how the connection is configured there. I can't help with the details, I don't use Fedora or Gnome. You could check if there are correct DNS addresses in /etc/resolve.conf but if it works automatically for the other installs Fedora should also get them automatically through DHCP. –  Leiaz Mar 17 '14 at 17:58
How do you configure the network? It should come up without any intervention on your part. Post the output of route -n, and ping (or whatever route tells you is your default router). What is in /etc/resolv.conf? –  vonbrand Mar 17 '14 at 18:42
Sorry vonbrand, I intended to upvote your comment, fluffed it and now not allowed to do it again, deserves 2 like the others. –  Harry Weston Mar 17 '14 at 18:48
I see you have a local ipv6 address, could yum be trying to connect using ipv6 ? (And fail as it is not configured to connect to the internet.) –  Leiaz Mar 17 '14 at 18:55

1 Answer 1

up vote 0 down vote accepted

At last, after almost six weeks of frustrated, numerous, attempted solutions based on suggestions by kind friends and Internet question sites, I have solved the problem (I think -- I am cautiously optimistic). The underlying symptom was that yum install emacs failed with a long list of errors,. Now it has finally worked, without hesitation. I don't know why, finding out is my next quest.

This is what I followed:


Answer by Stramash November 4, 2013 at 2:24 pm

Resolved this by adding nameserver above my router’s address in resolv.conf that was obtained by DHCP.

Not quite sure why it will not work with the automatic DHCP settings.


share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.