Tag Info

New answers tagged

0

There can be transient network errors between you and the pkg repo server which end up with you being unable to contact the repo server. A better place to ask this particular question would be the Oracle forums related to Solaris 11. (Try https://community.oracle.com/community/developer/english/server_%26_storage_systems/solaris/solaris_11 to start with). ...


0

PCRE devel package is not installed on your system. First install pcre-devel sudo yum install pcre-devel -y and then run sudo ./configure --with-included-apr it will automatically take pcre by default.


2

Jasonwryan (as per usual) was right on the mark with his initial comment. Arch's packages are supposed to be as close to "vanilla" as possible. Now, while you could use rpmextract or alien, there isn't really a good reason to do so. What you should do is create a PKGBUILD that uses the RPM as the source file and then installs everything that's needed where ...


0

The easiest method by far is to make use of a binary build that's provided through a YUM repository. One such option would be to use the hop5.inrepository. Specifically this page: gcc - Various compilers (C, C++, Objective-C, Java, ...). They're providing 4.8.2 which should work with CentOS 6.3 or 6.4. You might want to do an update prior: $ sudo yum ...


0

From https://forum.linode.com/viewtopic.php?t=8777 Changing el5 to el6 seems to do the job. su root cd /etc/yum.repos.d for FILE in *; do mv $FILE $FILE.old; sed 's/el5/el6/g' $FILE.old > $FILE; done yum clean all This fixed it for me.


1

Check out SUSE Studio. It basically lets you choose a base and customize it, appearance-wise and package-wise.


0

Other way to figure out what package you need pointing a file of its contents, is with the whatprovides yum option [root@host ~]# yum whatprovides /usr/sbin/sumtool But in your specific case, you got out of luck. I found no repos that have this package. I use the popular ones (CentOS-Vault, CentOS-Base, rpmforge, remi-repo and epel) and none of them ...


1

The tool sumtool is part of the package mtd-utils. $ sudo yum install mtd-utils How can you tell? You can use repoquery to find out what's in a particular package prior to installing it. $ repoquery -l mtd-utils | grep sumtool /usr/sbin/sumtool What package is command X part of? Again you can use repoquery to find out what package a given executable ...


1

In order to invoke a command by name, you need to put that symbolic link in one of the directories on the command search path. The environment variable PATH lists the directories in the command search path. The command in your question creates a symbolic link in the current directory, which is not useful. For system-wide commands not provided in a package, ...


1

Figured it out at last. I installed the EPEL Repos and then did the following (for CentOS anyway): # Install collectd-ping plugin. See http://www.rpmfind.net//linux/RPM/fedora/devel/rawhide/x86_64/c/collectd-ping-5.4.1-2.fc21.x86_64.html and above for versions yum install collectd-ping.x86_64 cp /usr/lib64/collectd/ping.so /opt/collectd/lib/collectd/ ...


1

You can enable any plugin you like in /etc/collectd/collectd.conf by uncommenting the plugin line: # /etc/collectd/collectd.conf #LoadPlugin perl LoadPlugin ping #LoadPlugin powerdns and the ping config: <Plugin ping> Host "host.foo.bar" Host "host.baz.qux" TTL 255 </Plugin> If you did not find a plugin, that possibly ...


0

The issue is in node-webkit https://groups.google.com/forum/#!topic/light-table-discussion/LTAXJKBuxbw from: https://github.com/LightTable/LightTable/issues/972 The quickest way I found to run LightTable on Debian: Copy the download link from LightTable site Paste the link in a browser (or wget) changing the version from 0.6.0 to 0.5.9 ...


7

If you bundle your binaries into your own RPMs then it's trivial to get a list of what they are and where they were installed. Example $ rpm -ql httpd| head -10 /etc/httpd /etc/httpd/conf /etc/httpd/conf.d /etc/httpd/conf.d/README /etc/httpd/conf.d/autoindex.conf /etc/httpd/conf.d/userdir.conf /etc/httpd/conf.d/welcome.conf /etc/httpd/conf.modules.d ...


4

An obvious suggestions is to name your binaries or your packages in a special way. So for example you could prefix them with cm-, per your initials as given in this post. If you are installing rpms they need to go into /usr/bin (if they are user level executables), per the FHS. They should not go into /usr/local/bin for example. That is for local installs ...


1

Binaries not part of the system or distribution are usually in /usr/local/bin the directory is usually in the standard $PATH so that your binaries will be found.


0

sudo apt-­get install network­-manager­-pptp  sudo apt­-get install network­-manager­-vpnc sudo apt-­get install network­-manager­-openvpn  then , follow this tuto: http://snnyc.com/wp-content/uploads/2012/06/connecting-to-cisco-vpn-from-ubuntu-linux-12-04.pdf


0

In this approach, on two occasions the output of ls, sorted on time and for directories only (-td), is piped to the head command which then selects the first line. The result is that the most recent version of both adt-bundle-linux-x86_64-*/sdk and of build-tools/android-*/ should be used for the modification of the PATH environment variable. ANDROID_SDK=$( ...


0

PATH="$PATH:$(printf '%s:' $ANDROID_SDK/build-tools/android-[0-9.]*/)" Should work. cd ~ ; mkdir dir1 dir2 dir3 ( PATH=$(printf %s: $HOME/dir[12]) echo "$PATH" ) /home/mikeserv/dir1:/home/mikeserv/dir2: Another method: PATH="$PATH:$( set -- `printf '%s\n' "$ANDROID_SDK"*/android[0-9.]*/ | sort -rV` ; [ -d "$1" ] && echo "$1" )" That takes ...


3

The ideal way is creating a soft link directory called adt, and re-creating it after SDK changes. Thus, your PATH will stay the same.


0

Source packages for linux are commonly structured with autotools and using them follows a common pattern. An easy way to tell is by looking at the INSTALL file inside, which will have instructions much like these. First you need to open the package up: tar -xzf foobar-1.2.3.gz Should construct a directory foobar-1.2.3. You can now cd foobar-1.2.3 and ...


2

There is no single way to 'install a .tgz file.' The .tgz format is just a compressed archive format (at tar archive that has been compressed with gzip) and can essentially contain anything. You can extract a .tgz file with the following command: tar -xvzf file.tgz If this contains software, usually there will be some text file inside with instructions ...


0

These days automated provision and configuration management (which includes installed software) has evolved greatly. To provision a CentOS machine I would typically: register a new system in cobbler create a A/AAA DNS entry, a host entry and a puppet service entry for this host in FreeeIPA PXE boot the machine (virtual or physical, the method applies) Sit ...


1

Did you try downloading the drivers from here? http://www.edimax.com/en/support_download.php?pl1_id=1. Also I found this thread titled: "Thread: Driver installation Probrema 7612UAn Edimax EW-v2" which covers the compiling and setting up of what looks to be the same card as yours.


1

It is not possible to filter by package size with --assume-yes, but you can query the packages by size with repoquery, and with some awk filtering, install just the files that have a download size with less than 1MB: yum -y install `repoquery --pkgnarrow=updates -a --qf "%{name} %{size}" | awk '$2 < 1048576 {print $1}'` What this command does: ...


1

Well the direct route would be to go to the PCRE website and download the package you want. http://www.pcre.org/ Or download from the SVN repo: $ svn co svn://vcs.exim.org/pcre/code/trunk pcre Since you're on SuSE you can probably go poking through the SuSE Build Service and get whatever version you need, or submit a build yourself. I found the PCRE ...



Top 50 recent answers are included