Tag Info

New answers tagged

1

Yes, this is like what a Live CD does. It's done using a special filesystem driver that's designed to overlay multiple filesystems on top of one another - in this case, a read-only file system with a RAM disk. There are lots of different choices of overlay filesystems - try searching around for UnionFS, aufs, and overlayfs to get an idea of what your ...


1

Turns out it was a SELinux permissions problem with files in ~/.ssh/. I copied the files from a backup stored on an external drive, where I guess they had received different labels, and they kept their incorrect labels. I should have known. When experiencing puzzling failures that could be permission issues, always check SELinux. The fix: restorecon -r ...


2

I think nlu has pointed into the right direction. The generally better way to trace down ssh problems is by stopping the sshd on the server side and then start it with $(which sshd) -d. That will give you more meaningful error messages in almost all cases. Update: Sorry - you already did this. There seems to be one difference between sshd on the cli and ...


1

You should check the ownership and permissions of /home/user/.ssh/authorized_keys too.


1

No, it is not possible. Note that it is not just a matter of formatting. The free program in procps-ng has been recently updated to include the new MemAvailable metric that was recently introduced in the Linux kernel and that is a much better indication of the memory actually available than the typical (and often wrong) "free + buffers/cache" estimation ...


0

Inspired by this thread try the following Pre-condition VM with 4GB Memory + 40GB harddrive Workflow 1. mkdir -pv ~/chromium 2. cd ~/chromium 3. git config --global user.name "You Name" 4. git config --global user.email “youremail@example.com” 5. git config --global core.autocrlf false 6. git config --global core.filemode false 7. git config ...


0

I've patched mysql-connector-c++.spec and mysql-workbench.spec so you can mock both spec files for Fedora 21. Please read the instructions in README.rst to build it yourself. https://github.com/meeuw/mysql-workbench-fedora-21


0

If you have one site, which you want to support on your machine, it is not necessary to configure(of course if you have a DNS record). But to answer on your question, you can put your domain name in /etc/httpd/conf/httpd.conf: ServerName: isaacme.com And in your host file, /etc/hosts: "your ip address" isaacme.com


1

The server has recently disabled SSLv3 and the client required TLSv2: https://bugzilla.redhat.com/show_bug.cgi?id=994599 The server supports all TLS versions now.


1

In the /etc/sudoers file (accessible with visudo), look for requiretty. You can turn this off for the program with: Defaults!/path/to/program !requiretty In your case - you are using bash, so this path would be whatever you get from which bash.


1

You should verify that sudo doesn't require tty in the /etc/sudoers file (requiretty should not be set).


0

Here's how to permanently change the context of a directory: # install semanage if you don't already have it: yum install policycoreutils-python # give the directory a new default context. The part at the end is a regex. semanage fcontext -a -t httpd_sys_rw_content_t "/path/to/directory(/.*)?" # apply the default context to the directory restorecon -R ...


0

To fix this error, I think the command you wish to find is "yum-config-manager --disable ". In your case, try both "yum-config-manager --disable ffmulticonverter_stable" or "yum-config-manager --disable ppa:ffmulticonverter/stable". What's more is, why are you trying to install a ppa on a Fedora system? It won't work, as ppa is for Debian based systems. If ...


0

The kernel-devel package only has what is required to build additional modules for the kernel, not the entire kernel sources. You need to get the source rpm to build the entire kernel. You can stick with using a vanilla kernel as you said you've done, or if you'd like to build your own kernel rpm see these instructions


0

/proc/sys/kernel/printk controls what is printed to the system's console, so the prints showing up with dmesg is expected behaviour. You can filter dmesg output with e.g. dmesg --level=warn (see dmesg --help). With a dynamic logging level, something has to filter out the printk messages. It seems the design decision is to not have the kernel do this, but ...


0

I think I experienced a similar issue which CentOS 7. Try to partition everything with another linux (install) media (gparted or arch for example) and then just enter the mountpoints and filesystems but not click on reformat


0

The udev rules work well on plug events, but to get the correct behaviour after resume from suspend I need somthing like this in /etc/pm/sleep.d: #!/bin/sh case "${1}" in resume|thaw) if grep -q 1 /sys/class/power_supply/ADP1/online; then /usr/sbin/hdparm -B 254 /dev/sda else /usr/sbin/hdparm -B 128 /dev/sda fi ;; esac


1

You need to close the <startupnotify> tag with a corresponding </startupnotify>. See the official documentation.


0

I've seen some odd activity like this when distributing identical VMs on EC2. I always assume something awful has happened and rebuild all of the nodes involved. I would suggest a yum clean all, also comparing the Linux kernel version, any associated yum repo configs and run reboots on both machines. If it's still behaving the same way, try removing the ...


0

If the systems are identical, run yum clean all which clears yum cache of packages , metadata and headers.


0

Check your settings in /etc/yum.conf and under the /etc/yum.repos.d/ directory. Check the main config of /etc/yum.conf since these settings take priority over others. The values you define in the [main] section of the /etc/yum.conf file may override values set in individual [repository] sections. Compare the one that updates to the one that doesn't. ...


0

The installer runs as root. It doesn't have a special user id. Fortunately, you don't need it -- this doesn't have anything to do with what you want. What you do want to do is look in /root on your installed system, for the file /root/anaconda-ks.cfg. The %packages section in this file will tell you what package groups are installed; but even better, if you ...


1

The closest thing I've found to Gentoo's etc-update is rpmconf: # yum install -y rpmconf # rpmconf -a Configuration file `/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.25-5.b18.fc21.x86_64/jre/lib/security/US_export_policy.jar' -rw-r--r--. 1 root root 620 Oct 2 16:38 ...


0

The solution was that the ftp service was not actually permanently added to the firewall, simply using the following line worked firewall-cmd --permanent --add-service=ftp and following that with firewall-cmd --reload


0

See the switchdesk tool. It alters your .Xclients.


2

Fedora 16 was end of life on 2013-02-12, almost 2 years ago. The repositories, for that version, probably have been taken offline. You should reinstall a supported version.


1

The easy way is to install grub-customizer tool: sudo add-apt-repository ppa:danielrichter2007/grub-customizer sudo apt-get update sudo apt-get install grub-customizer When Grub Customizer starts up, you’ll see a list of all the items that show up in the boot menu. To hide entries that you don’t want to see anymore, simply uncheck the checkbox next to ...


1

What about this? # yum shell > remove fedora-logos fedora-release fedora-release-notes > install generic-logos generic-release generic-release-notes > run --> Running transaction check ---> Package fedora-logos.x86_64 0:21.0.5-1.fc21 will be erased ---> Package fedora-release.noarch 0:21-2 will be erased --> Processing Dependency: ...


0

Download and install Workbench RPM for x86_64 sudo yum install tinyxml


0

I see the same thing on Fedora 21 using the XFCE distro. It turns out the default install does not include policyutils-python, which contains semanage. You can often find out which package contains an item by using: yum whatprovides semanage



Top 50 recent answers are included