Top Nav

Archive | CentOS

Yum 404 Error On Xen-46

On CentOS 7, getting 404 error from yum when trying to update.

The error is related to Xen-46 which has been removed. The fix is to remove the repo and add the new Xen-48 repo:

 

 

0

Install jpegtran on CentOS 7

CentOS uses the libjpeg-turbo fork of the original libjpeg project. So to install jpegtran do this:

 

0

Dirty Cow Vulnerability (CVE-2016-5195)

On October 19, 2016, a privilege escalation vulnerability in the Linux kernel was disclosed. The bug is nicknamed Dirty COW because the underlying issue was a race condition in the way kernel handles copy-on-write (COW). Dirty COW has existed for a long time — at least since 2007, with kernel version 2.6.22 — so the vast majority of servers are at risk.

Exploiting this bug means that a regular, unprivileged user on your server can gain write access to any file they can read, and can therefore increase their privileges on the system. More information can be found on CVE-2016-5195 from Canonical, Red Hat, and Debian.

Fortunately, most major distributions have already released a fix. You can follow this tutorial to see if your server is vulnerable and to apply updates as needed.

Check Vulnerability

Ubuntu/Debian

To find out if your server is affected, check your kernel version.

  • uname -rv

You’ll see output like this:

Output

If your version is earlier than the following, you are affected:

  • 4.8.0-26.28 for Ubuntu 16.10
  • 4.4.0-45.66 for Ubuntu 16.04 LTS
  • 3.13.0-100.147 for Ubuntu 14.04 LTS
  • 3.2.0-113.155 for Ubuntu 12.04 LTS
  • 3.16.36-1+deb8u2 for Debian 8
  • 3.2.82-1 for Debian 7
  • 4.7.8-1 for Debian unstable

CentOS

Some versions of CentOS can use this script provided by RedHat for RHEL to test your server’s vulnerability. To try it, first download the script.

  • wget https://access.redhat.com/sites/default/files/rh-cve-2016-5195_1.sh

Then run it with bash.

  • bash rh-cve-2016-5195_1.sh

If you’re vulnerable, you’ll see output like this:

Output

Fix Vulnerability

Fortunately, applying the fix is straightforward: update your system and reboot your server.

On Ubuntu and Debian, upgrade your packages using apt-get.

  • sudo apt-get update && sudo apt-get dist-upgrade

You can update all of your packages on CentOS 6 and 7 with sudo yum update, but if you only want to update the kernel to address this bug, run:

  • sudo yum update kernel

Right now, we’re still waiting on a fix for CentOS 5. In the interim, you can use this workaround from the Red Hat bug tracker.

Finally, on all distributions, you’ll need to reboot your server to apply the changes.

  • sudo reboot

Conclusion

Make sure to update your Linux servers to stay protected from this privilege escalation bug.

0

Slow DNS Lookups For Web Requests

Ran into a strange problem recently … web server behind a firewall was able to resolve names with “dig” sucessfully but attempts to fetch web pages with “wget” or “curl” was very slow … seemed to hang on name resolution.  So this would work fine:

but this would hang for several seconds:

This problem extended to curl requests from with in PHP … in this case a Magento website … various plugins in the site were “calling home” when loading admin pages which resulted in making the admin painfully slow.

After much debugging we concluded that the problem was due to the fact that certain versions of glibc run IPv4 and IPv6 requests in parallel which breaks some firewalls and/or DNS servers. The work around was to add this option in /etc/resolv.conf:

This forces the requests to be made sequentially instead of in parallel. Hope this helps other struggling with these weird symptoms.

0