0

I have recently upgraded centOS 6.6 to 7. It seemingly boots up, but it can not open the log in page, Actually the systems is frozen here. Does anybody have similar experience? If so, it would be great to help with this.

  • What procedure/document did you use to do the upgrade? Did You follow said document exactly? Do You have access to the console? What do you see in the logs when in single-user mode? – Sebastian Szary Jan 4 '15 at 21:56
  • I followed some typical procedure like this one: abi.io/blogs/… , Honestly, I have read that there are some risks to upgrade centOS 6 to 7, but I tried and it seems I messed up. – PeymanT Jan 4 '15 at 23:30
  • I only have access to UEFI and GNU GRUB, As a single user I faced with ''welcome to rescue mode! tyoe'' , I am needed to type the root password and finally three lines appear, the last one is: IPv6: ADDRCONF(NETDEV_CHANGE): eth0: link become ready – PeymanT Jan 4 '15 at 23:44
4

Had the same problem driving me nuts! It just takes forever to show the login prompt. On my VM it took around 25 minutes! The problem for me was on the audit package installed.

  • Problem #1 : yum not working - could not find libsasl2.so.2 The upgrade process resulted in me having /usr/lib64/libsasl2.so.3 . just create a symlink in /usr/lib64 named libsasl2.so.2 pointing to /usr/lib64/libsasl2.so.3 via:

    ln -s /usr/lib64/libsasl2.so.3 /usr/lib64/libsasl2.so.2

  • Problem #2 : yum was working again but could not complete most operations. Had to remove mod-perl-2.0.4-11.el6_5.x65_86. Please note that this may not apply to you.

  • Problem #3 : Delay during boot. removed audit package ( yum erase audit ) updated glibc package (yum update glibc ) and then update the system ( yum update ).

I rebooted and the system was up in less than 20 seconds!

2

While the symlink will work, I don't believe it's the complete solution. I'm not sure this solution is either. I had similar symptoms as above. Important programs like Python and sshd were linked with libsasl2.so.2, which is a el6 package.

It seems that in my case some of the post install scripts did not run correctly. I started the rescue system, and as root in the /root/preupgrade/postupgrade.d/pkgdowngrades, run the fixpkgdowngrades.sh script. This fixes all the references to libsasl2.so.2, correcting them to libsasl2.so.3. I then removed the libsasl2.so.2 symlink.

I also cd'ed into each of the other directories in /root/preupgrade/postupgrade.d and ran the scripts therein. The /root/preupgrade/postupgrade.d/clean_centos6_pkgs/postupgrade_cleanup.sh will need to be fixed near the bottom where it exits if not called with --cleanup.

  • This does not really answer the question. If you have a different question, you can ask it by clicking Ask Question. You can also add a bounty to draw more attention to this question once you have enough reputation. – n.st Jul 14 '15 at 22:14
  • I was addressing the libsasl2.so.2 library issue. This was one of the most relevant searches, and sshd wont start without this kludge.. I was trying to assist others. with similar issues – user123392 Jul 15 '15 at 14:45
0

I would use a Live CD to bypass the login page. Then, I would use an external hard drive to transfer your important data to. Once all of the essentials are backed up, I would reinstall your OS and transfer your data back to your computer.

While you are logged into your computer with a Live CD, I would take the opportunity to review your log files to learn what transpired before this happened so that you learn how to prevent this from happening again.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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