Issue
I created a CentOS 7 instance using Google's default boot disk x86_64 built on 20200714.
I successfully logged in via SSH via browser, then updated the packages using
`sudo yum -y update`
After rebooting the VM, whenever I tried logging in via SSH, the following error appears:
Connection Failed
We are unable to connect to the VM on port 22. Learn more about possible causes of this issue.
Any ideas why this would happen?
Solution
This bug is a hot topic at this time. RHEL and CentOS 7 and 8 may be affected. Red Hat engineers are working on this bug.
First precautionary measures at the moment are:
- DO NOT apply the affected Errata RHSA-2020:3217 for RHEL 7.
- DO NOT apply the affected Errata RHSA-2020:3216 for RHEL 8.
- DO NOT reboot the systems with applied updates.
More information about this bug and possible workarounds are provided here:
https://access.redhat.com/solutions/5272311
Also, there is a freshly registered issue on Google Issue Tracker. You can keep and eye on it:
https://issuetracker.google.com/162523000
If you have a healthy VM instance in the same zone and with the same OS version as the impacted VM but without erroneous updates applied, there is a chance to fix the damaged OS manually:
- Stop the impacted instance that can't boot anymore.
- Edit it so that detach the boot disk.
- Attach the impacted boot disk to the healthy instance with the same OS version.
- Use
lsblkto determine the attached disk name. - Mount the EFI partition of the impacted boot disk on the healthy instance (e.g.
mount /dev/sdb1 /mnt) - Copy the old EFI files from the boot disk of the healthy instance to the mounted impacted boot disk:
# CentOS
cp /boot/efi/EFI/centos/shimx64.efi /mnt/EFI/centos/
# RHEL
cp /boot/efi/EFI/redhat/shimx64.efi /mnt/EFI/redhat/
umount /mnt- Detach the impacted disk and attach it back to the impacted instance as a boot disk.
- Boot the impacted instance and check the result.
Answered By - mebius99 Answer Checked By - Dawn Plyler (WPSolving Volunteer)