Chapter 4. Upgrades from Debian 10 (buster)

Before upgrading your system, it is strongly recommended that you make a full backup, or at least back up any data or configuration information you can't afford to lose. The upgrade tools and process are quite reliable, but a hardware failure in the middle of an upgrade could result in a severely damaged system.

The main things you'll want to back up are the contents of /etc, /var/lib/dpkg, /var/lib/apt/extended_states and the output of dpkg --get-selections "*" (the quotes are important). If you use aptitude to manage packages on your system, you will also want to back up /var/lib/aptitude/pkgstates.

The upgrade process itself does not modify anything in the /home directory. However, some applications (e.g. parts of the Mozilla suite, and the GNOME and KDE desktop environments) are known to overwrite existing user settings with new defaults when a new version of the application is first started by a user. As a precaution, you may want to make a backup of the hidden files and directories (“dotfiles”) in users' home directories. This backup may help to restore or recreate the old settings. You may also want to inform users about this.

Any package installation operation must be run with superuser privileges, so either log in as root or use su or sudo to gain the necessary access rights.

The upgrade has a few preconditions; you should check them before actually executing the upgrade.

It's wise to inform all users in advance of any upgrades you're planning, although users accessing your system via an ssh connection should notice little during the upgrade, and should be able to continue working.

If you wish to take extra precautions, back up or unmount the /home partition before upgrading.

You will have to do a kernel upgrade when upgrading to bullseye, so a reboot will be necessary. Typically, this will be done after the upgrade is finished.

There might be services that are offered by the system which are associated with packages that will be included in the upgrade. If this is the case, please note that, during the upgrade, these services will be stopped while their associated packages are being replaced and configured. During this time, these services will not be available.

The precise downtime for these services will vary depending on the number of packages being upgraded in the system, and it also includes the time the system administrator spends answering any configuration questions from package upgrades. Notice that if the upgrade process is left unattended and the system requests input during the upgrade there is a high possibility of services being unavailable^[1] for a significant period of time.

If the system being upgraded provides critical services for your users or the network^[2], you can reduce the downtime if you do a minimal system upgrade, as described in Section 4.4.4, “Minimal system upgrade”, followed by a kernel upgrade and reboot, and then upgrade the packages associated with your critical services. Upgrade these packages prior to doing the full upgrade described in Section 4.4.5, “Upgrading the system”. This way you can ensure that these critical services are running and available through the full upgrade process, and their downtime is reduced.

Although Debian tries to ensure that your system stays bootable at all times, there is always a chance that you may experience problems rebooting your system after the upgrade. Known potential issues are documented in this and the next chapters of these Release Notes.

For this reason it makes sense to ensure that you will be able to recover if your system should fail to reboot or, for remotely managed systems, fail to bring up networking.

If you are upgrading remotely via an ssh link it is recommended that you take the necessary precautions to be able to access the server through a remote serial terminal. There is a chance that, after upgrading the kernel and rebooting, you will have to fix the system configuration through a local console. Also, if the system is rebooted accidentally in the middle of an upgrade there is a chance you will need to recover using a local console.

For emergency recovery we generally recommend using the rescue mode of the bullseye Debian Installer. The advantage of using the installer is that you can choose between its many methods to find one that best suits your situation. For more information, please consult the section “Recovering a Broken System” in chapter 8 of the Installation Guide and the Debian Installer FAQ.

If that fails, you will need an alternative way to boot your system so you can access and repair it. One option is to use a special rescue or live install image. After booting from that, you should be able to mount your root file system and chroot into it to investigate and fix the problem.

# mount -o remount,rw /
      

	Important
	If you are using some VPN services (such as tinc) consider that they might not be available throughout the upgrade process. Please see Section 4.1.3, “Prepare for downtime on services”.

In order to gain extra safety margin when upgrading remotely, we suggest that you run upgrade processes in the virtual console provided by the screen program, which enables safe reconnection and ensures the upgrade process is not interrupted even if the remote connection process temporarily fails.

Direct upgrades from Debian releases older than 10 (buster) are not supported. Display your Debian version with:

$ cat /etc/debian_version
      

Please follow the instructions in the Release Notes for Debian 10 to upgrade to Debian 10 first.

Below there are two methods for finding installed packages that did not come from Debian, using either aptitude or apt-forktracer. Please note that neither of them are 100% accurate (e.g. the aptitude example will list packages that were once provided by Debian but no longer are, such as old kernel packages).

$ aptitude search '?narrow(?installed, ?not(?origin(Debian)))'
$ apt-forktracer | sort
    

This procedure assumes your system has been updated to the latest point release of buster. If you have not done this or are unsure, follow the instructions in Section A.1, “Upgrading your buster system”.

You should make sure the package database is ready before proceeding with the upgrade. If you are a user of another package manager like aptitude or synaptic, review any pending actions. A package scheduled for installation or removal might interfere with the upgrade procedure. Note that correcting this is only possible if your APT source-list files still point to buster and not to stable or bullseye; see Section A.2, “Checking your APT source-list files”.

It is a good idea to remove obsolete packages from your system before upgrading. They may introduce complications during the upgrade process, and can present security risks as they are no longer maintained.

A previous upgrade may have left unused copies of configuration files; old versions of configuration files, versions supplied by the package maintainers, etc. Removing leftover files from previous upgrades can avoid confusion. Find such leftover files with:

# find /etc -name '*.dpkg-*' -o -name '*.ucf-*' -o -name '*.merge-error'
    

For APT source lines referencing the security archive, the format has changed slightly along with the release name, going from buster/updates to bullseye-security; see Section 5.1.2, “Changed security archive layout”.

If you have listed the proposed-updates section in your APT source-list files, you should remove it before attempting to upgrade your system. This is a precaution to reduce the likelihood of conflicts.

If you have any non-Debian packages on your system, you should be aware that these may be removed during the upgrade because of conflicting dependencies. If these packages were installed by adding an extra package archive in your APT source-list files, you should check if that archive also offers packages compiled for bullseye and change the source item accordingly at the same time as your source items for Debian packages.

Some users may have unofficial backported “newer” versions of packages that are in Debian installed on their buster system. Such packages are most likely to cause problems during an upgrade as they may result in file conflicts^[4]. Section 4.5, “Possible issues during upgrade” has some information on how to deal with file conflicts if they should occur.

If you have configured APT to install certain packages from a distribution other than stable (e.g. from testing), you may have to change your APT pinning configuration (stored in /etc/apt/preferences and /etc/apt/preferences.d/) to allow the upgrade of packages to the versions in the new stable release. Further information on APT pinning can be found in apt_preferences(5).

Regardless of the method used for upgrading, it is recommended that you check the status of all packages first, and verify that all packages are in an upgradable state. The following command will show any packages which have a status of Half-Installed or Failed-Config, and those with any error status.

# dpkg --audit
    

You could also inspect the state of all packages on your system using aptitude or with commands such as

# dpkg -l | pager
    

or

# dpkg --get-selections "*" > ~/curr-pkgs.txt
    

It is desirable to remove any holds before upgrading. If any package that is essential for the upgrade is on hold, the upgrade will fail.

Note that aptitude uses a different method for registering packages that are on hold than apt and dselect. You can identify packages on hold for aptitude with

# aptitude search "~ahold" 
    

If you want to check which packages you had on hold for apt, you should use

# dpkg --get-selections | grep 'hold$'
    

If you changed and recompiled a package locally, and didn't rename it or put an epoch in the version, you must put it on hold to prevent it from being upgraded.

The “hold” package state for apt can be changed using:

# echo package_name hold | dpkg --set-selections
    

Replace hold with install to unset the “hold” state.

If there is anything you need to fix, it is best to make sure your APT source-list files still refer to buster as explained in Section A.2, “Checking your APT source-list files”.

On new installations the default is for APT to be set up to use the Debian APT CDN service, which should ensure that packages are automatically downloaded from a server near you in network terms. As this is a relatively new service, older installations may have configuration that still points to one of the main Debian Internet servers or one of the mirrors. If you haven't done so yet, it is recommended to switch over to the use of the CDN service in your APT configuration.

To make use of the CDN service, add a line like this to your APT source configuration (assuming you are using main and contrib):

deb http://deb.debian.org/debian bullseye main contrib

After adding your new sources, disable the previously existing “deb” lines by placing a hash sign (#) in front of them.

However, if you get better results using a specific mirror that is close to you in network terms, this option is still available.

Debian mirror addresses can be found at https://www.debian.org/distrib/ftplist (look at the “list of Debian mirrors” section).

For example, suppose your closest Debian mirror is http://mirrors.kernel.org. If you inspect that mirror with a web browser, you will notice that the main directories are organized like this:

http://mirrors.kernel.org/debian/dists/bullseye/main/binary-ppc64el/...
http://mirrors.kernel.org/debian/dists/bullseye/contrib/binary-ppc64el/...
    

To configure APT to use a given mirror, add a line like this (again, assuming you are using main and contrib):

deb http://mirrors.kernel.org/debian bullseye main contrib

Note that the “dists” is added implicitly, and the arguments after the release name are used to expand the path into multiple directories.

Again, after adding your new sources, disable the previously existing archive entries.

Instead of using remote package mirrors, you may wish to modify the APT source-list files to use a mirror on a local disk (possibly mounted over NFS).

For example, your package mirror may be under /var/local/debian/, and have main directories like this:

/var/local/debian/dists/bullseye/main/binary-ppc64el/...
/var/local/debian/dists/bullseye/contrib/binary-ppc64el/...
    

To use this with apt, add this line to your sources.list file:

deb file:/var/local/debian bullseye main contrib

Note that the “dists” is added implicitly, and the arguments after the release name are used to expand the path into multiple directories.

After adding your new sources, disable the previously existing archive entries in the APT source-list files by placing a hash sign (#) in front of them.

If you want to use only DVDs (or CDs or Blu-ray Discs), comment out the existing entries in all the APT source-list files by placing a hash sign (#) in front of them.

Make sure there is a line in /etc/fstab that enables mounting your CD-ROM drive at the /media/cdrom mount point. For example, if /dev/sr0 is your CD-ROM drive, /etc/fstab should contain a line like:

/dev/sr0 /media/cdrom auto noauto,ro 0 0
    

Note that there must be no spaces between the words noauto,ro in the fourth field.

To verify it works, insert a CD and try running

# mount /media/cdrom    # this will mount the CD to the mount point
# ls -alF /media/cdrom  # this should show the CD's root directory
# umount /media/cdrom   # this will unmount the CD
    

Next, run:

# apt-cdrom add
    

for each Debian Binary CD-ROM you have, to add the data about each CD to APT's database.

It is strongly recommended that you use the /usr/bin/script program to record a transcript of the upgrade session. Then if a problem occurs, you will have a log of what happened, and if needed, can provide exact information in a bug report. To start the recording, type:

# script -t 2>~/upgrade-bullseyestep.time -a ~/upgrade-bullseyestep.script
    

or similar. If you have to rerun the typescript (e.g. if you have to reboot the system) use different step values to indicate which step of the upgrade you are logging. Do not put the typescript file in a temporary directory such as /tmp or /var/tmp (files in those directories may be deleted during the upgrade or during any restart).

The typescript will also allow you to review information that has scrolled off-screen. If you are at the system's console, just switch to VT2 (using Alt+F2) and, after logging in, use less -R ~root/upgrade-bullseye.script to view the file.

After you have completed the upgrade, you can stop script by typing exit at the prompt.

apt will also log the changed package states in /var/log/apt/history.log and the terminal output in /var/log/apt/term.log. dpkg will, in addition, log all package state changes in /var/log/dpkg.log. If you use aptitude, it will also log state changes in /var/log/aptitude.

If you have used the -t switch for script you can use the scriptreplay program to replay the whole session:

# scriptreplay ~/upgrade-bullseyestep.time ~/upgrade-bullseyestep.script
    

First the list of available packages for the new release needs to be fetched. This is done by executing:

# apt update
    

	Note
	Users of apt-secure may find issues when using aptitude or apt-get. For apt-get, you can use apt-get update --allow-releaseinfo-change.

You have to make sure before upgrading your system that you will have sufficient hard disk space when you start the full system upgrade described in Section 4.4.5, “Upgrading the system”. First, any package needed for installation that is fetched from the network is stored in /var/cache/apt/archives (and the partial/ subdirectory, during download), so you must make sure you have enough space on the file system partition that holds /var/ to temporarily download the packages that will be installed in your system. After the download, you will probably need more space in other file system partitions in order to both install upgraded packages (which might contain bigger binaries or more data) and new packages that will be pulled in for the upgrade. If your system does not have sufficient space you might end up with an incomplete upgrade that is difficult to recover from.

apt can show you detailed information about the disk space needed for the installation. Before executing the upgrade, you can see this estimate by running:

# apt -o APT::Get::Trivial-Only=true full-upgrade
[ ... ]
XXX upgraded, XXX newly installed, XXX to remove and XXX not upgraded.
Need to get xx.xMB of archives. 
After this operation, AAAMB of additional disk space will be used.
    

	Note
	Running this command at the beginning of the upgrade process may give an error, for the reasons described in the next sections. In that case you will need to wait until you've done the minimal system upgrade as in Section 4.4.4, “Minimal system upgrade” before running this command to estimate the disk space.

If you do not have enough space for the upgrade, apt will warn you with a message like this:

E: You don't have enough free space in /var/cache/apt/archives/.
    

In this situation, make sure you free up space beforehand. You can:

Note that in order to safely remove packages, it is advisable to switch your APT source-list files back to buster as described in Section A.2, “Checking your APT source-list files”.

	Important
	If you are upgrading remotely, be aware of Section 5.1.22, “No new SSH connections possible during upgrade”.

In some cases, doing the full upgrade (as described below) directly might remove large numbers of packages that you will want to keep. We therefore recommend a two-part upgrade process: first a minimal upgrade to overcome these conflicts, then a full upgrade as described in Section 4.4.5, “Upgrading the system”.

To do this, first run:

# apt upgrade --without-new-pkgs
    

This has the effect of upgrading those packages which can be upgraded without requiring any other packages to be removed or installed.

The minimal system upgrade can also be useful when the system is tight on space and a full upgrade cannot be run due to space constraints.

If the apt-listchanges package is installed, it will (in its default configuration) show important information about upgraded packages in a pager after downloading the packages. Press q after reading to exit the pager and continue the upgrade.

Once you have taken the previous steps, you are now ready to continue with the main part of the upgrade. Execute:

# apt full-upgrade
    

This will perform a complete upgrade of the system, installing the newest available versions of all packages, and resolving all possible dependency changes between packages in different releases. If necessary, it will install some new packages (usually new library versions, or renamed packages), and remove any conflicting obsoleted packages.

When upgrading from a set of CDs/DVDs/BDs, you will probably be asked to insert specific discs at several points during the upgrade. You might have to insert the same disc multiple times; this is due to inter-related packages that have been spread out over the discs.

New versions of currently installed packages that cannot be upgraded without changing the install status of another package will be left at their current version (displayed as “held back”). This can be resolved by either using aptitude to choose these packages for installation or by trying apt install package.

In some cases the apt full-upgrade step can fail after downloading packages with:

E: Could not perform immediate configuration on 'package'.  Please see man 5 apt.conf under APT::Immediate-Configure for details.
      

If that happens, running apt full-upgrade -o APT::Immediate-Configure=0 instead should allow the upgrade to proceed.

Another possible workaround for this problem is to temporarily add both buster and bullseye sources to your APT source-list files and run apt update.

The upgrade process to bullseye might ask for the removal of packages on the system. The precise list of packages will vary depending on the set of packages that you have installed. These release notes give general advice on these removals, but if in doubt, it is recommended that you examine the package removals proposed by each method before proceeding. For more information about packages obsoleted in bullseye, see Section 4.8, “Obsolete packages”.

Sometimes it's necessary to enable the APT::Force-LoopBreak option in APT to be able to temporarily remove an essential package due to a Conflicts/Pre-Depends loop. apt will alert you of this and abort the upgrade. You can work around this by specifying the option -o APT::Force-LoopBreak=1 on the apt command line.

It is possible that a system's dependency structure can be so corrupt as to require manual intervention. Usually this means using apt or

# dpkg --remove package_name
    

to eliminate some of the offending packages, or

# apt -f install
# dpkg --configure --pending
    

In extreme cases you might have to force re-installation with a command like

# dpkg --install /path/to/package_name.deb
    

File conflicts should not occur if you upgrade from a “pure” buster system, but can occur if you have unofficial backports installed. A file conflict will result in an error like:

Unpacking <package-foo> (from <package-foo-file>) ...
dpkg: error processing <package-foo> (--install):
trying to overwrite `<some-file-name>',
which is also in package <package-bar>
dpkg-deb: subprocess paste killed by signal (Broken pipe)
Errors were encountered while processing:
<package-foo>
    

You can try to solve a file conflict by forcibly removing the package mentioned on the last line of the error message:

# dpkg -r --force-depends package_name
    

After fixing things up, you should be able to resume the upgrade by repeating the previously described apt commands.

During the upgrade, you will be asked questions regarding the configuration or re-configuration of several packages. When you are asked if any file in the /etc/init.d directory, or the /etc/manpath.config file should be replaced by the package maintainer's version, it's usually necessary to answer “yes” to ensure system consistency. You can always revert to the old versions, since they will be saved with a .dpkg-old extension.

If you're not sure what to do, write down the name of the package or file and sort things out at a later time. You can search in the typescript file to review the information that was on the screen during the upgrade.

If you are running the upgrade using the system's local console you might find that at some points during the upgrade the console is shifted over to a different view and you lose visibility of the upgrade process. For example, this may happen in systems with a graphical interface when the display manager is restarted.

To recover the console where the upgrade was running you will have to use Ctrl+Alt+F1 (if in the graphical startup screen) or Alt+F1 (if in the local text-mode console) to switch back to the virtual terminal 1. Replace F1 with the function key with the same number as the virtual terminal the upgrade was running in. You can also use Alt+Left Arrow or Alt+Right Arrow to switch between the different text-mode terminals.

When you full-upgrade from buster to bullseye, it is strongly recommended that you install a linux-image-* metapackage, if you have not done so before. These metapackages will automatically pull in a newer version of the kernel during upgrades. You can verify whether you have one installed by running:

# dpkg -l "linux-image*" | grep ^ii | grep -i meta
    

If you do not see any output, then you will either need to install a new linux-image package by hand or install a linux-image metapackage. To see a list of available linux-image metapackages, run:

# apt-cache search linux-image- | grep -i meta | grep -v transition
    

If you are unsure about which package to select, run uname -r and look for a package with a similar name. For example, if you see “4.9.0-8-amd64”, it is recommended that you install linux-image-amd64. You may also use apt to see a long description of each package in order to help choose the best one available. For example:

# apt show linux-image-amd64
    

You should then use apt install to install it. Once this new kernel is installed you should reboot at the next available opportunity to get the benefits provided by the new kernel version. However, please have a look at Section 5.1.24, “Things to do post upgrade before rebooting” before performing the first reboot after the upgrade.

For the more adventurous there is an easy way to compile your own custom kernel on Debian. Install the kernel sources, provided in the linux-source package. You can make use of the deb-pkg target available in the sources' makefile for building a binary package. More information can be found in the Debian Linux Kernel Handbook, which can also be found as the debian-kernel-handbook package.

If possible, it is to your advantage to upgrade the kernel package separately from the main full-upgrade to reduce the chances of a temporarily non-bootable system. Note that this should only be done after the minimal upgrade process described in Section 4.4.4, “Minimal system upgrade”.

It is generally advisable to purge removed packages. This is especially true if these have been removed in an earlier release upgrade (e.g. from the upgrade to buster) or they were provided by third-party vendors. In particular, old init.d scripts have been known to cause issues.

	Caution
	Purging a package will generally also purge its log files, so you might want to back them up first.

The following command displays a list of all removed packages that may have configuration files left on the system (if any):

# dpkg -l | awk '/^rc/ { print $2 }'
    

The packages can be removed by using apt purge. Assuming you want to purge all of them in one go, you can use the following command:

# apt purge $(dpkg -l | awk '/^rc/ { print $2 }')
    

If you use aptitude, you can also use the following alternative to the commands above:

# aptitude search '~c'
# aptitude purge '~c'
    

Some packages from buster may have been replaced in bullseye by transitional dummy packages, which are empty placeholders designed to simplify upgrades. If for instance an application that was formerly a single package has been split into several, a transitional package may be provided with the same name as the old package and with appropriate dependencies to cause the new ones to be installed. After this has happened the redundant dummy package can be safely removed.

The package descriptions for transitional dummy packages usually indicate their purpose. However, they are not uniform; in particular, some “dummy” packages are designed to be kept installed, in order to pull in a full software suite, or track the current latest version of some program. You might also find deborphan with the --guess-* options (e.g. --guess-dummy) useful to detect transitional dummy packages on your system.