In addition to numerous new and upgraded packages, this release has the following highlights: Support is planned until the end of October 2021, handing over to 21.09. GNOME desktop environment was upgraded to 3.38, see its release notes.

The following new services were added since the last release: Keycloak, an open source identity and access management server with support for OpenID Connect, OAUTH 2.0 and SAML 2.0. See the Keycloak section of the NixOS manual for more information.

When upgrading from a previous release, please be aware of the following incompatible changes: systemd-journal2gelf no longer parses json and expects the receiving system to handle it. How to achieve this with Graylog is described in this GitHub issue. If the services.dbus module is enabled, then the user D-Bus session is now always socket activated. The associated options services.dbus.socketActivated and services.xserver.startDbusSession have therefore been removed and you will receive a warning if they are present in your configuration. This change makes the user D-Bus session available also for non-graphical logins. The option fonts.enableFontDir has been renamed to . The path of font directory has also been changed to /run/current-system/sw/share/X11/fonts, for consistency with other X11 resources. A number of options have been renamed in the kicad interface. oceSupport has been renamed to withOCE, withOCCT has been renamed to withOCC, ngspiceSupport has been renamed to withNgspice, and scriptingSupport has been renamed to withScripting. Additionally, kicad/base.nix no longer provides default argument values since these are provided by kicad/default.nix. The socket for the pdns-recursor module was moved from /var/lib/pdns-recursor to /run/pdns-recursor to match upstream. PowerDNS has been updated from 4.2.x to 4.3.x. Please be sure to review the Upgrade Notes provided by upstream before upgrading. Worth specifically noting is that the service now runs entirely as a dedicated pdns user, instead of starting as root and dropping privileges, as well as the default socket-dir location changing from /var/lib/powerdns to /run/pdns. btc1 has been abandoned upstream, and removed. riak-cs package removed along with services.riak-cs module. stanchion package removed along with services.stanchion module. mutt has been updated to a new major version (2.x), which comes with some backward incompatible changes that are described in the release notes for Mutt 2.0. vim switched to Python 3, dropping all Python 2 support. boot.zfs.forceImportAll previously did nothing, but has been fixed. However its default has been changed to false to preserve the existing default behaviour. If you have this explicitly set to true, please note that your non-root pools will now be forcibly imported.

The default-version of nextcloud is nextcloud20. Please note that it's not possible to upgrade nextcloud across multiple major versions! This means that it's e.g. not possible to upgrade from nextcloud18 to nextcloud20 in a single deploy. The package can be manually upgraded by setting to nextcloud20. The setting defaults to 127.0.0.1 now, making Redis listen on the loopback interface only, and not all public network interfaces. NixOS now emits a deprecation warning if systemd's StartLimitInterval setting is used in a serviceConfig section instead of in a unitConfig; that setting is deprecated and now undocumented for the service section by systemd upstream, but still effective and somewhat buggy there, which can be confusing. See #45785 for details. All services should use or StartLimitIntervalSec in instead. The Unbound DNS resolver service (services.unbound) has been refactored to allow reloading, control sockets and to fix startup ordering issues. It is now possible to enable a local UNIX control socket for unbound by setting the option. Previously we just applied a very minimal set of restrictions and trusted unbound to properly drop root privs and capabilities. As of this we are (for the most part) just using the upstream example unit file for unbound. The main difference is that we start unbound as unbound user with the required capabilities instead of letting unbound do the chroot & uid/gid changes. The upstream unit configuration this is based on is a lot stricter with all kinds of permissions then our previous variant. It also came with the default of having the Type set to notify, therefore we are now also using the unbound-with-systemd package here. Unbound will start up, read the configuration files and start listening on the configured ports before systemd will declare the unit active (running). This will likely help with startup order and the occasional race condition during system activation where the DNS service is started but not yet ready to answer queries. Services depending on nss-lookup.target or unbound.service are now be able to use unbound when those targets have been reached. Aditionally to the much stricter runtime environmet the /dev/urandom mount lines we previously had in the code (that would randomly failed during the stop-phase) have been removed as systemd will take care of those for us. The preStart script is now only required if we enabled the trust anchor updates (which are still enabled by default). Another benefit of the refactoring is that we can now issue reloads via either pkill -HUP unbound and systemctl reload unbound to reload the running configuration without taking the daemon offline. A prerequisite of this was that unbound configuration is available on a well known path on the file system. We are using the path /etc/unbound/unbound.conf as that is the default in the CLI tooling which in turn enables us to use unbound-control without passing a custom configuration location.