opensource/podman
1
0
Fork 0
mirror of https://github.com/containers/podman.git synced 2025-05-20 16:47:39 +08:00
Code Issues Packages Projects Releases Wiki Activity

Remove Libpod special-init conditions

Browse Source 
Before this, for some special Podman commands (system reset,
system migrate, system renumber), Podman would create a first
Libpod runtime to do initialization and flag parsing, then stop
that runtime and create an entirely new runtime to perform the
actual task. This is an artifact of the pre-Podman 2.0 days, when
there was almost no indirection between Libpod and the CLI, and
we only used one runtime because we didn't need a second runtime
for flag parsing and basic init.

This system was clunky, and apparently, very buggy. When we
migrated to SQLite, some logic was introduced where we'd select a
different database location based on whether or not Libpod's
StaticDir was manually set - which differed between the first
invocation of Libpod and the second. So we'd get a different
database for some commands (like `system reset`) and they would
not be able to see existing containers, meaning they would not
function properly.

The immediate cause is obviously the SQLite behavior, but I'm
certain there's a lot more baggage hiding behind this multiple
Libpod runtime logic, so let's just refactor it out. It doesn't
make sense, and complicates the code. Instead, make Reset,
Renumber, and Migrate methods of the libpod Runtime. For Reset
and Renumber, we can shut the runtime down afterwards to achieve
the desired effect (no valid runtime after). Then pipe all of
them through the ContainerEngine so cmd/podman can access them.

As part of this, remove the SystemEngine part of pkg/domain. This
was supposed to encompass these "special" commands, but every
command in SystemEngine is actually a ContainerEngine command.
Reset, Renumber, Migrate - they all need a full Libpod and access
to all containers. There's no point to a separate engine if it
just wraps Libpod in the exact same way as ContainerEngine. This
consolidation saves us a bit more code and complexity.

Signed-off-by: Matt Heon <mheon@redhat.com>
This commit is contained in:
Matt Heon
2024-01-11 13:53:34 -05:00
committed by Matthew Heon
parent a60fe34fde
commit b94be90a16
19 changed files with 174 additions and 284 deletions
Download Patch File Download Diff File Expand all files Collapse all files

50
libpod/options.go
View File

@ -454,9 +454,10 @@ func WithDefaultInfraCommand(cmd string) RuntimeOption {
}
}
// WithReset instructs libpod to reset all storage to factory defaults.
// All containers, pods, volumes, images, and networks will be removed.
// All directories created by Libpod will be removed.
// WithReset tells Libpod that the runtime will be used to perform a system
// reset. A number of checks at initialization are relaxed as the runtime is
// going to be used to remove all containers, pods, volumes, images, and
// networks.
func WithReset() RuntimeOption {
return func(rt *Runtime) error {
if rt.valid {
@ -469,10 +470,8 @@ func WithReset() RuntimeOption {
}
}
// WithRenumber instructs libpod to perform a lock renumbering while
// initializing. This will handle migrations from early versions of libpod with
// file locks to newer versions with SHM locking, as well as changes in the
// number of configured locks.
// WithRenumber tells Libpod that the runtime will be used to perform a system
// renumber. A number of checks on initialization related to locks are relaxed.
func WithRenumber() RuntimeOption {
return func(rt *Runtime) error {
if rt.valid {
@ -485,43 +484,6 @@ func WithRenumber() RuntimeOption {
}
}
// WithMigrate instructs libpod to migrate container configurations to account
// for changes between Engine versions. All running containers will be stopped
// during a migration, then restarted after the migration is complete.
func WithMigrate() RuntimeOption {
return func(rt *Runtime) error {
if rt.valid {
return define.ErrRuntimeFinalized
}
rt.doMigrate = true
return nil
}
}
// WithMigrateRuntime instructs Engine to change the default OCI runtime on all
// containers during a migration. This is not used if `MigrateRuntime()` is not
// also passed.
// Engine makes no promises that your containers continue to work with the new
// runtime - migrations between dissimilar runtimes may well break things.
// Use with caution.
func WithMigrateRuntime(requestedRuntime string) RuntimeOption {
return func(rt *Runtime) error {
if rt.valid {
return define.ErrRuntimeFinalized
}
if requestedRuntime == "" {
return fmt.Errorf("must provide a non-empty name for new runtime: %w", define.ErrInvalidArg)
}
rt.migrateRuntime = requestedRuntime
return nil
}
}
// WithEventsLogger sets the events backend to use.
// Currently supported values are "file" for file backend and "journald" for
// journald backend.