podman/libpod/define/errors.go

205 lines
9.2 KiB
Go
Raw Normal View History

package define
import (
"errors"
"fmt"
)
var (
// ErrNoSuchCtr indicates the requested container does not exist
ErrNoSuchCtr = errors.New("no such container")
// ErrNoSuchPod indicates the requested pod does not exist
ErrNoSuchPod = errors.New("no such pod")
// ErrNoSuchVolume indicates the requested volume does not exist
ErrNoSuchVolume = errors.New("no such volume")
// ErrNoSuchNetwork indicates the requested network does not exist
ErrNoSuchNetwork = errors.New("network not found")
// ErrNoSuchExecSession indicates that the requested exec session does
// not exist.
ErrNoSuchExecSession = errors.New("no such exec session")
// ErrDepExists indicates that the current object has dependencies and
// cannot be removed before them.
ErrDepExists = errors.New("dependency exists")
// ErrNoAliases indicates that the container does not have any network
// aliases.
ErrNoAliases = errors.New("no aliases for container")
Initial implementation of volume plugins This implements support for mounting and unmounting volumes backed by volume plugins. Support for actually retrieving plugins requires a pull request to land in containers.conf and then that to be vendored, and as such is not yet ready. Given this, this code is only compile tested. However, the code for everything past retrieving the plugin has been written - there is support for creating, removing, mounting, and unmounting volumes, which should allow full functionality once the c/common PR is merged. A major change is the signature of the MountPoint function for volumes, which now, by necessity, returns an error. Named volumes managed by a plugin do not have a mountpoint we control; instead, it is managed entirely by the plugin. As such, we need to cache the path in the DB, and calls to retrieve it now need to access the DB (and may fail as such). Notably absent is support for SELinux relabelling and chowning these volumes. Given that we don't manage the mountpoint for these volumes, I am extremely reluctant to try and modify it - we could easily break the plugin trying to chown or relabel it. Also, we had no less than *5* separate implementations of inspecting a volume floating around in pkg/infra/abi and pkg/api/handlers/libpod. And none of them used volume.Inspect(), the only correct way of inspecting volumes. Remove them all and consolidate to using the correct way. Compat API is likely still doing things the wrong way, but that is an issue for another day. Fixes #4304 Signed-off-by: Matthew Heon <matthew.heon@pm.me>
2020-12-04 21:24:56 +00:00
// ErrMissingPlugin indicates that the requested operation requires a
// plugin that is not present on the system or in the configuration.
ErrMissingPlugin = errors.New("required plugin missing")
// ErrCtrExists indicates a container with the same name or ID already
// exists
ErrCtrExists = errors.New("container already exists")
// ErrPodExists indicates a pod with the same name or ID already exists
ErrPodExists = errors.New("pod already exists")
// ErrImageExists indicates an image with the same ID already exists
ErrImageExists = errors.New("image already exists")
// ErrVolumeExists indicates a volume with the same name already exists
ErrVolumeExists = errors.New("volume already exists")
// ErrExecSessionExists indicates an exec session with the same ID
// already exists.
ErrExecSessionExists = errors.New("exec session already exists")
// ErrNetworkExists indicates that a network with the given name already
// exists.
ErrNetworkExists = errors.New("network already exists")
// ErrCtrStateInvalid indicates a container is in an improper state for
// the requested operation
ErrCtrStateInvalid = errors.New("container state improper")
// ErrExecSessionStateInvalid indicates that an exec session is in an
// improper state for the requested operation
ErrExecSessionStateInvalid = errors.New("exec session state improper")
// ErrVolumeBeingUsed indicates that a volume is being used by at least one container
ErrVolumeBeingUsed = errors.New("volume is being used")
// ErrRuntimeFinalized indicates that the runtime has already been
// created and cannot be modified
ErrRuntimeFinalized = errors.New("runtime has been finalized")
// ErrCtrFinalized indicates that the container has already been created
// and cannot be modified
ErrCtrFinalized = errors.New("container has been finalized")
// ErrPodFinalized indicates that the pod has already been created and
// cannot be modified
ErrPodFinalized = errors.New("pod has been finalized")
// ErrVolumeFinalized indicates that the volume has already been created and
// cannot be modified
ErrVolumeFinalized = errors.New("volume has been finalized")
// ErrInvalidArg indicates that an invalid argument was passed
ErrInvalidArg = errors.New("invalid argument")
// ErrEmptyID indicates that an empty ID was passed
ErrEmptyID = errors.New("name or ID cannot be empty")
// ErrInternal indicates an internal library error
ErrInternal = errors.New("internal libpod error")
// ErrPodPartialFail indicates that a pod operation was only partially
// successful, and some containers within the pod failed.
ErrPodPartialFail = errors.New("some containers failed")
// ErrDetach indicates that an attach session was manually detached by
// the user.
ErrDetach = errors.New("detached from container")
// ErrWillDeadlock indicates that the requested operation will cause a
// deadlock. This is usually caused by upgrade issues, and is resolved
// by renumbering the locks.
ErrWillDeadlock = errors.New("deadlock due to lock mismatch")
// ErrNoCgroups indicates that the container does not have its own
// CGroup.
ErrNoCgroups = errors.New("this container does not have a cgroup")
// ErrNoLogs indicates that this container is not creating a log so log
// operations cannot be performed on it
ErrNoLogs = errors.New("this container is not logging output")
// ErrRootless indicates that the given command cannot but run without
// root.
ErrRootless = errors.New("operation requires root privileges")
// ErrRuntimeStopped indicates that the runtime has already been shut
// down and no further operations can be performed on it
ErrRuntimeStopped = errors.New("runtime has already been stopped")
// ErrCtrStopped indicates that the requested container is not running
// and the requested operation cannot be performed until it is started
ErrCtrStopped = errors.New("container is stopped")
// ErrCtrRemoved indicates that the container has already been removed
// and no further operations can be performed on it
ErrCtrRemoved = errors.New("container has already been removed")
// ErrPodRemoved indicates that the pod has already been removed and no
// further operations can be performed on it
ErrPodRemoved = errors.New("pod has already been removed")
// ErrVolumeRemoved indicates that the volume has already been removed and
// no further operations can be performed on it
ErrVolumeRemoved = errors.New("volume has already been removed")
// ErrExecSessionRemoved indicates that the exec session has already
// been removed and no further operations can be performed on it.
ErrExecSessionRemoved = errors.New("exec session has already been removed")
// ErrDBClosed indicates that the connection to the state database has
// already been closed
ErrDBClosed = errors.New("database connection already closed")
// ErrDBBadConfig indicates that the database has a different schema or
// was created by a libpod with a different config
ErrDBBadConfig = errors.New("database configuration mismatch")
// ErrNSMismatch indicates that the requested pod or container is in a
// different namespace and cannot be accessed or modified.
ErrNSMismatch = errors.New("target is in a different namespace")
// ErrNotImplemented indicates that the requested functionality is not
// yet present
ErrNotImplemented = errors.New("not yet implemented")
// ErrOSNotSupported indicates the function is not available on the particular
// OS.
ErrOSNotSupported = errors.New("no support for this OS yet")
// ErrOCIRuntime indicates a generic error from the OCI runtime
ErrOCIRuntime = errors.New("OCI runtime error")
// ErrOCIRuntimePermissionDenied indicates the OCI runtime attempted to invoke a command that returned
// a permission denied error
ErrOCIRuntimePermissionDenied = errors.New("OCI permission denied")
// ErrOCIRuntimeNotFound indicates the OCI runtime attempted to invoke a command
// that was not found
ErrOCIRuntimeNotFound = errors.New("OCI runtime attempted to invoke a command that was not found")
// ErrOCIRuntimeUnavailable indicates that the OCI runtime associated to a container
// could not be found in the configuration
ErrOCIRuntimeUnavailable = errors.New("OCI unavailable")
// ErrConmonOutdated indicates the version of conmon found (whether via the configuration or $PATH)
// is out of date for the current podman version
ErrConmonOutdated = errors.New("outdated conmon version")
Ensure Conmon is alive before waiting for exit file This came out of a conversation with Valentin about systemd-managed Podman. He discovered that unit files did not properly handle cases where Conmon was dead - the ExecStopPost `podman rm --force` line was not actually removing the container, but interestingly, adding a `podman cleanup --rm` line would remove it. Both of these commands do the same thing (minus the `podman cleanup --rm` command not force-removing running containers). Without a running Conmon instance, the container process is still running (assuming you killed Conmon with SIGKILL and it had no chance to kill the container it managed), but you can still kill the container itself with `podman stop` - Conmon is not involved, only the OCI Runtime. (`podman rm --force` and `podman stop` use the same code to kill the container). The problem comes when we want to get the container's exit code - we expect Conmon to make us an exit file, which it's obviously not going to do, being dead. The first `podman rm` would fail because of this, but importantly, it would (after failing to retrieve the exit code correctly) set container status to Exited, so that the second `podman cleanup` process would succeed. To make sure the first `podman rm --force` succeeds, we need to catch the case where Conmon is already dead, and instead of waiting for an exit file that will never come, immediately set the Stopped state and remove an error that can be caught and handled. Signed-off-by: Matthew Heon <mheon@redhat.com>
2020-06-08 17:34:12 +00:00
// ErrConmonDead indicates that the container's conmon process has been
// killed, preventing normal operation.
ErrConmonDead = errors.New("conmon process killed")
// ErrNetworkOnPodContainer indicates the user wishes to alter network attributes on a container
// in a pod. This cannot be done as the infra container has all the network information
ErrNetworkOnPodContainer = errors.New("network cannot be configured when it is shared with a pod")
// ErrNetworkInUse indicates the requested operation failed because the network was in use
ErrNetworkInUse = errors.New("network is being used")
// ErrStoreNotInitialized indicates that the container storage was never
// initialized.
ErrStoreNotInitialized = errors.New("the container storage was never initialized")
// ErrNoNetwork indicates that a container has no net namespace, like network=none
ErrNoNetwork = errors.New("container has no network namespace")
// ErrNetworkModeInvalid indicates that a container has the wrong network mode for an operation
ErrNetworkModeInvalid = errors.New("invalid network mode")
// ErrSetSecurityAttribute indicates that a request to set a container's security attribute
// was not possible.
ErrSetSecurityAttribute = fmt.Errorf("%w: unable to assign security attribute", ErrOCIRuntime)
// ErrGetSecurityAttribute indicates that a request to get a container's security attribute
// was not possible.
ErrGetSecurityAttribute = fmt.Errorf("%w: unable to get security attribute", ErrOCIRuntime)
// ErrSecurityAttribute indicates that an error processing security attributes
// for the container
ErrSecurityAttribute = fmt.Errorf("%w: unable to process security attribute", ErrOCIRuntime)
// ErrCanceled indicates that an operation has been cancelled by a user.
// Useful for potentially long running tasks.
ErrCanceled = errors.New("cancelled by user")
// ErrConmonVersionFormat is used when the expected versio-format of conmon
// has changed.
ErrConmonVersionFormat = "conmon version changed format"
)