aboutsummaryrefslogtreecommitdiff
path: root/libpod/container.go
blob: c57250d728dcd3fc1154bdd17040f5f32e5cc632 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
package libpod

import (
	"bytes"
	"fmt"
	"io/ioutil"
	"net"
	"os"
	"time"

	"github.com/containernetworking/cni/pkg/types"
	cnitypes "github.com/containernetworking/cni/pkg/types/current"
	"github.com/containers/common/pkg/secrets"
	"github.com/containers/image/v5/manifest"
	"github.com/containers/podman/v3/libpod/define"
	"github.com/containers/podman/v3/libpod/lock"
	"github.com/containers/storage"
	"github.com/cri-o/ocicni/pkg/ocicni"
	spec "github.com/opencontainers/runtime-spec/specs-go"
	"github.com/pkg/errors"
	"github.com/sirupsen/logrus"
)

// CgroupfsDefaultCgroupParent is the cgroup parent for CGroupFS in libpod
const CgroupfsDefaultCgroupParent = "/libpod_parent"

// SystemdDefaultCgroupParent is the cgroup parent for the systemd cgroup
// manager in libpod
const SystemdDefaultCgroupParent = "machine.slice"

// SystemdDefaultRootlessCgroupParent is the cgroup parent for the systemd cgroup
// manager in libpod when running as rootless
const SystemdDefaultRootlessCgroupParent = "user.slice"

// DefaultWaitInterval is the default interval between container status checks
// while waiting.
const DefaultWaitInterval = 250 * time.Millisecond

// LinuxNS represents a Linux namespace
type LinuxNS int

const (
	// InvalidNS is an invalid namespace
	InvalidNS LinuxNS = iota
	// IPCNS is the IPC namespace
	IPCNS LinuxNS = iota
	// MountNS is the mount namespace
	MountNS LinuxNS = iota
	// NetNS is the network namespace
	NetNS LinuxNS = iota
	// PIDNS is the PID namespace
	PIDNS LinuxNS = iota
	// UserNS is the user namespace
	UserNS LinuxNS = iota
	// UTSNS is the UTS namespace
	UTSNS LinuxNS = iota
	// CgroupNS is the CGroup namespace
	CgroupNS LinuxNS = iota
)

// String returns a string representation of a Linux namespace
// It is guaranteed to be the name of the namespace in /proc for valid ns types
func (ns LinuxNS) String() string {
	switch ns {
	case InvalidNS:
		return "invalid"
	case IPCNS:
		return "ipc"
	case MountNS:
		return "mnt"
	case NetNS:
		return "net"
	case PIDNS:
		return "pid"
	case UserNS:
		return "user"
	case UTSNS:
		return "uts"
	case CgroupNS:
		return "cgroup"
	default:
		return "unknown"
	}
}

// Container is a single OCI container.
// All operations on a Container that access state must begin with a call to
// syncContainer().
// There is no guarantee that state exists in a readable state before
// syncContainer() is run, and even if it does, its contents will be out of date
// and must be refreshed from the database.
// Generally, this requirement applies only to top-level functions; helpers can
// assume that their callers handled this requirement. Generally speaking, if a
// function takes the container lock and accesses any part of state, it should
// syncContainer() immediately after locking.
type Container struct {
	config *ContainerConfig

	state *ContainerState

	// Batched indicates that a container has been locked as part of a
	// Batch() operation
	// Functions called on a batched container will not lock or sync
	batched bool

	valid      bool
	lock       lock.Locker
	runtime    *Runtime
	ociRuntime OCIRuntime

	rootlessSlirpSyncR *os.File
	rootlessSlirpSyncW *os.File

	rootlessPortSyncR *os.File
	rootlessPortSyncW *os.File

	// A restored container should have the same IP address as before
	// being checkpointed. If requestedIP is set it will be used instead
	// of config.StaticIP.
	requestedIP net.IP
	// A restored container should have the same MAC address as before
	// being checkpointed. If requestedMAC is set it will be used instead
	// of config.StaticMAC.
	requestedMAC net.HardwareAddr

	// This is true if a container is restored from a checkpoint.
	restoreFromCheckpoint bool

	// Used to query the NOTIFY_SOCKET once along with setting up
	// mounts etc.
	notifySocket string

	slirp4netnsSubnet *net.IPNet
}

// ContainerState contains the current state of the container
// It is stored on disk in a tmpfs and recreated on reboot
type ContainerState struct {
	// The current state of the running container
	State define.ContainerStatus `json:"state"`
	// The path to the JSON OCI runtime spec for this container
	ConfigPath string `json:"configPath,omitempty"`
	// RunDir is a per-boot directory for container content
	RunDir string `json:"runDir,omitempty"`
	// Mounted indicates whether the container's storage has been mounted
	// for use
	Mounted bool `json:"mounted,omitempty"`
	// Mountpoint contains the path to the container's mounted storage as given
	// by containers/storage.
	Mountpoint string `json:"mountPoint,omitempty"`
	// StartedTime is the time the container was started
	StartedTime time.Time `json:"startedTime,omitempty"`
	// FinishedTime is the time the container finished executing
	FinishedTime time.Time `json:"finishedTime,omitempty"`
	// ExitCode is the exit code returned when the container stopped
	ExitCode int32 `json:"exitCode,omitempty"`
	// Exited is whether the container has exited
	Exited bool `json:"exited,omitempty"`
	// OOMKilled indicates that the container was killed as it ran out of
	// memory
	OOMKilled bool `json:"oomKilled,omitempty"`
	// PID is the PID of a running container
	PID int `json:"pid,omitempty"`
	// ConmonPID is the PID of the container's conmon
	ConmonPID int `json:"conmonPid,omitempty"`
	// ExecSessions contains all exec sessions that are associated with this
	// container.
	ExecSessions map[string]*ExecSession `json:"newExecSessions,omitempty"`
	// LegacyExecSessions are legacy exec sessions from older versions of
	// Podman.
	// These are DEPRECATED and will be removed in a future release.
	LegacyExecSessions map[string]*legacyExecSession `json:"execSessions,omitempty"`
	// NetworkStatus contains the configuration results for all networks
	// the pod is attached to. Only populated if we created a network
	// namespace for the container, and the network namespace is currently
	// active
	NetworkStatus []*cnitypes.Result `json:"networkResults,omitempty"`
	// BindMounts contains files that will be bind-mounted into the
	// container when it is mounted.
	// These include /etc/hosts and /etc/resolv.conf
	// This maps the path the file will be mounted to in the container to
	// the path of the file on disk outside the container
	BindMounts map[string]string `json:"bindMounts,omitempty"`
	// StoppedByUser indicates whether the container was stopped by an
	// explicit call to the Stop() API.
	StoppedByUser bool `json:"stoppedByUser,omitempty"`
	// RestartPolicyMatch indicates whether the conditions for restart
	// policy have been met.
	RestartPolicyMatch bool `json:"restartPolicyMatch,omitempty"`
	// RestartCount is how many times the container was restarted by its
	// restart policy. This is NOT incremented by normal container restarts
	// (only by restart policy).
	RestartCount uint `json:"restartCount,omitempty"`

	// ExtensionStageHooks holds hooks which will be executed by libpod
	// and not delegated to the OCI runtime.
	ExtensionStageHooks map[string][]spec.Hook `json:"extensionStageHooks,omitempty"`

	// NetInterfaceDescriptions describe the relationship between a CNI
	// network and an interface names
	NetInterfaceDescriptions ContainerNetworkDescriptions `json:"networkDescriptions,omitempty"`

	// containerPlatformState holds platform-specific container state.
	containerPlatformState
}

// ContainerNamedVolume is a named volume that will be mounted into the
// container. Each named volume is a libpod Volume present in the state.
type ContainerNamedVolume struct {
	// Name is the name of the volume to mount in.
	// Must resolve to a valid volume present in this Podman.
	Name string `json:"volumeName"`
	// Dest is the mount's destination
	Dest string `json:"dest"`
	// Options are fstab style mount options
	Options []string `json:"options,omitempty"`
}

// ContainerOverlayVolume is a overlay volume that will be mounted into the
// container. Each volume is a libpod Volume present in the state.
type ContainerOverlayVolume struct {
	// Destination is the absolute path where the mount will be placed in the container.
	Dest string `json:"dest"`
	// Source specifies the source path of the mount.
	Source string `json:"source,omitempty"`
	// Options holds overlay volume options.
	Options []string `json:"options,omitempty"`
}

// ContainerImageVolume is a volume based on a container image.  The container
// image is first mounted on the host and is then bind-mounted into the
// container.
type ContainerImageVolume struct {
	// Source is the source of the image volume.  The image can be referred
	// to by name and by ID.
	Source string `json:"source"`
	// Dest is the absolute path of the mount in the container.
	Dest string `json:"dest"`
	// ReadWrite sets the volume writable.
	ReadWrite bool `json:"rw"`
}

// ContainerSecret is a secret that is mounted in a container
type ContainerSecret struct {
	// Secret is the secret
	*secrets.Secret
	// UID is the UID of the secret file
	UID uint32
	// GID is the GID of the secret file
	GID uint32
	// Mode is the mode of the secret file
	Mode uint32
}

// ContainerNetworkDescriptions describes the relationship between the CNI
// network and the ethN where N is an integer
type ContainerNetworkDescriptions map[string]int

// Config accessors
// Unlocked

// Config returns the configuration used to create the container
func (c *Container) Config() *ContainerConfig {
	returnConfig := new(ContainerConfig)
	if err := JSONDeepCopy(c.config, returnConfig); err != nil {
		return nil
	}

	return returnConfig
}

// Runtime returns the container's Runtime.
func (c *Container) Runtime() *Runtime {
	return c.runtime
}

// Spec returns the container's OCI runtime spec
// The spec returned is the one used to create the container. The running
// spec may differ slightly as mounts are added based on the image
func (c *Container) Spec() *spec.Spec {
	returnSpec := new(spec.Spec)
	if err := JSONDeepCopy(c.config.Spec, returnSpec); err != nil {
		return nil
	}

	return returnSpec
}

// specFromState returns the unmarshalled json config of the container.  If the
// config does not exist (e.g., because the container was never started) return
// the spec from the config.
func (c *Container) specFromState() (*spec.Spec, error) {
	returnSpec := c.config.Spec

	if f, err := os.Open(c.state.ConfigPath); err == nil {
		returnSpec = new(spec.Spec)
		content, err := ioutil.ReadAll(f)
		if err != nil {
			return nil, errors.Wrapf(err, "error reading container config")
		}
		if err := json.Unmarshal(content, &returnSpec); err != nil {
			return nil, errors.Wrapf(err, "error unmarshalling container config")
		}
	} else if !os.IsNotExist(err) {
		// ignore when the file does not exist
		return nil, errors.Wrapf(err, "error opening container config")
	}

	return returnSpec, nil
}

// ID returns the container's ID
func (c *Container) ID() string {
	return c.config.ID
}

// Name returns the container's name
func (c *Container) Name() string {
	return c.config.Name
}

// PodID returns the full ID of the pod the container belongs to, or "" if it
// does not belong to a pod
func (c *Container) PodID() string {
	return c.config.Pod
}

// Namespace returns the libpod namespace the container is in.
// Namespaces are used to logically separate containers and pods in the state.
func (c *Container) Namespace() string {
	return c.config.Namespace
}

// Image returns the ID and name of the image used as the container's rootfs.
func (c *Container) Image() (string, string) {
	return c.config.RootfsImageID, c.config.RootfsImageName
}

// RawImageName returns the unprocessed and not-normalized user-specified image
// name.
func (c *Container) RawImageName() string {
	return c.config.RawImageName
}

// ShmDir returns the sources path to be mounted on /dev/shm in container
func (c *Container) ShmDir() string {
	return c.config.ShmDir
}

// ShmSize returns the size of SHM device to be mounted into the container
func (c *Container) ShmSize() int64 {
	return c.config.ShmSize
}

// StaticDir returns the directory used to store persistent container files
func (c *Container) StaticDir() string {
	return c.config.StaticDir
}

// NamedVolumes returns the container's named volumes.
// The name of each is guaranteed to point to a valid libpod Volume present in
// the state.
func (c *Container) NamedVolumes() []*ContainerNamedVolume {
	volumes := []*ContainerNamedVolume{}
	for _, vol := range c.config.NamedVolumes {
		newVol := new(ContainerNamedVolume)
		newVol.Name = vol.Name
		newVol.Dest = vol.Dest
		newVol.Options = vol.Options
		volumes = append(volumes, newVol)
	}

	return volumes
}

// Privileged returns whether the container is privileged
func (c *Container) Privileged() bool {
	return c.config.Privileged
}

// ProcessLabel returns the selinux ProcessLabel of the container
func (c *Container) ProcessLabel() string {
	return c.config.ProcessLabel
}

// MountLabel returns the SELinux mount label of the container
func (c *Container) MountLabel() string {
	return c.config.MountLabel
}

// Systemd returns whether the container will be running in systemd mode
func (c *Container) Systemd() bool {
	return c.config.Systemd
}

// User returns the user who the container is run as
func (c *Container) User() string {
	return c.config.User
}

// Dependencies gets the containers this container depends upon
func (c *Container) Dependencies() []string {
	// Collect in a map first to remove dupes
	dependsCtrs := map[string]bool{}

	// First add all namespace containers
	if c.config.IPCNsCtr != "" {
		dependsCtrs[c.config.IPCNsCtr] = true
	}
	if c.config.MountNsCtr != "" {
		dependsCtrs[c.config.MountNsCtr] = true
	}
	if c.config.NetNsCtr != "" {
		dependsCtrs[c.config.NetNsCtr] = true
	}
	if c.config.PIDNsCtr != "" {
		dependsCtrs[c.config.PIDNsCtr] = true
	}
	if c.config.UserNsCtr != "" {
		dependsCtrs[c.config.UserNsCtr] = true
	}
	if c.config.UTSNsCtr != "" {
		dependsCtrs[c.config.UTSNsCtr] = true
	}
	if c.config.CgroupNsCtr != "" {
		dependsCtrs[c.config.CgroupNsCtr] = true
	}

	// Add all generic dependencies
	for _, id := range c.config.Dependencies {
		dependsCtrs[id] = true
	}

	if len(dependsCtrs) == 0 {
		return []string{}
	}

	depends := make([]string, 0, len(dependsCtrs))
	for ctr := range dependsCtrs {
		depends = append(depends, ctr)
	}

	return depends
}

// NewNetNS returns whether the container will create a new network namespace
func (c *Container) NewNetNS() bool {
	return c.config.CreateNetNS
}

// PortMappings returns the ports that will be mapped into a container if
// a new network namespace is created
// If NewNetNS() is false, this value is unused
func (c *Container) PortMappings() ([]ocicni.PortMapping, error) {
	// First check if the container belongs to a network namespace (like a pod)
	if len(c.config.NetNsCtr) > 0 {
		netNsCtr, err := c.runtime.GetContainer(c.config.NetNsCtr)
		if err != nil {
			return nil, errors.Wrapf(err, "unable to lookup network namespace for container %s", c.ID())
		}
		return netNsCtr.PortMappings()
	}
	return c.config.PortMappings, nil
}

// DNSServers returns DNS servers that will be used in the container's
// resolv.conf
// If empty, DNS server from the host's resolv.conf will be used instead
func (c *Container) DNSServers() []net.IP {
	return c.config.DNSServer
}

// DNSSearch returns the DNS search domains that will be used in the container's
// resolv.conf
// If empty, DNS Search domains from the host's resolv.conf will be used instead
func (c *Container) DNSSearch() []string {
	return c.config.DNSSearch
}

// DNSOption returns the DNS options that will be used in the container's
// resolv.conf
// If empty, options from the host's resolv.conf will be used instead
func (c *Container) DNSOption() []string {
	return c.config.DNSOption
}

// HostsAdd returns hosts that will be added to the container's hosts file
// The host system's hosts file is used as a base, and these are appended to it
func (c *Container) HostsAdd() []string {
	return c.config.HostAdd
}

// UserVolumes returns user-added volume mounts in the container.
// These are not added to the spec, but are used during image commit and to
// trigger some OCI hooks.
func (c *Container) UserVolumes() []string {
	volumes := make([]string, 0, len(c.config.UserVolumes))
	volumes = append(volumes, c.config.UserVolumes...)
	return volumes
}

// Entrypoint is the container's entrypoint.
// This is not added to the spec, but is instead used during image commit.
func (c *Container) Entrypoint() []string {
	entrypoint := make([]string, 0, len(c.config.Entrypoint))
	entrypoint = append(entrypoint, c.config.Entrypoint...)
	return entrypoint
}

// Command is the container's command
// This is not added to the spec, but is instead used during image commit
func (c *Container) Command() []string {
	command := make([]string, 0, len(c.config.Command))
	command = append(command, c.config.Command...)
	return command
}

// Stdin returns whether STDIN on the container will be kept open
func (c *Container) Stdin() bool {
	return c.config.Stdin
}

// Labels returns the container's labels
func (c *Container) Labels() map[string]string {
	labels := make(map[string]string)
	for key, value := range c.config.Labels {
		labels[key] = value
	}
	return labels
}

// StopSignal is the signal that will be used to stop the container
// If it fails to stop the container, SIGKILL will be used after a timeout
// If StopSignal is 0, the default signal of SIGTERM will be used
func (c *Container) StopSignal() uint {
	return c.config.StopSignal
}

// StopTimeout returns the container's stop timeout
// If the container's default stop signal fails to kill the container, SIGKILL
// will be used after this timeout
func (c *Container) StopTimeout() uint {
	return c.config.StopTimeout
}

// CreatedTime gets the time when the container was created
func (c *Container) CreatedTime() time.Time {
	return c.config.CreatedTime
}

// CgroupParent gets the container's CGroup parent
func (c *Container) CgroupParent() string {
	return c.config.CgroupParent
}

// LogPath returns the path to the container's log file
// This file will only be present after Init() is called to create the container
// in the runtime
func (c *Container) LogPath() string {
	return c.config.LogPath
}

// LogTag returns the tag to the container's log file
func (c *Container) LogTag() string {
	return c.config.LogTag
}

// RestartPolicy returns the container's restart policy.
func (c *Container) RestartPolicy() string {
	return c.config.RestartPolicy
}

// RestartRetries returns the number of retries that will be attempted when
// using the "on-failure" restart policy
func (c *Container) RestartRetries() uint {
	return c.config.RestartRetries
}

// LogDriver returns the log driver for this container
func (c *Container) LogDriver() string {
	return c.config.LogDriver
}

// RuntimeName returns the name of the runtime
func (c *Container) RuntimeName() string {
	return c.config.OCIRuntime
}

// Runtime spec accessors
// Unlocked

// Hostname gets the container's hostname
func (c *Container) Hostname() string {
	if c.config.Spec.Hostname != "" {
		return c.config.Spec.Hostname
	}

	if len(c.ID()) < 11 {
		return c.ID()
	}
	return c.ID()[:12]
}

// WorkingDir returns the containers working dir
func (c *Container) WorkingDir() string {
	if c.config.Spec.Process != nil {
		return c.config.Spec.Process.Cwd
	}
	return "/"
}

// State Accessors
// Require locking

// State returns the current state of the container
func (c *Container) State() (define.ContainerStatus, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return define.ContainerStateUnknown, err
		}
	}
	return c.state.State, nil
}

// Mounted returns whether the container is mounted and the path it is mounted
// at (if it is mounted).
// If the container is not mounted, no error is returned, and the mountpoint
// will be set to "".
func (c *Container) Mounted() (bool, string, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return false, "", errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	// We cannot directly return c.state.Mountpoint as it is not guaranteed
	// to be set if the container is mounted, only if the container has been
	// prepared with c.prepare().
	// Instead, let's call into c/storage
	mountedTimes, err := c.runtime.storageService.MountedContainerImage(c.ID())
	if err != nil {
		return false, "", err
	}

	if mountedTimes > 0 {
		mountPoint, err := c.runtime.storageService.GetMountpoint(c.ID())
		if err != nil {
			return false, "", err
		}

		return true, mountPoint, nil
	}

	return false, "", nil
}

// StartedTime is the time the container was started
func (c *Container) StartedTime() (time.Time, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return time.Time{}, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.state.StartedTime, nil
}

// FinishedTime is the time the container was stopped
func (c *Container) FinishedTime() (time.Time, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return time.Time{}, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.state.FinishedTime, nil
}

// ExitCode returns the exit code of the container as
// an int32, and whether the container has exited.
// If the container has not exited, exit code will always be 0.
// If the container restarts, the exit code is reset to 0.
func (c *Container) ExitCode() (int32, bool, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return 0, false, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.state.ExitCode, c.state.Exited, nil
}

// OOMKilled returns whether the container was killed by an OOM condition
func (c *Container) OOMKilled() (bool, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return false, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.state.OOMKilled, nil
}

// PID returns the PID of the container.
// If the container is not running, a pid of 0 will be returned. No error will
// occur.
func (c *Container) PID() (int, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return -1, err
		}
	}

	return c.state.PID, nil
}

// ConmonPID Returns the PID of the container's conmon process.
// If the container is not running, a PID of 0 will be returned. No error will
// occur.
func (c *Container) ConmonPID() (int, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return -1, err
		}
	}

	return c.state.ConmonPID, nil
}

// ExecSessions retrieves active exec sessions running in the container
func (c *Container) ExecSessions() ([]string, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}

	ids := make([]string, 0, len(c.state.ExecSessions))
	for id := range c.state.ExecSessions {
		ids = append(ids, id)
	}

	return ids, nil
}

// ExecSession retrieves detailed information on a single active exec session in
// a container
func (c *Container) ExecSession(id string) (*ExecSession, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}

	session, ok := c.state.ExecSessions[id]
	if !ok {
		return nil, errors.Wrapf(define.ErrNoSuchExecSession, "no exec session with ID %s found in container %s", id, c.ID())
	}

	returnSession := new(ExecSession)
	if err := JSONDeepCopy(session, returnSession); err != nil {
		return nil, errors.Wrapf(err, "error copying contents of container %s exec session %s", c.ID(), session.ID())
	}

	return returnSession, nil
}

// IPs retrieves a container's IP address(es)
// This will only be populated if the container is configured to created a new
// network namespace, and that namespace is presently active
func (c *Container) IPs() ([]net.IPNet, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}

	if !c.config.CreateNetNS {
		return nil, errors.Wrapf(define.ErrInvalidArg, "container %s network namespace is not managed by libpod", c.ID())
	}

	ips := make([]net.IPNet, 0)

	for _, r := range c.state.NetworkStatus {
		for _, ip := range r.IPs {
			ips = append(ips, ip.Address)
		}
	}

	return ips, nil
}

// Routes retrieves a container's routes
// This will only be populated if the container is configured to created a new
// network namespace, and that namespace is presently active
func (c *Container) Routes() ([]types.Route, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}

	if !c.config.CreateNetNS {
		return nil, errors.Wrapf(define.ErrInvalidArg, "container %s network namespace is not managed by libpod", c.ID())
	}

	routes := make([]types.Route, 0)

	for _, r := range c.state.NetworkStatus {
		for _, route := range r.Routes {
			newRoute := types.Route{
				Dst: route.Dst,
				GW:  route.GW,
			}
			routes = append(routes, newRoute)
		}
	}

	return routes, nil
}

// BindMounts retrieves bind mounts that were created by libpod and will be
// added to the container
// All these mounts except /dev/shm are ignored if a mount in the given spec has
// the same destination
// These mounts include /etc/resolv.conf, /etc/hosts, and /etc/hostname
// The return is formatted as a map from destination (mountpoint in the
// container) to source (path of the file that will be mounted into the
// container)
// If the container has not been started yet, an empty map will be returned, as
// the files in question are only created when the container is started.
func (c *Container) BindMounts() (map[string]string, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}

	newMap := make(map[string]string, len(c.state.BindMounts))

	for key, val := range c.state.BindMounts {
		newMap[key] = val
	}

	return newMap, nil
}

// StoppedByUser returns whether the container was last stopped by an explicit
// call to the Stop() API, or whether it exited naturally.
func (c *Container) StoppedByUser() (bool, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return false, err
		}
	}

	return c.state.StoppedByUser, nil
}

// Misc Accessors
// Most will require locking

// NamespacePath returns the path of one of the container's namespaces
// If the container is not running, an error will be returned
func (c *Container) NamespacePath(linuxNS LinuxNS) (string, error) { //nolint:interfacer
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return "", errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}

	return c.namespacePath(linuxNS)
}

// namespacePath returns the path of one of the container's namespaces
// If the container is not running, an error will be returned
func (c *Container) namespacePath(linuxNS LinuxNS) (string, error) { //nolint:interfacer
	if c.state.State != define.ContainerStateRunning && c.state.State != define.ContainerStatePaused {
		return "", errors.Wrapf(define.ErrCtrStopped, "cannot get namespace path unless container %s is running", c.ID())
	}

	if linuxNS == InvalidNS {
		return "", errors.Wrapf(define.ErrInvalidArg, "invalid namespace requested from container %s", c.ID())
	}

	return fmt.Sprintf("/proc/%d/ns/%s", c.state.PID, linuxNS.String()), nil
}

// CgroupManager returns the cgroup manager used by the given container.
func (c *Container) CgroupManager() string {
	cgroupManager := c.config.CgroupManager
	if cgroupManager == "" {
		cgroupManager = c.runtime.config.Engine.CgroupManager
	}
	return cgroupManager
}

// CGroupPath returns a cgroups "path" for the given container.
// Note that the container must be running.  Otherwise, an error
// is returned.
func (c *Container) CGroupPath() (string, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return "", errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.cGroupPath()
}

// cGroupPath returns a cgroups "path" for the given container.
// Note that the container must be running.  Otherwise, an error
// is returned.
// NOTE: only call this when owning the container's lock.
func (c *Container) cGroupPath() (string, error) {
	if c.config.NoCgroups || c.config.CgroupsMode == "disabled" {
		return "", errors.Wrapf(define.ErrNoCgroups, "this container is not creating cgroups")
	}
	if c.state.State != define.ContainerStateRunning && c.state.State != define.ContainerStatePaused {
		return "", errors.Wrapf(define.ErrCtrStopped, "cannot get cgroup path unless container %s is running", c.ID())
	}

	// Read /proc/{PID}/cgroup and find the *longest* cgroup entry.  That's
	// needed to account for hacks in cgroups v1, where each line in the
	// file could potentially point to a cgroup.  The longest one, however,
	// is the libpod-specific one we're looking for.
	//
	// See #8397 on the need for the longest-path look up.
	//
	// And another workaround for containers running systemd as the payload.
	// containers running systemd moves themselves into a child subgroup of
	// the named systemd cgroup hierarchy.  Ignore any named cgroups during
	// the lookup.
	// See #10602 for more details.
	procPath := fmt.Sprintf("/proc/%d/cgroup", c.state.PID)
	lines, err := ioutil.ReadFile(procPath)
	if err != nil {
		return "", err
	}

	var cgroupPath string
	for _, line := range bytes.Split(lines, []byte("\n")) {
		// cgroups(7) nails it down to three fields with the 3rd
		// pointing to the cgroup's path which works both on v1 and v2.
		fields := bytes.Split(line, []byte(":"))
		if len(fields) != 3 {
			logrus.Debugf("Error parsing cgroup: expected 3 fields but got %d: %s", len(fields), procPath)
			continue
		}
		// Ignore named cgroups like name=systemd.
		if bytes.Contains(fields[1], []byte("=")) {
			continue
		}
		path := string(fields[2])
		if len(path) > len(cgroupPath) {
			cgroupPath = path
		}
	}

	if len(cgroupPath) == 0 {
		return "", errors.Errorf("could not find any cgroup in %q", procPath)
	}

	return cgroupPath, nil
}

// RootFsSize returns the root FS size of the container
func (c *Container) RootFsSize() (int64, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return -1, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.rootFsSize()
}

// RWSize returns the rw size of the container
func (c *Container) RWSize() (int64, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()
		if err := c.syncContainer(); err != nil {
			return -1, errors.Wrapf(err, "error updating container %s state", c.ID())
		}
	}
	return c.rwSize()
}

// IDMappings returns the UID/GID mapping used for the container
func (c *Container) IDMappings() storage.IDMappingOptions {
	return c.config.IDMappings
}

// RootUID returns the root user mapping from container
func (c *Container) RootUID() int {
	if len(c.config.IDMappings.UIDMap) == 1 && c.config.IDMappings.UIDMap[0].Size == 1 {
		return c.config.IDMappings.UIDMap[0].HostID
	}
	for _, uidmap := range c.config.IDMappings.UIDMap {
		if uidmap.ContainerID == 0 {
			return uidmap.HostID
		}
	}
	return 0
}

// RootGID returns the root user mapping from container
func (c *Container) RootGID() int {
	if len(c.config.IDMappings.GIDMap) == 1 && c.config.IDMappings.GIDMap[0].Size == 1 {
		return c.config.IDMappings.GIDMap[0].HostID
	}
	for _, gidmap := range c.config.IDMappings.GIDMap {
		if gidmap.ContainerID == 0 {
			return gidmap.HostID
		}
	}
	return 0
}

// IsInfra returns whether the container is an infra container
func (c *Container) IsInfra() bool {
	return c.config.IsInfra
}

// IsReadOnly returns whether the container is running in read only mode
func (c *Container) IsReadOnly() bool {
	return c.config.Spec.Root.Readonly
}

// NetworkDisabled returns whether the container is running with a disabled network
func (c *Container) NetworkDisabled() (bool, error) {
	if c.config.NetNsCtr != "" {
		container, err := c.runtime.state.Container(c.config.NetNsCtr)
		if err != nil {
			return false, err
		}
		return container.NetworkDisabled()
	}
	return networkDisabled(c)
}

func networkDisabled(c *Container) (bool, error) {
	if c.config.CreateNetNS {
		return false, nil
	}
	if !c.config.PostConfigureNetNS {
		for _, ns := range c.config.Spec.Linux.Namespaces {
			if ns.Type == spec.NetworkNamespace {
				return ns.Path == "", nil
			}
		}
	}
	return false, nil
}

func (c *Container) HostNetwork() bool {
	if c.config.CreateNetNS || c.config.NetNsCtr != "" {
		return false
	}
	for _, ns := range c.config.Spec.Linux.Namespaces {
		if ns.Type == spec.NetworkNamespace {
			return false
		}
	}
	return true
}

// ContainerState returns containerstate struct
func (c *Container) ContainerState() (*ContainerState, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, err
		}
	}
	returnConfig := new(ContainerState)
	if err := JSONDeepCopy(c.state, returnConfig); err != nil {
		return nil, errors.Wrapf(err, "error copying container %s state", c.ID())
	}
	return c.state, nil
}

// HasHealthCheck returns bool as to whether there is a health check
// defined for the container
func (c *Container) HasHealthCheck() bool {
	return c.config.HealthCheckConfig != nil
}

// HealthCheckConfig returns the command and timing attributes of the health check
func (c *Container) HealthCheckConfig() *manifest.Schema2HealthConfig {
	return c.config.HealthCheckConfig
}

// AutoRemove indicates whether the container will be removed after it is executed
func (c *Container) AutoRemove() bool {
	spec := c.config.Spec
	if spec.Annotations == nil {
		return false
	}
	return c.Spec().Annotations[define.InspectAnnotationAutoremove] == define.InspectResponseTrue
}

// Timezone returns the timezone configured inside the container.
// Local means it has the same timezone as the host machine
func (c *Container) Timezone() string {
	return c.config.Timezone
}

// Umask returns the Umask bits configured inside the container.
func (c *Container) Umask() string {
	return c.config.Umask
}

//Secrets return the secrets in the container
func (c *Container) Secrets() []*ContainerSecret {
	return c.config.Secrets
}

// Networks gets all the networks this container is connected to.
// Please do NOT use ctr.config.Networks, as this can be changed from those
// values at runtime via network connect and disconnect.
// If the container is configured to use CNI and this function returns an empty
// array, the container will still be connected to the default network.
// The second return parameter, a bool, indicates that the container container
// is joining the default CNI network - the network name will be included in the
// returned array of network names, but the container did not explicitly join
// this network.
func (c *Container) Networks() ([]string, bool, error) {
	if !c.batched {
		c.lock.Lock()
		defer c.lock.Unlock()

		if err := c.syncContainer(); err != nil {
			return nil, false, err
		}
	}

	return c.networks()
}

// NetworkMode gets the configured network mode for the container.
// Get actual value from the database
func (c *Container) NetworkMode() string {
	networkMode := ""
	ctrSpec := c.config.Spec

	switch {
	case c.config.CreateNetNS:
		// We actually store the network
		// mode for Slirp and Bridge, so
		// we can just use that
		networkMode = string(c.config.NetMode)
	case c.config.NetNsCtr != "":
		networkMode = fmt.Sprintf("container:%s", c.config.NetNsCtr)
	default:
		// Find the spec's network namespace.
		// If there is none, it's host networking.
		// If there is one and it has a path, it's "ns:".
		foundNetNS := false
		for _, ns := range ctrSpec.Linux.Namespaces {
			if ns.Type == spec.NetworkNamespace {
				foundNetNS = true
				if ns.Path != "" {
					networkMode = fmt.Sprintf("ns:%s", ns.Path)
				} else {
					// We're making a network ns,  but not
					// configuring with Slirp or CNI. That
					// means it's --net=none
					networkMode = "none"
				}
				break
			}
		}
		if !foundNetNS {
			networkMode = "host"
		}
	}
	return networkMode
}

// Unlocked accessor for networks
func (c *Container) networks() ([]string, bool, error) {
	networks, err := c.runtime.state.GetNetworks(c)
	if err != nil && errors.Cause(err) == define.ErrNoSuchNetwork {
		if len(c.config.Networks) == 0 && c.config.NetMode.IsBridge() {
			return []string{c.runtime.netPlugin.GetDefaultNetworkName()}, true, nil
		}
		return c.config.Networks, false, nil
	}

	return networks, false, err
}

// networksByNameIndex provides us with a map of container networks where key
// is network name and value is the index position
func (c *Container) networksByNameIndex() (map[string]int, error) {
	networks, _, err := c.networks()
	if err != nil {
		return nil, err
	}
	networkNamesByIndex := make(map[string]int, len(networks))
	for index, name := range networks {
		networkNamesByIndex[name] = index
	}
	return networkNamesByIndex, nil
}

// add puts the new given CNI network name into the tracking map
// and assigns it a new integer based on the map length
func (d ContainerNetworkDescriptions) add(networkName string) {
	d[networkName] = len(d)
}

// getInterfaceByName returns a formatted interface name for a given
// network along with a bool as to whether the network existed
func (d ContainerNetworkDescriptions) getInterfaceByName(networkName string) (string, bool) {
	val, exists := d[networkName]
	if !exists {
		return "", exists
	}
	return fmt.Sprintf("eth%d", val), exists
}