X-Git-Url: http://git.ipfire.org/?p=thirdparty%2Fmdadm.git;a=blobdiff_plain;f=mdadm.h;h=a44bb2a98a21cef296d49e074f006c13d4a9d456;hp=c7f864bba4485a80239a85b85dd9e04834a982ca;hb=e5408a3202c09267f780d53334c369793534a7e8;hpb=9277cc7752ae6eb4b10dff7b5fc9ec75d3d52b66 diff --git a/mdadm.h b/mdadm.h index c7f864bb..a44bb2a9 100644 --- a/mdadm.h +++ b/mdadm.h @@ -68,6 +68,39 @@ extern __off64_t lseek64 __P ((int __fd, __off64_t __offset, int __whence)); #define DEFAULT_BITMAP_DELAY 5 #define DEFAULT_MAX_WRITE_BEHIND 256 +/* MAP_DIR should be somewhere that persists across the pivotroot + * from early boot to late boot. + * Currently /dev seems to be the only option on most distros. + */ +#ifndef MAP_DIR +#define MAP_DIR "/dev/.mdadm" +#endif /* MAP_DIR */ +/* MAP_FILE is what we name the map file we put in MAP_DIR, in case you + * want something other than the default of "map" + */ +#ifndef MAP_FILE +#define MAP_FILE "map" +#endif /* MAP_FILE */ +/* MDMON_DIR is where pid and socket files used for communicating + * with mdmon normally live. It *should* be /var/run, but when + * mdmon is needed at early boot then it needs to write there prior + * to /var/run being mounted read/write, and it also then needs to + * persist beyond when /var/run is mounter read-only. So, to be + * safe, the default is somewhere that is read/write early in the + * boot process and stays up as long as possible during shutdown. + */ +#ifndef MDMON_DIR +#define MDMON_DIR "/dev/.mdadm/" +#endif /* MDMON_DIR */ + +/* FAILED_SLOTS is where to save files storing recent removal of array + * member in order to allow future reuse of disk inserted in the same + * slot for array recovery + */ +#ifndef FAILED_SLOTS_DIR +#define FAILED_SLOTS_DIR "/dev/.mdadm/failed-slots" +#endif /* FAILED_SLOTS */ + #include "md_u.h" #include "md_p.h" #include "bitmap.h" @@ -129,6 +162,22 @@ extern __off64_t lseek64 __P ((int __fd, __off64_t __offset, int __whence)); #endif /* __KLIBC__ */ +/* + * min()/max()/clamp() macros that also do + * strict type-checking.. See the + * "unnecessary" pointer comparison. + */ +#define min(x, y) ({ \ + typeof(x) _min1 = (x); \ + typeof(y) _min2 = (y); \ + (void) (&_min1 == &_min2); \ + _min1 < _min2 ? _min1 : _min2; }) + +#define max(x, y) ({ \ + typeof(x) _max1 = (x); \ + typeof(y) _max2 = (y); \ + (void) (&_max1 == &_max2); \ + _max1 > _max2 ? _max1 : _max2; }) /* general information that might be extracted from a superblock */ struct mdinfo { @@ -146,11 +195,15 @@ struct mdinfo { */ int reshape_active; unsigned long long reshape_progress; - unsigned long long resync_start; + union { + unsigned long long resync_start; /* per-array resync position */ + unsigned long long recovery_start; /* per-device rebuild position */ + #define MaxSector (~0ULL) /* resync/recovery complete position */ + }; unsigned long safe_mode_delay; /* ms delay to mark clean */ int new_level, delta_disks, new_layout, new_chunk; int errors; - int cache_size; /* size of raid456 stripe cache*/ + unsigned long cache_size; /* size of raid456 stripe cache*/ int mismatch_cnt; char text_version[50]; void *update_private; /* for passing metadata-format @@ -162,12 +215,15 @@ struct mdinfo { int container_member; /* for assembling external-metatdata arrays * This is to be used internally by metadata * handler only */ - + int container_enough; /* flag external handlers can set to + * indicate that subarrays have not enough (-1), + * enough to start (0), or all expected disks (1) */ char sys_name[20]; struct mdinfo *devs; struct mdinfo *next; /* Device info for mdmon: */ + int recovery_fd; int state_fd; #define DS_FAULTY 1 #define DS_INSYNC 2 @@ -214,6 +270,7 @@ extern char Version[], Usage[], Help[], OptionHelp[], /* for option that don't have short equivilents, we assign arbitrary * small numbers. '1' means an undecorated option, so we start at '2'. + * (note we must stop before we get to 65 i.e. 'A') */ enum special_options { AssumeClean = 2, @@ -222,13 +279,17 @@ enum special_options { ReAdd, NoDegraded, Sparc22, - BackupFile, + BackupFile, /* 8 */ HomeHost, AutoHomeHost, Symlinks, AutoDetect, Waitclean, DetailPlatform, + KillSubarray, + UpdateSubarray, /* 16 */ + IncrementalPath, + NoSharing }; /* structures read from config file */ @@ -242,21 +303,21 @@ enum special_options { * devices is considered */ #define UnSet (0xfffe) -typedef struct mddev_ident_s { +struct mddev_ident { char *devname; int uuid_set; int uuid[4]; char name[33]; - unsigned int super_minor; + int super_minor; char *devices; /* comma separated list of device * names with wild cards */ int level; - unsigned int raid_disks; - unsigned int spare_disks; + int raid_disks; + int spare_disks; struct supertype *st; int autof; /* 1 for normal, 2 for partitioned */ char *spare_group; @@ -270,11 +331,15 @@ typedef struct mddev_ident_s { */ char *member; /* subarray within a container */ - struct mddev_ident_s *next; -} *mddev_ident_t; + struct mddev_ident *next; + union { + /* fields needed by different users of this structure */ + int assembled; /* set when assembly succeeds */ + }; +}; /* List of device names - wildcards expanded */ -typedef struct mddev_dev_s { +struct mddev_dev { char *devname; char disposition; /* 'a' for add, 'r' for remove, 'f' for fail. * Not set for names read from .config @@ -282,10 +347,8 @@ typedef struct mddev_dev_s { char writemostly; /* 1 for 'set writemostly', 2 for 'clear writemostly' */ char re_add; char used; /* set when used */ - struct mdinfo *content; /* If devname is a container, this might list - * the remaining member arrays. */ - struct mddev_dev_s *next; -} *mddev_dev_t; + struct mddev_dev *next; +}; typedef struct mapping { char *name; @@ -300,11 +363,15 @@ struct mdstat_ent { char *level; char *pattern; /* U or up, _ for down */ int percent; /* -1 if no resync */ - int resync; /* 1 if resync, 0 if recovery */ + int resync; /* 3 if check, 2 if reshape, 1 if resync, 0 if recovery */ int devcnt; int raid_disks; int chunk_size; char * metadata_version; + struct dev_member { + char *name; + struct dev_member *next; + } *members; struct mdstat_ent *next; }; @@ -313,6 +380,7 @@ extern void free_mdstat(struct mdstat_ent *ms); extern void mdstat_wait(int seconds); extern void mdstat_wait_fd(int fd, const sigset_t *sigmask); extern int mddev_busy(int devnum); +extern struct mdstat_ent *mdstat_by_component(char *name); struct map_ent { struct map_ent *next; @@ -324,6 +392,7 @@ struct map_ent { }; extern int map_update(struct map_ent **mpp, int devnum, char *metadata, int uuid[4], char *path); +extern void map_remove(struct map_ent **map, int devnum); extern struct map_ent *map_by_uuid(struct map_ent **map, int uuid[4]); extern struct map_ent *map_by_devnum(struct map_ent **map, int devnum); extern struct map_ent *map_by_name(struct map_ent **map, char *name); @@ -353,7 +422,6 @@ enum sysfs_read_flags { GET_SIZE = (1 << 12), GET_STATE = (1 << 13), GET_ERROR = (1 << 14), - SKIP_GONE_DEVS = (1 << 15), }; /* If fd >= 0, get the array it is open on, @@ -380,8 +448,7 @@ extern int sysfs_get_str(struct mdinfo *sra, struct mdinfo *dev, char *name, char *val, int size); extern int sysfs_set_safemode(struct mdinfo *sra, unsigned long ms); extern int sysfs_set_array(struct mdinfo *info, int vers); -extern int sysfs_add_disk(struct mdinfo *sra, struct mdinfo *sd, - int in_sync); +extern int sysfs_add_disk(struct mdinfo *sra, struct mdinfo *sd, int resume); extern int sysfs_disk_to_scsi_id(int fd, __u32 *id); extern int sysfs_unique_holder(int devnum, long rdev); extern int load_sys(char *path, char *buf); @@ -475,9 +542,13 @@ extern struct superswitch { * The particular device should be: * The last device added by add_to_super * The device the metadata was loaded from by load_super + * If 'map' is present, then it is an array raid_disks long + * (raid_disk must already be set and correct) and it is filled + * with 1 for slots that are thought to be active and 0 for slots which + * appear to be failed/missing. */ - void (*getinfo_super)(struct supertype *st, struct mdinfo *info); - + void (*getinfo_super)(struct supertype *st, struct mdinfo *info, char *map); + struct mdinfo *(*getinfo_super_disks)(struct supertype *st); /* Check if the given metadata is flagged as belonging to "this" * host. 0 for 'no', 1 for 'yes', -1 for "Don't record homehost" */ @@ -534,8 +605,10 @@ extern struct superswitch { int (*write_init_super)(struct supertype *st); int (*compare_super)(struct supertype *st, struct supertype *tst); int (*load_super)(struct supertype *st, int fd, char *devname); + int (*load_container)(struct supertype *st, int fd, char *devname); struct supertype * (*match_metadata_desc)(char *arg); __u64 (*avail_size)(struct supertype *st, __u64 size); + unsigned long long (*min_acceptable_spare_size)(struct supertype *st); int (*add_internal_bitmap)(struct supertype *st, int *chunkp, int delay, int write_behind, unsigned long long size, int may_change, int major); @@ -558,9 +631,16 @@ extern struct superswitch { char *subdev, unsigned long long *freesize, int verbose); - struct mdinfo *(*container_content)(struct supertype *st); + struct mdinfo *(*container_content)(struct supertype *st, char *subarray); /* Allow a metadata handler to override mdadm's default layouts */ int (*default_layout)(int level); /* optional */ + /* query the supertype for default chunk size */ + int (*default_chunk)(struct supertype *st); /* optional */ + /* Permit subarray's to be deleted from inactive containers */ + int (*kill_subarray)(struct supertype *st); /* optional */ + /* Permit subarray's to be modified */ + int (*update_subarray)(struct supertype *st, char *subarray, + char *update, struct mddev_ident *ident); /* optional */ /* for mdmon */ int (*open_new)(struct supertype *c, struct active_array *a, @@ -602,13 +682,30 @@ extern struct superswitch { */ struct mdinfo *(*activate_spare)(struct active_array *a, struct metadata_update **updates); + /* + * Return statically allocated string that represents metadata specific + * controller domain of the disk. The domain is used in disk domain + * matching functions. Disks belong to the same domain if the they have + * the same domain from mdadm.conf and belong the same metadata domain. + * Returning NULL or not providing this handler means that metadata + * does not distinguish the differences between disks that belong to + * different controllers. They are in the domain specified by + * configuration file (mdadm.conf). + * In case when the metadata has the notion of domains based on disk + * it shall return NULL for disks that do not belong to the controller + * the supported domains. Such disks will form another domain and won't + * be mixed with supported ones. + */ + const char *(*get_disk_controller_domain)(const char *path); int swapuuid; /* true if uuid is bigending rather than hostendian */ int external; const char *name; /* canonical metadata name */ -} super0, super1, super_ddf, *superlist[]; +} *superlist[]; -extern struct superswitch super_imsm; +extern struct superswitch super0, super1; +extern struct superswitch super_imsm, super_ddf; +extern struct superswitch mbr, gpt; struct metadata_update { int len; @@ -639,11 +736,8 @@ struct supertype { int minor_version; int max_devs; int container_dev; /* devnum of container */ - char subarray[32]; /* name of array inside container */ void *sb; void *info; - int loaded_container; /* Set if load_super found a container, - * not just one device */ struct metadata_update *updates; struct metadata_update **update_tail; @@ -661,14 +755,117 @@ struct supertype { }; -extern struct supertype *super_by_fd(int fd); -extern struct supertype *guess_super(int fd); +extern struct supertype *super_by_fd(int fd, char **subarray); +enum guess_types { guess_any, guess_array, guess_partitions }; +extern struct supertype *guess_super_type(int fd, enum guess_types guess_type); +static inline struct supertype *guess_super(int fd) { + return guess_super_type(fd, guess_any); +} extern struct supertype *dup_super(struct supertype *st); extern int get_dev_size(int fd, char *dname, unsigned long long *sizep); +extern int must_be_container(int fd); +extern int dev_size_from_id(unsigned id, unsigned long long *size); extern void get_one_disk(int mdfd, mdu_array_info_t *ainf, mdu_disk_info_t *disk); void wait_for(char *dev, int fd); +/* + * Data structures for policy management. + * Each device can have a policy structure that lists + * various name/value pairs each possibly with a metadata associated. + * The policy list is sorted by name/value/metadata + */ +struct dev_policy { + struct dev_policy *next; + char *name; /* None of these strings are allocated. They are + * all just references to strings which are known + * to exist elsewhere. + * name and metadata can be compared by address equality. + */ + const char *metadata; + const char *value; +}; + +extern char pol_act[], pol_domain[], pol_metadata[], pol_auto[]; + +/* iterate over the sublist starting at list, having the same + * 'name' as 'list', and matching the given metadata (Where + * NULL matches anything + */ +#define pol_for_each(item, list, _metadata) \ + for (item = list; \ + item && item->name == list->name; \ + item = item->next) \ + if (!(!_metadata || !item->metadata || _metadata == item->metadata)) \ + ; else + +/* + * policy records read from mdadm are largely just name-value pairs. + * The names are constants, not strdupped + */ +struct pol_rule { + struct pol_rule *next; + char *type; /* rule_policy or rule_part */ + struct rule { + struct rule *next; + char *name; + char *value; + char *dups; /* duplicates of 'value' with a partNN appended */ + } *rule; +}; + +extern char rule_policy[], rule_part[]; +extern char rule_path[], rule_type[]; +extern char type_part[], type_disk[]; + +extern void policyline(char *line, char *type); +extern void policy_add(char *type, ...); +extern void policy_free(void); + +extern struct dev_policy *path_policy(char *path, char *type); +extern struct dev_policy *disk_policy(struct mdinfo *disk); +extern struct dev_policy *devnum_policy(int dev); +extern void dev_policy_free(struct dev_policy *p); + +//extern void pol_new(struct dev_policy **pol, char *name, char *val, char *metadata); +extern void pol_add(struct dev_policy **pol, char *name, char *val, char *metadata); +extern struct dev_policy *pol_find(struct dev_policy *pol, char *name); + +enum policy_action { + act_default, + act_include, + act_re_add, + act_spare, /* This only applies to bare devices */ + act_spare_same_slot, /* this allows non-bare devices, + * but only if recent removal */ + act_force_spare, /* this allow non-bare devices in any case */ + act_err +}; + +extern int policy_action_allows(struct dev_policy *plist, const char *metadata, + enum policy_action want); +extern int disk_action_allows(struct mdinfo *disk, const char *metadata, + enum policy_action want); + +struct domainlist { + struct domainlist *next; + const char *dom; +}; + +extern int domain_test(struct domainlist *dom, struct dev_policy *pol, + const char *metadata); +extern struct domainlist *domain_from_array(struct mdinfo *mdi, + const char *metadata); +extern void domainlist_add_dev(struct domainlist **dom, int devnum, + const char *metadata); +extern void domain_free(struct domainlist *dl); +extern void domain_merge(struct domainlist **domp, struct dev_policy *pol, + const char *metadata); +void domain_add(struct domainlist **domp, char *domain); + +extern void policy_save_path(char *id_path, struct map_ent *array); +extern int policy_check_path(struct mdinfo *disk, struct map_ent *array); + #if __GNUC__ < 3 struct stat64; #endif @@ -714,7 +911,7 @@ extern int Manage_ro(char *devname, int fd, int readonly); extern int Manage_runstop(char *devname, int fd, int runstop, int quiet); extern int Manage_resize(char *devname, int fd, long long size, int raid_disks); extern int Manage_subdevs(char *devname, int fd, - mddev_dev_t devlist, int verbose); + struct mddev_dev *devlist, int verbose, int test); extern int autodetect(void); extern int Grow_Add_device(char *devname, int fd, char *newdev); extern int Grow_addbitmap(char *devname, int fd, char *file, int chunk, int delay, int write_behind, int force); @@ -727,14 +924,14 @@ extern int Grow_continue(int mdfd, struct supertype *st, struct mdinfo *info, char *backup_file); extern int Assemble(struct supertype *st, char *mddev, - mddev_ident_t ident, - mddev_dev_t devlist, char *backup_file, + struct mddev_ident *ident, + struct mddev_dev *devlist, char *backup_file, int readonly, int runstop, char *update, char *homehost, int require_homehost, int verbose, int force); extern int Build(char *mddev, int chunk, int level, int layout, - int raiddisks, mddev_dev_t devlist, int assume_clean, + int raiddisks, struct mddev_dev *devlist, int assume_clean, char *bitmap_file, int bitmap_chunk, int write_behind, int delay, int verbose, int autof, unsigned long long size); @@ -742,33 +939,33 @@ extern int Build(char *mddev, int chunk, int level, int layout, extern int Create(struct supertype *st, char *mddev, int chunk, int level, int layout, unsigned long long size, int raiddisks, int sparedisks, char *name, char *homehost, int *uuid, - int subdevs, mddev_dev_t devlist, + int subdevs, struct mddev_dev *devlist, int runstop, int verbose, int force, int assume_clean, char *bitmap_file, int bitmap_chunk, int write_behind, int delay, int autof); extern int Detail(char *dev, int brief, int export, int test, char *homehost); extern int Detail_Platform(struct superswitch *ss, int scan, int verbose); extern int Query(char *dev); -extern int Examine(mddev_dev_t devlist, int brief, int export, int scan, +extern int Examine(struct mddev_dev *devlist, int brief, int export, int scan, int SparcAdjust, struct supertype *forcest, char *homehost); -extern int Monitor(mddev_dev_t devlist, +extern int Monitor(struct mddev_dev *devlist, char *mailaddr, char *alert_cmd, int period, int daemonise, int scan, int oneshot, - int dosyslog, int test, char *pidfile, int increments); + int dosyslog, int test, char *pidfile, int increments, + int share); extern int Kill(char *dev, struct supertype *st, int force, int quiet, int noexcl); +extern int Kill_subarray(char *dev, char *subarray, int quiet); +extern int Update_subarray(char *dev, char *subarray, char *update, struct mddev_ident *ident, int quiet); extern int Wait(char *dev); extern int WaitClean(char *dev, int sock, int verbose); extern int Incremental(char *devname, int verbose, int runstop, struct supertype *st, char *homehost, int require_homehost, int autof); -extern int Incremental_container(struct supertype *st, char *devname, - int verbose, int runstop, int autof, - int trustworthy); extern void RebuildMap(void); extern int IncrementalScan(int verbose); - +extern int IncrementalRemove(char *devname, char *path, int verbose); extern int CreateBitmap(char *filename, int force, char uuid[16], unsigned long chunksize, unsigned long daemon_sleep, unsigned long write_behind, @@ -787,6 +984,7 @@ extern int parse_layout_faulty(char *layout); extern int check_ext2(int fd, char *name); extern int check_reiser(int fd, char *name); extern int check_raid(int fd, char *name); +extern int check_partitions(int fd, char *dname, unsigned long long freesize); extern int get_mdp_major(void); extern int dev_open(char *dev, int flags); @@ -796,10 +994,10 @@ extern int is_standard(char *dev, int *nump); extern int same_dev(char *one, char *two); extern int parse_auto(char *str, char *msg, int config); -extern mddev_ident_t conf_get_ident(char *dev); -extern mddev_dev_t conf_get_devs(void); +extern struct mddev_ident *conf_get_ident(char *dev); +extern struct mddev_dev *conf_get_devs(void); extern int conf_test_dev(char *devname); -extern int conf_test_metadata(const char *version); +extern int conf_test_metadata(const char *version, struct dev_policy *pol, int is_homehost); extern struct createinfo *conf_get_create_info(void); extern void set_conffile(char *file); extern char *conf_get_mailaddr(void); @@ -810,7 +1008,7 @@ extern char *conf_line(FILE *file); extern char *conf_word(FILE *file, int allow_key); extern int conf_name_is_free(char *name); extern int devname_matches(char *name, char *match); -extern struct mddev_ident_s *conf_match(struct mdinfo *info, struct supertype *st); +extern struct mddev_ident *conf_match(struct mdinfo *info, struct supertype *st); extern void free_line(char *line); extern int match_oneof(char *devices, char *devname); @@ -824,9 +1022,11 @@ extern char *fname_from_uuid(struct supertype *st, extern unsigned long calc_csum(void *super, int bytes); extern int enough(int level, int raid_disks, int layout, int clean, char *avail, int avail_disks); +extern int enough_fd(int fd); extern int ask(char *mesg); extern unsigned long long get_component_size(int fd); extern void remove_partitions(int fd); +extern int test_partition(int fd); extern unsigned long long calc_array_size(int level, int raid_disks, int layout, int chunksize, unsigned long long devsize); extern int flush_metadata_updates(struct supertype *st); @@ -838,6 +1038,7 @@ extern int assemble_container_content(struct supertype *st, int mdfd, extern int add_disk(int mdfd, struct supertype *st, struct mdinfo *sra, struct mdinfo *info); extern int set_array_info(int mdfd, struct supertype *st, struct mdinfo *info); +unsigned long long min_recovery_start(struct mdinfo *array); extern char *human_size(long long bytes); extern char *human_size_brief(long long bytes); @@ -855,13 +1056,19 @@ extern int create_mddev(char *dev, char *name, int autof, int trustworthy, char *chosen); /* values for 'trustworthy' */ #define LOCAL 1 +#define LOCAL_ANY 10 #define FOREIGN 2 #define METADATA 3 extern int open_mddev(char *dev, int report_errors); extern int open_container(int fd); +extern int is_container_member(struct mdstat_ent *ent, char *devname); +extern int is_subarray_active(char *subarray, char *devname); +int is_container_active(char *devname); +extern int open_subarray(char *dev, char *subarray, struct supertype *st, int quiet); +extern struct superswitch *version_to_superswitch(char *vers); extern int mdmon_running(int devnum); -extern int signal_mdmon(int devnum); +extern int mdmon_pid(int devnum); extern int check_env(char *name); extern __u32 random32(void); extern int start_mdmon(int devnum); @@ -903,6 +1110,11 @@ static inline int is_subarray(char *vers) return (*vers == '/' || *vers == '-'); } +static inline char *to_subarray(struct mdstat_ent *ent, char *container) +{ + return &ent->metadata_version[10+strlen(container)+1]; +} + #ifdef DEBUG #define dprintf(fmt, arg...) \ fprintf(stderr, fmt, ##arg) @@ -1001,3 +1213,10 @@ static inline int xasprintf(char **strp, const char *fmt, ...) { #define ALGORITHM_PARITY_0_6 20 #define ALGORITHM_PARITY_N_6 ALGORITHM_PARITY_N +/* Define PATH_MAX in case we don't use glibc or standard library does + * not have PATH_MAX defined. Assume max path length is 4K characters. + */ +#ifndef PATH_MAX +#define PATH_MAX 4096 +#endif +