X-Git-Url: https://git.camperquake.de/gitweb.cgi?a=blobdiff_plain;f=lib%2Flibzfs%2Flibzfs_pool.c;h=e2a67c6a7b06306199a78f52ff3e7b4217792d65;hb=c372b36e3eba18d1a8747b8375e3e7c049d40c89;hp=3cb7bbc911e51994744409855240ea084d6ed72e;hpb=b8864a233c569edcc57c686f3ea8cd1ae3b89153;p=zfs.git diff --git a/lib/libzfs/libzfs_pool.c b/lib/libzfs/libzfs_pool.c index 3cb7bbc..e2a67c6 100644 --- a/lib/libzfs/libzfs_pool.c +++ b/lib/libzfs/libzfs_pool.c @@ -21,6 +21,8 @@ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. + * Copyright 2011 Nexenta Systems, Inc. All rights reserved. + * Copyright (c) 2012 by Delphix. All rights reserved. */ #include @@ -32,6 +34,8 @@ #include #include #include +#include +#include #include #include #include @@ -44,10 +48,6 @@ static int read_efi_label(nvlist_t *config, diskaddr_t *sb); -#define DISK_ROOT "/dev/dsk" -#define RDISK_ROOT "/dev/rdsk" -#define BACKUP_SLICE "s2" - typedef struct prop_flags { int create:1; /* Validate property on creation */ int import:1; /* Validate property on import */ @@ -62,7 +62,7 @@ typedef struct prop_flags { static int zpool_get_all_props(zpool_handle_t *zhp) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; libzfs_handle_t *hdl = zhp->zpool_hdl; (void) strlcpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); @@ -178,6 +178,8 @@ char * zpool_state_to_name(vdev_state_t state, vdev_aux_t aux) { switch (state) { + default: + break; case VDEV_STATE_CLOSED: case VDEV_STATE_OFFLINE: return (gettext("OFFLINE")); @@ -233,6 +235,7 @@ zpool_get_prop(zpool_handle_t *zhp, zpool_prop_t prop, char *buf, size_t len, case ZPOOL_PROP_ALTROOT: case ZPOOL_PROP_CACHEFILE: + case ZPOOL_PROP_COMMENT: if (zhp->zpool_props != NULL || zpool_get_all_props(zhp) == 0) { (void) strlcpy(buf, @@ -270,6 +273,8 @@ zpool_get_prop(zpool_handle_t *zhp, zpool_prop_t prop, char *buf, size_t len, case ZPOOL_PROP_SIZE: case ZPOOL_PROP_ALLOCATED: case ZPOOL_PROP_FREE: + case ZPOOL_PROP_EXPANDSZ: + case ZPOOL_PROP_ASHIFT: (void) zfs_nicenum(intval, buf, len); break; @@ -336,6 +341,7 @@ bootfs_name_valid(const char *pool, char *bootfs) return (B_FALSE); } +#if defined(__sun__) || defined(__sun) /* * Inspect the configuration to determine if any of the devices contain * an EFI label. @@ -356,9 +362,10 @@ pool_uses_efi(nvlist_t *config) } return (B_FALSE); } +#endif -static boolean_t -pool_is_bootable(zpool_handle_t *zhp) +boolean_t +zpool_is_bootable(zpool_handle_t *zhp) { char bootfs[ZPOOL_MAXNAMELEN]; @@ -382,7 +389,7 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, zpool_prop_t prop; char *strval; uint64_t intval; - char *slash; + char *slash, *check; struct stat64 statbuf; zpool_handle_t *zhp; nvlist_t *nvroot; @@ -421,6 +428,8 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, * Perform additional checking for specific properties. */ switch (prop) { + default: + break; case ZPOOL_PROP_VERSION: if (intval < version || intval > SPA_VERSION) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, @@ -431,6 +440,24 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, } break; + case ZPOOL_PROP_ASHIFT: + if (!flags.create) { + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, + "property '%s' can only be set at " + "creation time"), propname); + (void) zfs_error(hdl, EZFS_BADPROP, errbuf); + goto error; + } + + if (intval != 0 && (intval < 9 || intval > 13)) { + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, + "property '%s' number %d is invalid."), + propname, intval); + (void) zfs_error(hdl, EZFS_BADPROP, errbuf); + goto error; + } + break; + case ZPOOL_PROP_BOOTFS: if (flags.create || flags.import) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, @@ -469,6 +496,7 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, verify(nvlist_lookup_nvlist(zpool_get_config(zhp, NULL), ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); +#if defined(__sun__) || defined(__sun) /* * bootfs property cannot be set on a disk which has * been EFI labeled. @@ -481,6 +509,7 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, zpool_close(zhp); goto error; } +#endif zpool_close(zhp); break; @@ -541,6 +570,26 @@ zpool_valid_proplist(libzfs_handle_t *hdl, const char *poolname, *slash = '/'; break; + case ZPOOL_PROP_COMMENT: + for (check = strval; *check != '\0'; check++) { + if (!isprint(*check)) { + zfs_error_aux(hdl, + dgettext(TEXT_DOMAIN, + "comment may only have printable " + "characters")); + (void) zfs_error(hdl, EZFS_BADPROP, + errbuf); + goto error; + } + } + if (strlen(strval) > ZPROP_MAX_COMMENT) { + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, + "comment must not exceed %d characters"), + ZPROP_MAX_COMMENT); + (void) zfs_error(hdl, EZFS_BADPROP, errbuf); + goto error; + } + break; case ZPOOL_PROP_READONLY: if (!flags.import) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, @@ -565,7 +614,7 @@ error: int zpool_set_prop(zpool_handle_t *zhp, const char *propname, const char *propval) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; int ret = -1; char errbuf[1024]; nvlist_t *nvl = NULL; @@ -647,9 +696,14 @@ zpool_expand_proplist(zpool_handle_t *zhp, zprop_list_t **plp) /* * Don't start the slice at the default block of 34; many storage - * devices will use a stripe width of 128k, so start there instead. + * devices will use a stripe width of 128k, other vendors prefer a 1m + * alignment. It is best to play it safe and ensure a 1m alignment + * given 512B blocks. When the block size is larger by a power of 2 + * we will still be 1m aligned. Some devices are sensitive to the + * partition ending alignment as well. */ -#define NEW_START_BLOCK 256 +#define NEW_START_BLOCK 2048 +#define PARTITION_END_ALIGNMENT 2048 /* * Validate the given pool name, optionally putting an extended error message in @@ -730,7 +784,10 @@ zpool_name_valid(libzfs_handle_t *hdl, boolean_t isopen, const char *pool) zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "multiple '@' delimiters in name")); break; - + case NAME_ERR_NO_AT: + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, + "permission set is missing '@'")); + break; } } return (B_FALSE); @@ -877,7 +934,7 @@ int zpool_create(libzfs_handle_t *hdl, const char *pool, nvlist_t *nvroot, nvlist_t *props, nvlist_t *fsprops) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; nvlist_t *zc_fsprops = NULL; nvlist_t *zc_props = NULL; char msg[1024]; @@ -941,10 +998,12 @@ zpool_create(libzfs_handle_t *hdl, const char *pool, nvlist_t *nvroot, * This can happen if the user has specified the same * device multiple times. We can't reliably detect this * until we try to add it and see we already have a - * label. + * label. This can also happen under if the device is + * part of an active md or lvm device. */ zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "one or more vdevs refer to the same device")); + "one or more vdevs refer to the same device, or one of\n" + "the devices is part of an active md or lvm device")); return (zfs_error(hdl, EZFS_BADDEV, msg)); case EOVERFLOW: @@ -1009,7 +1068,7 @@ create_failed: int zpool_destroy(zpool_handle_t *zhp) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; zfs_handle_t *zfp = NULL; libzfs_handle_t *hdl = zhp->zpool_hdl; char msg[1024]; @@ -1052,7 +1111,7 @@ zpool_destroy(zpool_handle_t *zhp) int zpool_add(zpool_handle_t *zhp, nvlist_t *nvroot) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; int ret; libzfs_handle_t *hdl = zhp->zpool_hdl; char msg[1024]; @@ -1071,7 +1130,8 @@ zpool_add(zpool_handle_t *zhp, nvlist_t *nvroot) return (zfs_error(hdl, EZFS_BADVERSION, msg)); } - if (pool_is_bootable(zhp) && nvlist_lookup_nvlist_array(nvroot, +#if defined(__sun__) || defined(__sun) + if (zpool_is_bootable(zhp) && nvlist_lookup_nvlist_array(nvroot, ZPOOL_CONFIG_SPARES, &spares, &nspares) == 0) { uint64_t s; @@ -1089,6 +1149,7 @@ zpool_add(zpool_handle_t *zhp, nvlist_t *nvroot) } } } +#endif if (zpool_get_prop_int(zhp, ZPOOL_PROP_VERSION, NULL) < SPA_VERSION_L2CACHE && @@ -1176,7 +1237,7 @@ zpool_add(zpool_handle_t *zhp, nvlist_t *nvroot) int zpool_export_common(zpool_handle_t *zhp, boolean_t force, boolean_t hardforce) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; (void) snprintf(msg, sizeof (msg), dgettext(TEXT_DOMAIN, @@ -1422,7 +1483,7 @@ int zpool_import_props(libzfs_handle_t *hdl, nvlist_t *config, const char *newname, nvlist_t *props, int flags) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; zpool_rewind_policy_t policy; nvlist_t *nv = NULL; nvlist_t *nvinfo = NULL; @@ -1552,6 +1613,12 @@ zpool_import_props(libzfs_handle_t *hdl, nvlist_t *config, const char *newname, (void) zpool_standard_error(hdl, error, desc); break; + case EBUSY: + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, + "one or more devices are already in use\n")); + (void) zfs_error(hdl, EZFS_BADDEV, desc); + break; + default: (void) zpool_standard_error(hdl, error, desc); zpool_explain_recover(hdl, @@ -1592,7 +1659,7 @@ zpool_import_props(libzfs_handle_t *hdl, nvlist_t *config, const char *newname, int zpool_scan(zpool_handle_t *zhp, pool_scan_func_t func) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; libzfs_handle_t *hdl = zhp->zpool_hdl; @@ -1635,38 +1702,6 @@ zpool_scan(zpool_handle_t *zhp, pool_scan_func_t func) } /* - * This provides a very minimal check whether a given string is likely a - * c#t#d# style string. Users of this are expected to do their own - * verification of the s# part. - */ -#define CTD_CHECK(str) (str && str[0] == 'c' && isdigit(str[1])) - -/* - * More elaborate version for ones which may start with "/dev/dsk/" - * and the like. - */ -static int -ctd_check_path(char *str) { - /* - * If it starts with a slash, check the last component. - */ - if (str && str[0] == '/') { - char *tmp = strrchr(str, '/'); - - /* - * If it ends in "/old", check the second-to-last - * component of the string instead. - */ - if (tmp != str && strcmp(tmp, "/old") == 0) { - for (tmp--; *tmp != '/'; tmp--) - ; - } - str = tmp + 1; - } - return (CTD_CHECK(str)); -} - -/* * Find a vdev that matches the search criteria specified. We use the * the nvpair name to determine how we should look for the device. * 'avail_spare' is set to TRUE if the provided guid refers to an AVAIL @@ -1714,50 +1749,22 @@ vdev_to_nvlist_iter(nvlist_t *nv, nvlist_t *search, boolean_t *avail_spare, * Search for the requested value. Special cases: * * - ZPOOL_CONFIG_PATH for whole disk entries. These end in - * "s0" or "s0/old". The "s0" part is hidden from the user, + * "-part1", or "p1". The suffix is hidden from the user, * but included in the string, so this matches around it. + * - ZPOOL_CONFIG_PATH for short names zfs_strcmp_shortname() + * is used to check all possible expanded paths. * - looking for a top-level vdev name (i.e. ZPOOL_CONFIG_TYPE). * * Otherwise, all other searches are simple string compares. */ - if (strcmp(srchkey, ZPOOL_CONFIG_PATH) == 0 && - ctd_check_path(val)) { + if (strcmp(srchkey, ZPOOL_CONFIG_PATH) == 0) { uint64_t wholedisk = 0; (void) nvlist_lookup_uint64(nv, ZPOOL_CONFIG_WHOLE_DISK, &wholedisk); - if (wholedisk) { - int slen = strlen(srchval); - int vlen = strlen(val); - - if (slen != vlen - 2) - break; - - /* - * make_leaf_vdev() should only set - * wholedisk for ZPOOL_CONFIG_PATHs which - * will include "/dev/dsk/", giving plenty of - * room for the indices used next. - */ - ASSERT(vlen >= 6); - - /* - * strings identical except trailing "s0" - */ - if (strcmp(&val[vlen - 2], "s0") == 0 && - strncmp(srchval, val, slen) == 0) - return (nv); - - /* - * strings identical except trailing "s0/old" - */ - if (strcmp(&val[vlen - 6], "s0/old") == 0 && - strcmp(&srchval[slen - 4], "/old") == 0 && - strncmp(srchval, val, slen - 4) == 0) - return (nv); + if (zfs_strcmp_pathname(srchval, val, wholedisk) == 0) + return (nv); - break; - } } else if (strcmp(srchkey, ZPOOL_CONFIG_TYPE) == 0 && val) { char *type, *idx, *end, *p; uint64_t id, vdev_id; @@ -1908,7 +1915,6 @@ nvlist_t * zpool_find_vdev(zpool_handle_t *zhp, const char *path, boolean_t *avail_spare, boolean_t *l2cache, boolean_t *log) { - char buf[MAXPATHLEN]; char *end; nvlist_t *nvroot, *search, *ret; uint64_t guid; @@ -1920,9 +1926,6 @@ zpool_find_vdev(zpool_handle_t *zhp, const char *path, boolean_t *avail_spare, verify(nvlist_add_uint64(search, ZPOOL_CONFIG_GUID, guid) == 0); } else if (zpool_vdev_is_interior(path)) { verify(nvlist_add_string(search, ZPOOL_CONFIG_TYPE, path) == 0); - } else if (path[0] != '/') { - (void) snprintf(buf, sizeof (buf), "%s%s", "/dev/dsk/", path); - verify(nvlist_add_string(search, ZPOOL_CONFIG_PATH, buf) == 0); } else { verify(nvlist_add_string(search, ZPOOL_CONFIG_PATH, path) == 0); } @@ -2060,6 +2063,7 @@ zpool_get_config_physpath(nvlist_t *config, char *physpath, size_t phypath_size) &child, &count) != 0) return (EZFS_INVALCONFIG); +#if defined(__sun__) || defined(__sun) /* * root pool can not have EFI labeled disks and can only have * a single top-level vdev. @@ -2067,6 +2071,7 @@ zpool_get_config_physpath(nvlist_t *config, char *physpath, size_t phypath_size) if (strcmp(type, VDEV_TYPE_ROOT) != 0 || count != 1 || pool_uses_efi(vdev_root)) return (EZFS_POOL_INVALARG); +#endif (void) vdev_get_physpaths(child[0], physpath, phypath_size, &rsz, B_FALSE); @@ -2094,36 +2099,30 @@ zpool_get_physpath(zpool_handle_t *zhp, char *physpath, size_t phypath_size) * the disk to use the new unallocated space. */ static int -zpool_relabel_disk(libzfs_handle_t *hdl, const char *name) +zpool_relabel_disk(libzfs_handle_t *hdl, const char *path, const char *msg) { - char path[MAXPATHLEN]; - char errbuf[1024]; int fd, error; - int (*_efi_use_whole_disk)(int); - if ((_efi_use_whole_disk = (int (*)(int))dlsym(RTLD_DEFAULT, - "efi_use_whole_disk")) == NULL) - return (-1); - - (void) snprintf(path, sizeof (path), "%s/%s", RDISK_ROOT, name); - - if ((fd = open(path, O_RDWR | O_NDELAY)) < 0) { + if ((fd = open(path, O_RDWR|O_DIRECT)) < 0) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot " - "relabel '%s': unable to open device"), name); - return (zfs_error(hdl, EZFS_OPENFAILED, errbuf)); + "relabel '%s': unable to open device: %d"), path, errno); + return (zfs_error(hdl, EZFS_OPENFAILED, msg)); } /* * It's possible that we might encounter an error if the device * does not have any unallocated space left. If so, we simply * ignore that error and continue on. + * + * Also, we don't call efi_rescan() - that would just return EBUSY. + * The module will do it for us in vdev_disk_open(). */ - error = _efi_use_whole_disk(fd); + error = efi_use_whole_disk(fd); (void) close(fd); if (error && error != VT_ENOSPC) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot " - "relabel '%s': unable to read disk capacity"), name); - return (zfs_error(hdl, EZFS_NOCAP, errbuf)); + "relabel '%s': unable to read disk capacity"), path); + return (zfs_error(hdl, EZFS_NOCAP, msg)); } return (0); } @@ -2136,11 +2135,12 @@ int zpool_vdev_online(zpool_handle_t *zhp, const char *path, int flags, vdev_state_t *newstate) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tgt; boolean_t avail_spare, l2cache, islog; libzfs_handle_t *hdl = zhp->zpool_hdl; + int error; if (flags & ZFS_ONLINE_EXPAND) { (void) snprintf(msg, sizeof (msg), @@ -2162,13 +2162,10 @@ zpool_vdev_online(zpool_handle_t *zhp, const char *path, int flags, if (flags & ZFS_ONLINE_EXPAND || zpool_get_prop_int(zhp, ZPOOL_PROP_AUTOEXPAND, NULL)) { - char *pathname = NULL; uint64_t wholedisk = 0; (void) nvlist_lookup_uint64(tgt, ZPOOL_CONFIG_WHOLE_DISK, &wholedisk); - verify(nvlist_lookup_string(tgt, ZPOOL_CONFIG_PATH, - &pathname) == 0); /* * XXX - L2ARC 1.0 devices can't support expansion. @@ -2180,8 +2177,22 @@ zpool_vdev_online(zpool_handle_t *zhp, const char *path, int flags, } if (wholedisk) { - pathname += strlen(DISK_ROOT) + 1; - (void) zpool_relabel_disk(hdl, pathname); + const char *fullpath = path; + char buf[MAXPATHLEN]; + + if (path[0] != '/') { + error = zfs_resolve_shortname(path, buf, + sizeof(buf)); + if (error != 0) + return (zfs_error(hdl, EZFS_NODEVICE, + msg)); + + fullpath = buf; + } + + error = zpool_relabel_disk(hdl, fullpath, msg); + if (error != 0) + return (error); } } @@ -2208,7 +2219,7 @@ zpool_vdev_online(zpool_handle_t *zhp, const char *path, int flags, int zpool_vdev_offline(zpool_handle_t *zhp, const char *path, boolean_t istmp) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tgt; boolean_t avail_spare, l2cache; @@ -2258,7 +2269,7 @@ zpool_vdev_offline(zpool_handle_t *zhp, const char *path, boolean_t istmp) int zpool_vdev_fault(zpool_handle_t *zhp, uint64_t guid, vdev_aux_t aux) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; libzfs_handle_t *hdl = zhp->zpool_hdl; @@ -2293,7 +2304,7 @@ zpool_vdev_fault(zpool_handle_t *zhp, uint64_t guid, vdev_aux_t aux) int zpool_vdev_degrade(zpool_handle_t *zhp, uint64_t guid, vdev_aux_t aux) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; libzfs_handle_t *hdl = zhp->zpool_hdl; @@ -2347,7 +2358,7 @@ int zpool_vdev_attach(zpool_handle_t *zhp, const char *old_disk, const char *new_disk, nvlist_t *nvroot, int replacing) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; int ret; nvlist_t *tgt; @@ -2358,7 +2369,7 @@ zpool_vdev_attach(zpool_handle_t *zhp, uint_t children; nvlist_t *config_root; libzfs_handle_t *hdl = zhp->zpool_hdl; - boolean_t rootpool = pool_is_bootable(zhp); + boolean_t rootpool = zpool_is_bootable(zhp); if (replacing) (void) snprintf(msg, sizeof (msg), dgettext(TEXT_DOMAIN, @@ -2367,6 +2378,7 @@ zpool_vdev_attach(zpool_handle_t *zhp, (void) snprintf(msg, sizeof (msg), dgettext(TEXT_DOMAIN, "cannot attach %s to %s"), new_disk, old_disk); +#if defined(__sun__) || defined(__sun) /* * If this is a root pool, make sure that we're not attaching an * EFI labeled device. @@ -2376,6 +2388,7 @@ zpool_vdev_attach(zpool_handle_t *zhp, "EFI labeled devices are not supported on root pools.")); return (zfs_error(hdl, EZFS_POOL_NOTSUP, msg)); } +#endif (void) strlcpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); if ((tgt = zpool_find_vdev(zhp, old_disk, &avail_spare, &l2cache, @@ -2521,7 +2534,7 @@ zpool_vdev_attach(zpool_handle_t *zhp, int zpool_vdev_detach(zpool_handle_t *zhp, const char *path) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tgt; boolean_t avail_spare, l2cache; @@ -2619,7 +2632,7 @@ int zpool_vdev_split(zpool_handle_t *zhp, char *newname, nvlist_t **newroot, nvlist_t *props, splitflags_t flags) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tree, *config, **child, **newchild, *newconfig = NULL; nvlist_t **varray = NULL, *zc_props = NULL; @@ -2830,7 +2843,7 @@ out: int zpool_vdev_remove(zpool_handle_t *zhp, const char *path) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tgt; boolean_t avail_spare, l2cache, islog; @@ -2875,7 +2888,7 @@ zpool_vdev_remove(zpool_handle_t *zhp, const char *path) int zpool_clear(zpool_handle_t *zhp, const char *path, nvlist_t *rewindnvl) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; nvlist_t *tgt; zpool_rewind_policy_t policy; @@ -2951,7 +2964,7 @@ zpool_clear(zpool_handle_t *zhp, const char *path, nvlist_t *rewindnvl) int zpool_vdev_clear(zpool_handle_t *zhp, uint64_t guid) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; char msg[1024]; libzfs_handle_t *hdl = zhp->zpool_hdl; @@ -2970,6 +2983,46 @@ zpool_vdev_clear(zpool_handle_t *zhp, uint64_t guid) } /* + * Change the GUID for a pool. + */ +int +zpool_reguid(zpool_handle_t *zhp) +{ + char msg[1024]; + libzfs_handle_t *hdl = zhp->zpool_hdl; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; + + (void) snprintf(msg, sizeof (msg), + dgettext(TEXT_DOMAIN, "cannot reguid '%s'"), zhp->zpool_name); + + (void) strlcpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); + if (zfs_ioctl(hdl, ZFS_IOC_POOL_REGUID, &zc) == 0) + return (0); + + return (zpool_standard_error(hdl, errno, msg)); +} + +/* + * Reopen the pool. + */ +int +zpool_reopen(zpool_handle_t *zhp) +{ + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; + char msg[1024]; + libzfs_handle_t *hdl = zhp->zpool_hdl; + + (void) snprintf(msg, sizeof (msg), + dgettext(TEXT_DOMAIN, "cannot reopen '%s'"), + zhp->zpool_name); + + (void) strlcpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); + if (zfs_ioctl(hdl, ZFS_IOC_POOL_REOPEN, &zc) == 0) + return (0); + return (zpool_standard_error(hdl, errno, msg)); +} + +/* * Convert from a devid string to a path. */ static char * @@ -3035,7 +3088,7 @@ path_to_devid(const char *path) static void set_path(zpool_handle_t *zhp, nvlist_t *nv, const char *path) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; (void) strncpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); (void) strncpy(zc.zc_value, path, sizeof (zc.zc_value)); @@ -3046,6 +3099,37 @@ set_path(zpool_handle_t *zhp, nvlist_t *nv, const char *path) } /* + * Remove partition suffix from a vdev path. Partition suffixes may take three + * forms: "-partX", "pX", or "X", where X is a string of digits. The second + * case only occurs when the suffix is preceded by a digit, i.e. "md0p0" The + * third case only occurs when preceded by a string matching the regular + * expression "^[hs]d[a-z]+", i.e. a scsi or ide disk. + */ +static char * +strip_partition(libzfs_handle_t *hdl, char *path) +{ + char *tmp = zfs_strdup(hdl, path); + char *part = NULL, *d = NULL; + + if ((part = strstr(tmp, "-part")) && part != tmp) { + d = part + 5; + } else if ((part = strrchr(tmp, 'p')) && + part > tmp + 1 && isdigit(*(part-1))) { + d = part + 1; + } else if ((tmp[0] == 'h' || tmp[0] == 's') && tmp[1] == 'd') { + for (d = &tmp[2]; isalpha(*d); part = ++d); + } + if (part && d && *d != '\0') { + for (; isdigit(*d); d++); + if (*d == '\0') + *part = '\0'; + } + return (tmp); +} + +#define PATH_BUF_LEN 64 + +/* * Given a vdev, return the name to display in iostat. If the vdev has a path, * we use that, stripping off any leading "/dev/dsk/"; if not, we use the type. * We also check if this is a whole disk, in which case we strip off the @@ -3064,9 +3148,10 @@ char * zpool_vdev_name(libzfs_handle_t *hdl, zpool_handle_t *zhp, nvlist_t *nv, boolean_t verbose) { - char *path, *devid; + char *path, *devid, *type; uint64_t value; - char buf[64]; + char buf[PATH_BUF_LEN]; + char tmpbuf[PATH_BUF_LEN]; vdev_stat_t *vs; uint_t vsc; @@ -3078,7 +3163,6 @@ zpool_vdev_name(libzfs_handle_t *hdl, zpool_handle_t *zhp, nvlist_t *nv, (u_longlong_t)value); path = buf; } else if (nvlist_lookup_string(nv, ZPOOL_CONFIG_PATH, &path) == 0) { - /* * If the device is dead (faulted, offline, etc) then don't * bother opening it. Otherwise we may be forcing the user to @@ -3117,29 +3201,21 @@ zpool_vdev_name(libzfs_handle_t *hdl, zpool_handle_t *zhp, nvlist_t *nv, devid_str_free(newdevid); } - if (strncmp(path, "/dev/dsk/", 9) == 0) - path += 9; + /* + * For a block device only use the name. + */ + verify(nvlist_lookup_string(nv, ZPOOL_CONFIG_TYPE, &type) == 0); + if (strcmp(type, VDEV_TYPE_DISK) == 0) { + path = strrchr(path, '/'); + path++; + } + /* + * Remove the partition from the path it this is a whole disk. + */ if (nvlist_lookup_uint64(nv, ZPOOL_CONFIG_WHOLE_DISK, &value) == 0 && value) { - int pathlen = strlen(path); - char *tmp = zfs_strdup(hdl, path); - - /* - * If it starts with c#, and ends with "s0", chop - * the "s0" off, or if it ends with "s0/old", remove - * the "s0" from the middle. - */ - if (CTD_CHECK(tmp)) { - if (strcmp(&tmp[pathlen - 2], "s0") == 0) { - tmp[pathlen - 2] = '\0'; - } else if (pathlen > 6 && - strcmp(&tmp[pathlen - 6], "s0/old") == 0) { - (void) strcpy(&tmp[pathlen - 6], - "/old"); - } - } - return (tmp); + return strip_partition(hdl, path); } } else { verify(nvlist_lookup_string(nv, ZPOOL_CONFIG_TYPE, &path) == 0); @@ -3148,6 +3224,7 @@ zpool_vdev_name(libzfs_handle_t *hdl, zpool_handle_t *zhp, nvlist_t *nv, * If it's a raidz device, we need to stick in the parity level. */ if (strcmp(path, VDEV_TYPE_RAIDZ) == 0) { + verify(nvlist_lookup_uint64(nv, ZPOOL_CONFIG_NPARITY, &value) == 0); (void) snprintf(buf, sizeof (buf), "%s%llu", path, @@ -3164,9 +3241,9 @@ zpool_vdev_name(libzfs_handle_t *hdl, zpool_handle_t *zhp, nvlist_t *nv, verify(nvlist_lookup_uint64(nv, ZPOOL_CONFIG_ID, &id) == 0); - (void) snprintf(buf, sizeof (buf), "%s-%llu", path, - (u_longlong_t)id); - path = buf; + (void) snprintf(tmpbuf, sizeof (tmpbuf), "%s-%llu", + path, (u_longlong_t)id); + path = tmpbuf; } } @@ -3186,7 +3263,7 @@ zbookmark_compare(const void *a, const void *b) int zpool_get_errlog(zpool_handle_t *zhp, nvlist_t **nverrlistp) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; uint64_t count; zbookmark_t *zb = NULL; int i; @@ -3282,7 +3359,7 @@ nomem: int zpool_upgrade(zpool_handle_t *zhp, uint64_t new_version) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; libzfs_handle_t *hdl = zhp->zpool_hdl; (void) strcpy(zc.zc_name, zhp->zpool_name); @@ -3344,7 +3421,7 @@ zpool_stage_history(libzfs_handle_t *hdl, const char *history_str) static int get_history(zpool_handle_t *zhp, char *buf, uint64_t *off, uint64_t *len) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; libzfs_handle_t *hdl = zhp->zpool_hdl; (void) strlcpy(zc.zc_name, zhp->zpool_name, sizeof (zc.zc_name)); @@ -3467,11 +3544,99 @@ zpool_get_history(zpool_handle_t *zhp, nvlist_t **nvhisp) return (err); } +/* + * Retrieve the next event. If there is a new event available 'nvp' will + * contain a newly allocated nvlist and 'dropped' will be set to the number + * of missed events since the last call to this function. When 'nvp' is + * set to NULL it indicates no new events are available. In either case + * the function returns 0 and it is up to the caller to free 'nvp'. In + * the case of a fatal error the function will return a non-zero value. + * When the function is called in blocking mode it will not return until + * a new event is available. + */ +int +zpool_events_next(libzfs_handle_t *hdl, nvlist_t **nvp, + int *dropped, int block, int cleanup_fd) +{ + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; + int error = 0; + + *nvp = NULL; + *dropped = 0; + zc.zc_cleanup_fd = cleanup_fd; + + if (!block) + zc.zc_guid = ZEVENT_NONBLOCK; + + if (zcmd_alloc_dst_nvlist(hdl, &zc, ZEVENT_SIZE) != 0) + return (-1); + +retry: + if (zfs_ioctl(hdl, ZFS_IOC_EVENTS_NEXT, &zc) != 0) { + switch (errno) { + case ESHUTDOWN: + error = zfs_error_fmt(hdl, EZFS_POOLUNAVAIL, + dgettext(TEXT_DOMAIN, "zfs shutdown")); + goto out; + case ENOENT: + /* Blocking error case should not occur */ + if (block) + error = zpool_standard_error_fmt(hdl, errno, + dgettext(TEXT_DOMAIN, "cannot get event")); + + goto out; + case ENOMEM: + if (zcmd_expand_dst_nvlist(hdl, &zc) != 0) { + error = zfs_error_fmt(hdl, EZFS_NOMEM, + dgettext(TEXT_DOMAIN, "cannot get event")); + goto out; + } else { + goto retry; + } + default: + error = zpool_standard_error_fmt(hdl, errno, + dgettext(TEXT_DOMAIN, "cannot get event")); + goto out; + } + } + + error = zcmd_read_dst_nvlist(hdl, &zc, nvp); + if (error != 0) + goto out; + + *dropped = (int)zc.zc_cookie; +out: + zcmd_free_nvlists(&zc); + + return (error); +} + +/* + * Clear all events. + */ +int +zpool_events_clear(libzfs_handle_t *hdl, int *count) +{ + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; + char msg[1024]; + + (void) snprintf(msg, sizeof (msg), dgettext(TEXT_DOMAIN, + "cannot clear events")); + + if (zfs_ioctl(hdl, ZFS_IOC_EVENTS_CLEAR, &zc) != 0) + return (zpool_standard_error_fmt(hdl, errno, msg)); + + if (count != NULL) + *count = (int)zc.zc_cookie; /* # of events cleared */ + + return (0); +} + void zpool_obj_to_path(zpool_handle_t *zhp, uint64_t dsobj, uint64_t obj, char *pathname, size_t len) { - zfs_cmd_t zc = { 0 }; + zfs_cmd_t zc = { "\0", "\0", "\0", "\0", 0 }; boolean_t mounted = B_FALSE; char *mntpnt = NULL; char dsname[MAXNAMELEN]; @@ -3532,9 +3697,9 @@ read_efi_label(nvlist_t *config, diskaddr_t *sb) if (nvlist_lookup_string(config, ZPOOL_CONFIG_PATH, &path) != 0) return (err); - (void) snprintf(diskname, sizeof (diskname), "%s%s", RDISK_ROOT, + (void) snprintf(diskname, sizeof (diskname), "%s%s", DISK_ROOT, strrchr(path, '/')); - if ((fd = open(diskname, O_RDONLY|O_NDELAY)) >= 0) { + if ((fd = open(diskname, O_RDWR|O_DIRECT)) >= 0) { struct dk_gpt *vtoc; if ((err = efi_alloc_and_read(fd, &vtoc)) >= 0) { @@ -3580,6 +3745,54 @@ find_start_block(nvlist_t *config) return (MAXOFFSET_T); } +int +zpool_label_disk_wait(char *path, int timeout) +{ + struct stat64 statbuf; + int i; + + /* + * Wait timeout miliseconds for a newly created device to be available + * from the given path. There is a small window when a /dev/ device + * will exist and the udev link will not, so we must wait for the + * symlink. Depending on the udev rules this may take a few seconds. + */ + for (i = 0; i < timeout; i++) { + usleep(1000); + + errno = 0; + if ((stat64(path, &statbuf) == 0) && (errno == 0)) + return (0); + } + + return (ENOENT); +} + +int +zpool_label_disk_check(char *path) +{ + struct dk_gpt *vtoc; + int fd, err; + + if ((fd = open(path, O_RDWR|O_DIRECT)) < 0) + return errno; + + if ((err = efi_alloc_and_read(fd, &vtoc)) != 0) { + (void) close(fd); + return err; + } + + if (vtoc->efi_flags & EFI_GPT_PRIMARY_CORRUPT) { + efi_free(vtoc); + (void) close(fd); + return EIDRM; + } + + efi_free(vtoc); + (void) close(fd); + return 0; +} + /* * Label an individual disk. The name provided is the short name, * stripped of any leading /dev path. @@ -3589,7 +3802,7 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) { char path[MAXPATHLEN]; struct dk_gpt *vtoc; - int fd; + int rval, fd; size_t resv = EFI_MIN_RESV_SIZE; uint64_t slice_size; diskaddr_t start_block; @@ -3602,12 +3815,14 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) if (zhp) { nvlist_t *nvroot; - if (pool_is_bootable(zhp)) { +#if defined(__sun__) || defined(__sun) + if (zpool_is_bootable(zhp)) { zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "EFI labeled devices are not supported on root " "pools.")); return (zfs_error(hdl, EZFS_POOL_NOTSUP, errbuf)); } +#endif verify(nvlist_lookup_nvlist(zhp->zpool_config, ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); @@ -3622,16 +3837,15 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) start_block = NEW_START_BLOCK; } - (void) snprintf(path, sizeof (path), "%s/%s%s", RDISK_ROOT, name, - BACKUP_SLICE); + (void) snprintf(path, sizeof (path), "%s/%s", DISK_ROOT, name); - if ((fd = open(path, O_RDWR | O_NDELAY)) < 0) { + if ((fd = open(path, O_RDWR|O_DIRECT)) < 0) { /* * This shouldn't happen. We've long since verified that this * is a valid device. */ - zfs_error_aux(hdl, - dgettext(TEXT_DOMAIN, "unable to open device")); + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot " + "label '%s': unable to open device: %d"), path, errno); return (zfs_error(hdl, EZFS_OPENFAILED, errbuf)); } @@ -3644,8 +3858,8 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) (void) no_memory(hdl); (void) close(fd); - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "unable to read disk capacity"), name); + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "cannot " + "label '%s': unable to read disk capacity"), path); return (zfs_error(hdl, EZFS_NOCAP, errbuf)); } @@ -3655,6 +3869,7 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) if (start_block == MAXOFFSET_T) start_block = NEW_START_BLOCK; slice_size -= start_block; + slice_size = P2ALIGN(slice_size, PARTITION_END_ALIGNMENT); vtoc->efi_parts[0].p_start = start_block; vtoc->efi_parts[0].p_size = slice_size; @@ -3674,7 +3889,7 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) vtoc->efi_parts[8].p_size = resv; vtoc->efi_parts[8].p_tag = V_RESERVED; - if (efi_write(fd, vtoc) != 0) { + if ((rval = efi_write(fd, vtoc)) != 0 || (rval = efi_rescan(fd)) != 0) { /* * Some block drivers (like pcata) may not support EFI * GPT labels. Print out a helpful error message dir- @@ -3684,123 +3899,36 @@ zpool_label_disk(libzfs_handle_t *hdl, zpool_handle_t *zhp, char *name) (void) close(fd); efi_free(vtoc); - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "try using fdisk(1M) and then provide a specific slice")); + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "try using " + "parted(8) and then provide a specific slice: %d"), rval); return (zfs_error(hdl, EZFS_LABELFAILED, errbuf)); } (void) close(fd); efi_free(vtoc); - return (0); -} -static boolean_t -supported_dump_vdev_type(libzfs_handle_t *hdl, nvlist_t *config, char *errbuf) -{ - char *type; - nvlist_t **child; - uint_t children, c; - - verify(nvlist_lookup_string(config, ZPOOL_CONFIG_TYPE, &type) == 0); - if (strcmp(type, VDEV_TYPE_RAIDZ) == 0 || - strcmp(type, VDEV_TYPE_FILE) == 0 || - strcmp(type, VDEV_TYPE_LOG) == 0 || - strcmp(type, VDEV_TYPE_HOLE) == 0 || - strcmp(type, VDEV_TYPE_MISSING) == 0) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "vdev type '%s' is not supported"), type); - (void) zfs_error(hdl, EZFS_VDEVNOTSUP, errbuf); - return (B_FALSE); - } - if (nvlist_lookup_nvlist_array(config, ZPOOL_CONFIG_CHILDREN, - &child, &children) == 0) { - for (c = 0; c < children; c++) { - if (!supported_dump_vdev_type(hdl, child[c], errbuf)) - return (B_FALSE); - } - } - return (B_TRUE); -} + /* Wait for the first expected partition to appear. */ -/* - * check if this zvol is allowable for use as a dump device; zero if - * it is, > 0 if it isn't, < 0 if it isn't a zvol - */ -int -zvol_check_dump_config(char *arg) -{ - zpool_handle_t *zhp = NULL; - nvlist_t *config, *nvroot; - char *p, *volname; - nvlist_t **top; - uint_t toplevels; - libzfs_handle_t *hdl; - char errbuf[1024]; - char poolname[ZPOOL_MAXNAMELEN]; - int pathlen = strlen(ZVOL_FULL_DEV_DIR); - int ret = 1; - - if (strncmp(arg, ZVOL_FULL_DEV_DIR, pathlen)) { - return (-1); - } - - (void) snprintf(errbuf, sizeof (errbuf), dgettext(TEXT_DOMAIN, - "dump is not supported on device '%s'"), arg); - - if ((hdl = libzfs_init()) == NULL) - return (1); - libzfs_print_on_error(hdl, B_TRUE); - - volname = arg + pathlen; - - /* check the configuration of the pool */ - if ((p = strchr(volname, '/')) == NULL) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "malformed dataset name")); - (void) zfs_error(hdl, EZFS_INVALIDNAME, errbuf); - return (1); - } else if (p - volname >= ZFS_MAXNAMELEN) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "dataset name is too long")); - (void) zfs_error(hdl, EZFS_NAMETOOLONG, errbuf); - return (1); - } else { - (void) strncpy(poolname, volname, p - volname); - poolname[p - volname] = '\0'; - } - - if ((zhp = zpool_open(hdl, poolname)) == NULL) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "could not open pool '%s'"), poolname); - (void) zfs_error(hdl, EZFS_OPENFAILED, errbuf); - goto out; - } - config = zpool_get_config(zhp, NULL); - if (nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, - &nvroot) != 0) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "could not obtain vdev configuration for '%s'"), poolname); - (void) zfs_error(hdl, EZFS_INVALCONFIG, errbuf); - goto out; - } + (void) snprintf(path, sizeof (path), "%s/%s", DISK_ROOT, name); + (void) zfs_append_partition(path, MAXPATHLEN); - verify(nvlist_lookup_nvlist_array(nvroot, ZPOOL_CONFIG_CHILDREN, - &top, &toplevels) == 0); - if (toplevels != 1) { - zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, - "'%s' has multiple top level vdevs"), poolname); - (void) zfs_error(hdl, EZFS_DEVOVERFLOW, errbuf); - goto out; + rval = zpool_label_disk_wait(path, 3000); + if (rval) { + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "failed to " + "detect device partitions on '%s': %d"), path, rval); + return (zfs_error(hdl, EZFS_LABELFAILED, errbuf)); } - if (!supported_dump_vdev_type(hdl, top[0], errbuf)) { - goto out; + /* We can't be to paranoid. Read the label back and verify it. */ + (void) snprintf(path, sizeof (path), "%s/%s", DISK_ROOT, name); + rval = zpool_label_disk_check(path); + if (rval) { + zfs_error_aux(hdl, dgettext(TEXT_DOMAIN, "freshly written " + "EFI label on '%s' is damaged. Ensure\nthis device " + "is not in in use, and is functioning properly: %d"), + path, rval); + return (zfs_error(hdl, EZFS_LABELFAILED, errbuf)); } - ret = 0; -out: - if (zhp) - zpool_close(zhp); - libzfs_fini(hdl); - return (ret); + return 0; }