/* * Detect if any leaf devices that have seen errors or could not be opened. */ static boolean_t find_vdev_problem(nvlist_t *vdev, int (*func)(vdev_stat_t *, uint_t), boolean_t ignore_replacing) { nvlist_t **child; vdev_stat_t *vs; uint_t c, vsc, children; /* * Ignore problems within a 'replacing' vdev, since we're presumably in * the process of repairing any such errors, and don't want to call them * out again. We'll pick up the fact that a resilver is happening * later. */ if (ignore_replacing == B_TRUE) { char *type; verify(nvlist_lookup_string(vdev, ZPOOL_CONFIG_TYPE, &type) == 0); if (strcmp(type, VDEV_TYPE_REPLACING) == 0) return (B_FALSE); } if (nvlist_lookup_nvlist_array(vdev, ZPOOL_CONFIG_CHILDREN, &child, &children) == 0) { for (c = 0; c < children; c++) if (find_vdev_problem(child[c], func, ignore_replacing)) return (B_TRUE); } else { verify(nvlist_lookup_uint64_array(vdev, ZPOOL_CONFIG_VDEV_STATS, (uint64_t **)&vs, &vsc) == 0); if (func(vs, vsc) != 0) return (B_TRUE); } /* * Check any L2 cache devs */ if (nvlist_lookup_nvlist_array(vdev, ZPOOL_CONFIG_L2CACHE, &child, &children) == 0) { for (c = 0; c < children; c++) if (find_vdev_problem(child[c], func, ignore_replacing)) return (B_TRUE); } return (B_FALSE); }
/* * Detect if any leaf devices that have seen errors or could not be opened. */ static boolean_t find_vdev_problem(nvlist_t *vdev, int (*func)(uint64_t, uint64_t, uint64_t)) { nvlist_t **child; vdev_stat_t *vs; uint_t c, children; char *type; /* * Ignore problems within a 'replacing' vdev, since we're presumably in * the process of repairing any such errors, and don't want to call them * out again. We'll pick up the fact that a resilver is happening * later. */ verify(nvlist_lookup_string(vdev, ZPOOL_CONFIG_TYPE, &type) == 0); if (strcmp(type, VDEV_TYPE_REPLACING) == 0) return (B_FALSE); if (nvlist_lookup_nvlist_array(vdev, ZPOOL_CONFIG_CHILDREN, &child, &children) == 0) { for (c = 0; c < children; c++) if (find_vdev_problem(child[c], func)) return (B_TRUE); } else { verify(nvlist_lookup_uint64_array(vdev, ZPOOL_CONFIG_STATS, (uint64_t **)&vs, &c) == 0); if (func(vs->vs_state, vs->vs_aux, vs->vs_read_errors + vs->vs_write_errors + vs->vs_checksum_errors)) return (B_TRUE); } return (B_FALSE); }
/* * Active pool health status. * * To determine the status for a pool, we make several passes over the config, * picking the most egregious error we find. In order of importance, we do the * following: * * - Check for a complete and valid configuration * - Look for any missing devices in a non-replicated config * - Check for any data errors * - Check for any missing devices in a replicated config * - Look for any devices showing errors * - Check for any resilvering devices * * There can obviously be multiple errors within a single pool, so this routine * only picks the most damaging of all the current errors to report. */ static zpool_status_t check_status(nvlist_t *config, boolean_t isimport) { nvlist_t *nvroot; vdev_stat_t *vs; uint_t vsc; uint64_t nerr; uint64_t version; verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, &version) == 0); verify(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); verify(nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_STATS, (uint64_t **)&vs, &vsc) == 0); /* * Newer on-disk version. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_VERSION_NEWER) return (ZPOOL_STATUS_VERSION_NEWER); /* * Check that the config is complete. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_GUID_SUM) return (ZPOOL_STATUS_BAD_GUID_SUM); /* * Missing devices in non-replicated config. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_NR); /* * Corrupted pool metadata */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_CORRUPT_DATA) return (ZPOOL_STATUS_CORRUPT_POOL); /* * Persistent data errors. */ if (!isimport) { if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_ERRCOUNT, &nerr) == 0 && nerr != 0) return (ZPOOL_STATUS_CORRUPT_DATA); } /* * Missing devices in a replicated config. */ if (find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_R); if (find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_R); /* * Devices with errors */ if (!isimport && find_vdev_problem(nvroot, vdev_errors)) return (ZPOOL_STATUS_FAILING_DEV); /* * Offlined devices */ if (find_vdev_problem(nvroot, vdev_offlined)) return (ZPOOL_STATUS_OFFLINE_DEV); /* * Currently resilvering */ if (!vs->vs_scrub_complete && vs->vs_scrub_type == POOL_SCRUB_RESILVER) return (ZPOOL_STATUS_RESILVERING); /* * Outdated, but usable, version */ if (version < ZFS_VERSION) return (ZPOOL_STATUS_VERSION_OLDER); return (ZPOOL_STATUS_OK); }
/* * Active pool health status. * * To determine the status for a pool, we make several passes over the config, * picking the most egregious error we find. In order of importance, we do the * following: * * - Check for a complete and valid configuration * - Look for any faulted or missing devices in a non-replicated config * - Check for any data errors * - Check for any faulted or missing devices in a replicated config * - Look for any devices showing errors * - Check for any resilvering devices * * There can obviously be multiple errors within a single pool, so this routine * only picks the most damaging of all the current errors to report. */ static zpool_status_t check_status(nvlist_t *config, boolean_t isimport) { nvlist_t *nvroot; vdev_stat_t *vs; pool_scan_stat_t *ps = NULL; uint_t vsc, psc; uint64_t nerr; uint64_t version; uint64_t stateval; uint64_t suspended; uint64_t hostid = 0; verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, &version) == 0); verify(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); verify(nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_VDEV_STATS, (uint64_t **)&vs, &vsc) == 0); verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_STATE, &stateval) == 0); /* * Currently resilvering a vdev */ (void) nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_SCAN_STATS, (uint64_t **)&ps, &psc); if (ps && ps->pss_func == POOL_SCAN_RESILVER && ps->pss_state == DSS_SCANNING) return (ZPOOL_STATUS_RESILVERING); /* * Pool last accessed by another system. */ (void) nvlist_lookup_uint64(config, ZPOOL_CONFIG_HOSTID, &hostid); if (hostid != 0 && (unsigned long)hostid != gethostid() && stateval == POOL_STATE_ACTIVE) return (ZPOOL_STATUS_HOSTID_MISMATCH); /* * Newer on-disk version. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_VERSION_NEWER) return (ZPOOL_STATUS_VERSION_NEWER); /* * Unsupported feature(s). */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_UNSUP_FEAT) { nvlist_t *nvinfo; verify(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_LOAD_INFO, &nvinfo) == 0); if (nvlist_exists(nvinfo, ZPOOL_CONFIG_CAN_RDONLY)) return (ZPOOL_STATUS_UNSUP_FEAT_WRITE); return (ZPOOL_STATUS_UNSUP_FEAT_READ); } /* * Check that the config is complete. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_GUID_SUM) return (ZPOOL_STATUS_BAD_GUID_SUM); /* * Check whether the pool has suspended due to failed I/O. */ if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_SUSPENDED, &suspended) == 0) { if (suspended == ZIO_FAILURE_MODE_CONTINUE) return (ZPOOL_STATUS_IO_FAILURE_CONTINUE); return (ZPOOL_STATUS_IO_FAILURE_WAIT); } /* * Could not read a log. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_LOG) { return (ZPOOL_STATUS_BAD_LOG); } /* * Bad devices in non-replicated config. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_faulted, B_TRUE)) return (ZPOOL_STATUS_FAULTED_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_missing, B_TRUE)) return (ZPOOL_STATUS_MISSING_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_broken, B_TRUE)) return (ZPOOL_STATUS_CORRUPT_LABEL_NR); /* * Corrupted pool metadata */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_CORRUPT_DATA) return (ZPOOL_STATUS_CORRUPT_POOL); /* * Persistent data errors. */ if (!isimport) { if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_ERRCOUNT, &nerr) == 0 && nerr != 0) return (ZPOOL_STATUS_CORRUPT_DATA); } /* * Missing devices in a replicated config. */ if (find_vdev_problem(nvroot, vdev_faulted, B_TRUE)) return (ZPOOL_STATUS_FAULTED_DEV_R); if (find_vdev_problem(nvroot, vdev_missing, B_TRUE)) return (ZPOOL_STATUS_MISSING_DEV_R); if (find_vdev_problem(nvroot, vdev_broken, B_TRUE)) return (ZPOOL_STATUS_CORRUPT_LABEL_R); /* * Devices with errors */ if (!isimport && find_vdev_problem(nvroot, vdev_errors, B_TRUE)) return (ZPOOL_STATUS_FAILING_DEV); /* * Offlined devices */ if (find_vdev_problem(nvroot, vdev_offlined, B_TRUE)) return (ZPOOL_STATUS_OFFLINE_DEV); /* * Removed device */ if (find_vdev_problem(nvroot, vdev_removed, B_TRUE)) return (ZPOOL_STATUS_REMOVED_DEV); /* * Suboptimal, but usable, ashift configuration. */ if (find_vdev_problem(nvroot, vdev_non_native_ashift, B_FALSE)) return (ZPOOL_STATUS_NON_NATIVE_ASHIFT); /* * Outdated, but usable, version */ if (SPA_VERSION_IS_SUPPORTED(version) && version != SPA_VERSION) return (ZPOOL_STATUS_VERSION_OLDER); /* * Usable pool with disabled features */ if (version >= SPA_VERSION_FEATURES) { int i; nvlist_t *feat; if (isimport) { feat = fnvlist_lookup_nvlist(config, ZPOOL_CONFIG_LOAD_INFO); feat = fnvlist_lookup_nvlist(feat, ZPOOL_CONFIG_ENABLED_FEAT); } else { feat = fnvlist_lookup_nvlist(config, ZPOOL_CONFIG_FEATURE_STATS); } for (i = 0; i < SPA_FEATURES; i++) { zfeature_info_t *fi = &spa_feature_table[i]; if (!nvlist_exists(feat, fi->fi_guid)) return (ZPOOL_STATUS_FEAT_DISABLED); } } return (ZPOOL_STATUS_OK); }
/* * Active pool health status. * * To determine the status for a pool, we make several passes over the config, * picking the most egregious error we find. In order of importance, we do the * following: * * - Check for a complete and valid configuration * - Look for any faulted or missing devices in a non-replicated config * - Check for any data errors * - Check for any faulted or missing devices in a replicated config * - Look for any devices showing errors * - Check for any resilvering devices * * There can obviously be multiple errors within a single pool, so this routine * only picks the most damaging of all the current errors to report. */ static zpool_status_t check_status(zpool_handle_t *zhp, nvlist_t *config, boolean_t isimport) { nvlist_t *nvroot; vdev_stat_t *vs; uint_t vsc; uint64_t nerr; uint64_t version; uint64_t stateval; uint64_t hostid = 0; verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, &version) == 0); verify(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); verify(nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_STATS, (uint64_t **)&vs, &vsc) == 0); verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_STATE, &stateval) == 0); (void) nvlist_lookup_uint64(config, ZPOOL_CONFIG_HOSTID, &hostid); /* * Pool last accessed by another system. */ if (hostid != 0 && (unsigned long)hostid != gethostid() && stateval == POOL_STATE_ACTIVE) return (ZPOOL_STATUS_HOSTID_MISMATCH); /* * Newer on-disk version. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_VERSION_NEWER) return (ZPOOL_STATUS_VERSION_NEWER); /* * Check that the config is complete. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_GUID_SUM) return (ZPOOL_STATUS_BAD_GUID_SUM); /* * Pool has experienced failed I/O. */ if (stateval == POOL_STATE_IO_FAILURE) { zpool_handle_t *tmp_zhp = NULL; libzfs_handle_t *hdl = NULL; char property[ZPOOL_MAXPROPLEN]; char *failmode = NULL; if (zhp == NULL) { char *poolname; verify(nvlist_lookup_string(config, ZPOOL_CONFIG_POOL_NAME, &poolname) == 0); if ((hdl = libzfs_init()) == NULL) return (ZPOOL_STATUS_IO_FAILURE_WAIT); tmp_zhp = zpool_open_canfail(hdl, poolname); if (tmp_zhp == NULL) { libzfs_fini(hdl); return (ZPOOL_STATUS_IO_FAILURE_WAIT); } } if (zpool_get_prop(zhp ? zhp : tmp_zhp, ZPOOL_PROP_FAILUREMODE, property, sizeof (property), NULL) == 0) failmode = property; if (tmp_zhp != NULL) zpool_close(tmp_zhp); if (hdl != NULL) libzfs_fini(hdl); if (failmode == NULL) return (ZPOOL_STATUS_IO_FAILURE_WAIT); if (strncmp(failmode, "continue", strlen("continue")) == 0) return (ZPOOL_STATUS_IO_FAILURE_CONTINUE); else return (ZPOOL_STATUS_IO_FAILURE_WAIT); } /* * Could not read a log. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_LOG) { return (ZPOOL_STATUS_BAD_LOG); } /* * Bad devices in non-replicated config. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_faulted)) return (ZPOOL_STATUS_FAULTED_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_NR); /* * Corrupted pool metadata */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_CORRUPT_DATA) return (ZPOOL_STATUS_CORRUPT_POOL); /* * Persistent data errors. */ if (!isimport) { if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_ERRCOUNT, &nerr) == 0 && nerr != 0) return (ZPOOL_STATUS_CORRUPT_DATA); } /* * Missing devices in a replicated config. */ if (find_vdev_problem(nvroot, vdev_faulted)) return (ZPOOL_STATUS_FAULTED_DEV_R); if (find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_R); if (find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_R); /* * Devices with errors */ if (!isimport && find_vdev_problem(nvroot, vdev_errors)) return (ZPOOL_STATUS_FAILING_DEV); /* * Offlined devices */ if (find_vdev_problem(nvroot, vdev_offlined)) return (ZPOOL_STATUS_OFFLINE_DEV); /* * Currently resilvering */ if (!vs->vs_scrub_complete && vs->vs_scrub_type == POOL_SCRUB_RESILVER) return (ZPOOL_STATUS_RESILVERING); /* * Outdated, but usable, version */ if (version < SPA_VERSION) return (ZPOOL_STATUS_VERSION_OLDER); return (ZPOOL_STATUS_OK); }
/* * Active pool health status. * * To determine the status for a pool, we make several passes over the config, * picking the most egregious error we find. In order of importance, we do the * following: * * - Check for a complete and valid configuration * - Look for any faulted or missing devices in a non-replicated config * - Check for any data errors * - Check for any faulted or missing devices in a replicated config * - Look for any devices showing errors * - Check for any resilvering devices * * There can obviously be multiple errors within a single pool, so this routine * only picks the most damaging of all the current errors to report. */ static zpool_status_t check_status(nvlist_t *config, boolean_t isimport) { nvlist_t *nvroot; vdev_stat_t *vs; pool_scan_stat_t *ps = NULL; uint_t vsc, psc; uint64_t nerr; uint64_t version; uint64_t stateval; uint64_t suspended; uint64_t hostid = 0; unsigned long system_hostid = gethostid() & 0xffffffff; verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_VERSION, &version) == 0); verify(nvlist_lookup_nvlist(config, ZPOOL_CONFIG_VDEV_TREE, &nvroot) == 0); verify(nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_VDEV_STATS, (uint64_t **)&vs, &vsc) == 0); verify(nvlist_lookup_uint64(config, ZPOOL_CONFIG_POOL_STATE, &stateval) == 0); /* * Currently resilvering a vdev */ (void) nvlist_lookup_uint64_array(nvroot, ZPOOL_CONFIG_SCAN_STATS, (uint64_t **)&ps, &psc); if (ps && ps->pss_func == POOL_SCAN_RESILVER && ps->pss_state == DSS_SCANNING) return (ZPOOL_STATUS_RESILVERING); /* * Pool last accessed by another system. */ (void) nvlist_lookup_uint64(config, ZPOOL_CONFIG_HOSTID, &hostid); if (hostid != 0 && (unsigned long)hostid != system_hostid && stateval == POOL_STATE_ACTIVE) return (ZPOOL_STATUS_HOSTID_MISMATCH); /* * Newer on-disk version. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_VERSION_NEWER) return (ZPOOL_STATUS_VERSION_NEWER); /* * Check that the config is complete. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_GUID_SUM) return (ZPOOL_STATUS_BAD_GUID_SUM); /* * Check whether the pool has suspended due to failed I/O. */ if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_SUSPENDED, &suspended) == 0) { if (suspended == ZIO_FAILURE_MODE_CONTINUE) return (ZPOOL_STATUS_IO_FAILURE_CONTINUE); return (ZPOOL_STATUS_IO_FAILURE_WAIT); } /* * Could not read a log. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_BAD_LOG) { return (ZPOOL_STATUS_BAD_LOG); } /* * Bad devices in non-replicated config. */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_faulted)) return (ZPOOL_STATUS_FAULTED_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_NR); if (vs->vs_state == VDEV_STATE_CANT_OPEN && find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_NR); /* * Corrupted pool metadata */ if (vs->vs_state == VDEV_STATE_CANT_OPEN && vs->vs_aux == VDEV_AUX_CORRUPT_DATA) return (ZPOOL_STATUS_CORRUPT_POOL); /* * Persistent data errors. */ if (!isimport) { if (nvlist_lookup_uint64(config, ZPOOL_CONFIG_ERRCOUNT, &nerr) == 0 && nerr != 0) return (ZPOOL_STATUS_CORRUPT_DATA); } /* * Missing devices in a replicated config. */ if (find_vdev_problem(nvroot, vdev_faulted)) return (ZPOOL_STATUS_FAULTED_DEV_R); if (find_vdev_problem(nvroot, vdev_missing)) return (ZPOOL_STATUS_MISSING_DEV_R); if (find_vdev_problem(nvroot, vdev_broken)) return (ZPOOL_STATUS_CORRUPT_LABEL_R); /* * Devices with errors */ if (!isimport && find_vdev_problem(nvroot, vdev_errors)) return (ZPOOL_STATUS_FAILING_DEV); /* * Offlined devices */ if (find_vdev_problem(nvroot, vdev_offlined)) return (ZPOOL_STATUS_OFFLINE_DEV); /* * Removed device */ if (find_vdev_problem(nvroot, vdev_removed)) return (ZPOOL_STATUS_REMOVED_DEV); /* * Outdated, but usable, version */ if (version < SPA_VERSION) return (ZPOOL_STATUS_VERSION_OLDER); return (ZPOOL_STATUS_OK); }