Outils de lecture (read)
Ces outils lisent l’état des sauvegardes. Ils ne changent rien. Un jeton OAuth avec la portée read voit ces 32 outils. Les textes de la source sont en anglais.
get_backup_health
Backup health. Portée read. Lecture seule : oui. Destructif : non.
Health of the backups of the MSP: agents online and offline, M365 tenants, stored bytes, the open backup problems of the last 7 days (failed, partial, missed or interrupted jobs) grouped by client: a problem is resolved, and not shown, when a later job of the same endpoint, type and policy succeeded (for an agent update, also when the agent runs its version); each problem has resolved_by_job_id (null). A problem acknowledged by a person in the portal (same endpoint, type and policy, failures up to the time of the acknowledgement) is not shown either, as on the portal home; a failure after the acknowledgement is shown. With include_acknowledged: also the acknowledged problems, each with acknowledged (by, at, comment; null when not acknowledged). No tool acknowledges a problem. Also the open "first backup unfinished" alerts. With client_id: the problem jobs and the open "first backup unfinished" alerts of that client only. With tag or no_tag: only the endpoints with these tags (a node of get_fleet_tree), for example the problems of the site Lyon; the first backup alerts too. Also the open SQL Server alerts (mssql_alerts): mssql_rights_missing (the login of an instance is not sysadmin) and mssql_log_growing (the transaction log of a database grows: a full backup never empties it). A "first backup unfinished" alert is a policy (files or bare metal image) with no restore point N days after its first run started (N: the endpoint's first_backup_alert days, else the stack default of get_alert_settings). Until a first restore point, nothing is pruned: the alert gives the bytes already stored and last_progress_at, the last progress of its runs. It closes at the first restore point. With no progress, the platform deletes the data of the abandoned first backup after reset_after (null: never), on a repo with no restore point only; a "first backup reset" alert (first_backup_resets) then tells it, until the next restore point. No tool starts or delays a deletion.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | non | Id of the client. |
tag | array of string | non | Tags the endpoint must have, all of them (AND). Each one is key:value, for example site:Lyon. Key and value compare without case. Use list_tag_keys to get the keys and values that exist; never guess a name. Éléments : 1 à 30. |
no_tag | array of string | non | Tag keys the endpoint must not have, all of them (AND), for example site: the endpoints without a site. Éléments : 1 à 30. |
include_acknowledged | boolean | non | Also the problems acknowledged in the portal (not resolved), with who, when and the comment. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
list_clients
List clients. Portée read. Lecture seule : oui. Destructif : non.
Lists the clients (customers) of the MSP, or finds a client from an approximate name. With several matches, the answer lists them: ask the user which one, never choose.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
query | string | non | Name or part of a name, any case. When nothing matches, each word is tried alone. Longueur : 1 à 255. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_client
Get a client. Portée read. Lecture seule : oui. Destructif : non.
One client of the MSP: name, number of endpoints, stored bytes, pending deletion.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | oui | Id of the client. |
list_endpoints
List endpoints. Portée read. Lecture seule : oui. Destructif : non.
Lists the endpoints (machines with the agent, M365 tenants) of the MSP, or finds a machine from an approximate name. Filters: client, type, presence, problems only, boot mode (boot), agent version too old (too_old_only), tags (tag: has all these key:value tags; no_tag: has none of these keys). The filter of a node of get_fleet_tree gives its endpoints. Each item has its client name, its tags, and for an agent agent_version, min_agent_version and agent_version_state (too_old: the agent is blocked, no backup or restore until it is updated; unknown: no connection yet). With several matches, ask the user which one, never choose.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
query | string | non | Name or part of a name, any case. When nothing matches, each word is tried alone. Longueur : 1 à 255. |
client_id | string (uuid) | non | Id of the client. |
type | "agent" | "m365" | non | |
agent_status | "online" | "offline" | non | |
problems_only | boolean | non | Only endpoints with a problem: agent offline, or last backup failed, interrupted, missed or partial. |
boot | "uefi" | "bios" | "unknown" | non | Boot mode of the machine (agent.boot): uefi (UEFI and a GPT system disk), bios (BIOS firmware or MBR system disk: no image backup), unknown (an older agent that does not send it, a disk it could not read, or an M365 tenant) |
too_old_only | boolean | non | Only agents whose version is below the minimum version of the platform (agent_version_state too_old): they are blocked. |
tag | array of string | non | Tags the endpoint must have, all of them (AND). Each one is key:value, for example site:Lyon. Key and value compare without case. Use list_tag_keys to get the keys and values that exist; never guess a name. Éléments : 1 à 30. |
no_tag | array of string | non | Tag keys the endpoint must not have, all of them (AND), for example site: the endpoints without a site. Éléments : 1 à 30. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_endpoint
Get an endpoint. Portée read. Lecture seule : oui. Destructif : non.
One endpoint: its agent (OS, version, online or offline, boot: firmware uefi or bios and system disk gpt, mbr or unknown, null when the agent does not send it), agent_version, min_agent_version and agent_version_state (too_old: below the minimum version of the platform, the agent is blocked: no backup or restore until it is updated by hand; doc_url of an agent_too_old error gives the procedure), its last backups, its file backup policy and its bare metal (image) policy, with blocked_reason baremetal_unsupported_boot when the machine is BIOS or MBR: no image backup runs, with the pending changes of these policies and schedule_effective (the window_minutes and first_backup_window its runs use: the schedule's value, else the policy's), and first_backup_alert (agents only): days, the effective days before the "first backup unfinished" alert; override_days, the value set on this endpoint, or null for the stack default (default_days); and first_backup_reset (agents only): days with no progress before the data of an abandoned first backup is deleted (0: never), override_days, default_days, and planned_at, the date after which it is deleted while the "first backup unfinished" alert is open (null: none planned). Read only: no tool starts or delays a deletion. On a Windows agent, mssql: the SQL Server instances with their databases (state, recovery model, sizes, last backup, excluded such as tde or availability_group, warnings such as log_growing) and rights_missing when the login is not sysadmin, mssql_service_account, the Windows account of the agent service (for example NT SERVICE\BaaStide, or the service name of a white label agent): with rights_missing and a connection_auth of service, this account needs a SQL Server login with the sysadmin role, and mssql_policies, the SQL Server policies with their pending changes. Never a SQL Server login or password. For a Hyper-V host, hyperv.vms: each VM with its vm_key, name, state, protectable or not and why, excluded or not, last restore point and last result (list_vm_snapshots, preview_vm_restore). For a Proxmox VE node, proxmox: the node (version, pve_supported, storages with free space, bridges), unprotectable_reason (pve_unsupported_version: no guest is backed up) and vms: each VM and container (kind: ct) with the same fields, plus tpm (a VM with a TPM is restored in the portal only), the lock, and for a container the mount points not backed up (bind mounts, devices).
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the endpoint. |
list_jobs
List jobs. Portée read. Lecture seule : oui. Destructif : non.
Jobs of an endpoint (backups, image backups, maintenance), newest first: status, cause, dates, counters. No log lines. A failed, partial, interrupted or missed job has an error_summary: a typed cause, a fixed message for that cause, file_error_codes (number of unreadable files per typed code, such as permission_denied or file_in_use) and portal_url, the job page of the portal. It never holds file names, paths or the text of the agent: they can hold customer data and text that is not from BaaStide. To see which files failed, the user opens portal_url. A succeeded or partial backup has a volume: source (bytes and count read, unit files, databases, disks or volumes), changed and new (files changed and new since the previous backup; null for a full backup: SQL Server, image, VM) and sent (new bytes stored; compressed false when the agent sent only the size before compression). Counters only.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the endpoint. |
type | "backup" | "baremetal_backup" | "mssql_backup" | "vm_backup" | "init" | "unlock" | "forget" | "prune" | "check" | "migrate" | "update" | "reset_first_backup" | non | |
status | "scheduled" | "dispatched" | "running" | "succeeded" | "partial" | "failed" | "cancelled" | "interrupted" | "preempted" | "missed" | "superseded" | non | |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_job
Get a job. Portée read. Lecture seule : oui. Destructif : non.
One job: status, typed cause, dates, backup counters, live progress while it runs. The free text of an error and the log lines are not returned. A failed, partial, interrupted or missed job has an error_summary: a typed cause, a fixed message for that cause, file_error_codes (number of unreadable files per typed code, such as permission_denied or file_in_use) and portal_url, the job page of the portal. It never holds file names, paths or the text of the agent: they can hold customer data and text that is not from BaaStide. To see which files failed, the user opens portal_url. A succeeded or partial backup has a volume: source (bytes and count read, unit files, databases, disks or volumes), changed and new (files changed and new since the previous backup; null for a full backup: SQL Server, image, VM) and sent (new bytes stored; compressed false when the agent sent only the size before compression). Counters only.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
job_id | string (uuid) | oui | Id of the job. |
list_snapshots
List restore points. Portée read. Lecture seule : oui. Destructif : non.
Restore points (snapshots) of an agent endpoint, newest first: date, kind (files or bare metal image), sizes, number of files. Filter by date: date (one UTC day), or after / before. Without a match, the answer gives the nearest older and newer points. To find a file of a restore point, use search_snapshot; to list a folder, browse_snapshot. With backup_kind mssql: the SQL Server restore points, with their instance, number of databases and total size; list_snapshot_databases gives their databases.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the endpoint. |
date | string (date) | non | A UTC day, YYYY-MM-DD. |
after | string (date-time) | non | Only points at or after this time (ISO 8601) |
before | string (date-time) | non | Only points before this time (ISO 8601) |
backup_kind | "files" | "baremetal" | "mssql" | "hyperv" | "proxmox" | non | files (default for a file restore), baremetal or mssql (SQL Server databases) |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
list_restore_sessions
List restores. Portée read. Lecture seule : oui. Destructif : non.
Restore sessions where the agent is the source or the target, newest first: state, counters, error codes. Never the failed paths. A SQL Server restore (kind mssql) gives its instance and its databases with their new names, and the result per database once known.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the endpoint. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_restore_session
Follow a restore. Portée read. Lecture seule : oui. Destructif : non.
State of one restore session: progress, files and bytes restored, skipped files, number of errors and their codes, duration, destination. The destination says where the files are written on the target (the folder, and the rule: the full path of each file is kept under it). The answer says if the restore is finished; if not, call again later. Never the failed paths. A SQL Server restore gives the result of each database (stats.databases: status, typed code, state of the new database, DBCC CHECKDB result, number of orphaned users).
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
restore_session_id | string (uuid) | oui | Id of the restore session. |
list_pending_actions
List delayed changes. Portée read. Lecture seule : oui. Destructif : non.
Destructive changes delayed by 72 hours (shorter retention, removed paths, endpoint deletion, ...): what changes, when it applies. A pending one can be cancelled with cancel_pending_action. update_file_policy and update_baremetal_policy also cancel the delayed changes the new policy no longer asks for (status cancelled, audit reason policy_updated), without cancel_pending_action.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
status | "pending" | "cancelled" | "applied" | non | Défaut : "pending". |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_usage
Storage use. Portée read. Lecture seule : oui. Destructif : non.
Daily storage use per endpoint between two UTC days (stored bytes, source bytes, M365 billable users). Filter by client or endpoint.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
from | string (date) | oui | First UTC day, YYYY-MM-DD. |
to | string (date) | oui | Last UTC day, YYYY-MM-DD. |
client_id | string (uuid) | non | Id of the client. |
endpoint_id | string (uuid) | non | Id of the endpoint. |
limit | integer | non | Number of items, 1 to 100. Défaut : 100. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_audit_log
Audit log. Portée read. Lecture seule : oui. Destructif : non.
Audit log of the MSP, newest first: who did what, on which object, when. The details of each entry are not returned.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
action | string | non | An action, for example policy.updated or restore_session.requested. |
endpoint_id | string (uuid) | non | Id of the endpoint. |
from | string (date-time) | non | |
to | string (date-time) | non | |
limit | integer | non | Number of items, 1 to 100. Défaut : 50. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
get_alert_settings
Alert settings. Portée read. Lecture seule : oui. Destructif : non.
Alert rules of the MSP (which events send an email, thresholds). With client_id: the overrides of that client too. first_backup_alert_default_days: the stack default of the days before the "first backup unfinished" alert; an endpoint can override it (get_endpoint, first_backup_alert).
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | non | Id of the client. |
get_install_profile
Install profile. Portée read. Lecture seule : oui. Destructif : non.
Install profile of the MSP (white label of the agent): the names a new agent gets (product, publisher, short name, Windows service name, description, support URL), its own domain (public_host: the install commands, the agents it enrolls and the email links of the MSP are on https://<public_host>; null: the platform URL), its revision, and the URLs of the Linux and Windows install scripts. profile_id null: no saved profile, new agents get the BaaStide names. A change applies to new installs only: get_endpoint gives the profile id and revision each agent was installed with. Read only: the profile is changed in the portal, by an MSP admin. Never an enrollment token: create_enrollment_token gives the install commands.
Paramètres
Aucun paramètre.
get_m365_status
M365 tenant status. Portée read. Lecture seule : oui. Destructif : non.
State of one M365 tenant (an endpoint of type m365): the connection to Microsoft 365 (connected, reconnect required, permission missing), the certificate expiry, the Graph permissions granted, and the latest backup runs with their number of units by state. Aggregated only: no mailbox, drive or item.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the M365 endpoint. |
runs | integer | non | Number of latest runs, 1 to 30. Défaut : 7. Valeur : 1 à 30. |
license_status
Licence status. Portée read. Lecture seule : oui. Destructif : non.
State of the licence of the BaaStide platform: valid, grace, overdraft, exhausted, none (restore only) or trial_expired; if backups and new machines are allowed; type, dates, annual commitment and balance (in millipoints: 1 point = 1000), the usage of the month and its cap (postpaid), the trial end and its data (locked or not), and the current alerts. A backup tool answers license_required when the licence does not allow backups. Only a platform administrator imports a licence, in the portal.
Paramètres
Aucun paramètre.
license_usage
Licence consumption. Portée read. Lecture seule : oui. Destructif : non.
Consumption of points of the BaaStide platform: per day and per workload type (agent_server, agent_workstation, vm, m365_user), the count of protected workloads and the points consumed; the local balance and the projection to the end of the annual commitment (pace of the last 30 days). Amounts in millipoints: 1 point = 1000. Counts only: no machine, file or user names. A day with charged: false had no backup allowed and consumed nothing.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
days | integer | non | Number of days, newest first (default 30, at most 90). Valeur : 1 à 90. |
get_fleet_tree
Fleet tree. Portée read. Lecture seule : oui. Destructif : non.
The tree of the fleet (the "Parc" page of the portal): the clients, then the values of the tree tag keys of the MSP (for example site, then role), in the order the MSP set. Each node gives its number of endpoints, of failed endpoints (red: last backup failed) and of warning endpoints (orange: offline, last backup interrupted, missed or partial, or no succeeded backup for 7 days). A node with value null holds the endpoints without that key. Without client_id: the clients, also the clients without an endpoint (with endpoints: 0). To open a node, give its filter (client_id, tag, no_tag). To list the endpoints of a node, give its filter to list_endpoints; its problems, to get_backup_health.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | non | Id of the client of the node to open. |
tag | array of string | non | With client_id: the tag of the filter of the node to open, key:value items. Éléments : 1 à 30. |
no_tag | array of string | non | With client_id: the no_tag of the filter of the node to open. Éléments : 1 à 30. |
query | string | non | Keeps the nodes with an endpoint whose name, client name or tree tag value holds this text, any case. Longueur : 1 à 255. |
problems_only | boolean | non | Keeps the nodes with a failed or warning endpoint. |
depth | integer | non | Levels to give under the node: 1 gives its children only. 1 to 3; at most 20 nodes are opened in one call. Défaut : 1. Valeur : 1 à 3. |
list_tag_keys
List tag keys and values. Portée read. Lecture seule : oui. Destructif : non.
The tag keys of the MSP (for example site, role) and their values (for example Lyon, Paris), with the number of endpoints of each. tree_level tells if a key is a level of the fleet tree (1 is the first level under the client), null for a free key. Without key: every key, with the 10 most used values of the first 10 keys. With key: the values of that key. Use it before a tag or no_tag filter, so that the names are the real ones.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
key | string | non | A tag key, any case: gives its values, most used first. Longueur : 1 à 64. |
query | string | non | With key: only the values that start with this text, any case. Longueur : 1 à 128. |
limit | integer | non | With key: number of values, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
list_m365_restores
List M365 restores. Portée read. Lecture seule : oui. Destructif : non.
Lists the latest OneDrive / SharePoint restores of one M365 tenant (an endpoint of type m365), newest first: state, progress (items restored, skipped, failed, to review), totals and the typed error. Never the names of the restored files.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the M365 endpoint. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
get_m365_restore
M365 restore. Portée read. Lecture seule : oui. Destructif : non.
State of one OneDrive / SharePoint restore: status, progress, totals, restoration folder and typed error. Never the names of the restored files.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the M365 endpoint. |
restore_id | string (uuid) | oui | Id of the restore. |
list_m365_exports
List M365 exports. Portée read. Lecture seule : oui. Destructif : non.
Lists the latest exports of one M365 tenant (an endpoint of type m365), newest first: content, state, progress (items, bytes), typed error, end of the download window (expires_at) and portal_url, the export page of the portal. The file is downloaded in the BaaStide portal only, by a person who logs in: give the user portal_url.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the M365 endpoint. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
get_m365_export
M365 export. Portée read. Lecture seule : oui. Destructif : non.
State of one export of an M365 tenant: status, progress (items, bytes), number of files, typed error, end of the download window (expires_at) and portal_url, the export page of the portal. The file is downloaded in the BaaStide portal only, by a person who logs in: give the user portal_url.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the M365 endpoint. |
export_id | string (uuid) | oui | Id of the export. |
browse_snapshot
List a folder of a restore point. Portée read. Lecture seule : oui. Destructif : non.
Lists the files and folders of one folder of a restore point (snapshot) of an agent: name, type, size, modification date. Never the content of a file. Start at / and go down; to find a file from its name, use search_snapshot instead. A path of the answer can be given as is to start_restore, after the user confirms it.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the agent endpoint that made the backup. |
snapshot_id | string | oui | Id of the restore point (list_snapshots) |
path | any | non | The folder to list, in the form of the backup: /etc/compta on Linux, /C/Users/bob for C:\Users\bob on Windows. Default: / Défaut : "/". |
limit | integer | non | Number of entries, 1 to 200. Défaut : 100. Valeur : 1 à 200. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur maximum : 1024. |
search_snapshot
Find a file in a restore point. Portée read. Lecture seule : oui. Destructif : non.
Finds files and folders by name in a restore point (snapshot) of an agent, in all its folders: path, type, size, modification date. Use it when the user knows the name of a file but not its folder ("restore factures.txt"). The name matches a part of the name, any case; when the whole name finds nothing, each word is tried alone. At most 50 paths, with a flag when there are more. Never the content of a file. On a large restore point (hundreds of thousands of files, such as a server or a web site), give the folder to search in (path): a search of the whole restore point can take a minute and stop before its end. Confirm the path with the user, then give it as is to start_restore.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the agent endpoint that made the backup. |
snapshot_id | string | oui | Id of the restore point (list_snapshots) |
name | string | oui | Name or part of a name, any case, for example factures.txt or factures Longueur : 1 à 255. |
path | any | non | The folder to search in, with all its subfolders. Give it on a large restore point, in the form of the backup: /etc/compta on Linux, /C/Users/bob for C:\Users\bob on Windows. Default: / Défaut : "/". |
type | "file" | "dir" | non | Only files, or only folders. |
limit | integer | non | Number of paths, 1 to 50. Défaut : 50. Valeur : 1 à 50. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur maximum : 8192. |
list_snapshot_databases
Databases of a SQL Server restore point. Portée read. Lecture seule : oui. Destructif : non.
The databases of one SQL Server restore point (list_snapshots with backup_kind mssql): name, instance, status, size of the backup, size of the data (the space a restore needs), recovery model and SQL Server version. Never a table, a row or a file of a database. Use the names with preview_mssql_restore.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the agent endpoint. |
snapshot_id | string | oui | Id of the SQL Server restore point (list_snapshots, backup_kind mssql) |
list_vm_snapshots
List the restore points of a VM. Portée read. Lecture seule : oui. Destructif : non.
Restore points of one VM of a Hyper-V host, or of one VM or container (kind: ct) of a Proxmox VE node, newest first: date, consistency (app_consistent, fs_frozen, crash, stopped), bytes of the disks read and bytes added. One restore point per guest and backup run. Never the content of the disks.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
endpoint_id | string (uuid) | oui | Id of the Hyper-V host or Proxmox node (agent endpoint) |
vm_key | any | oui | vm_key of the VM or container, as get_endpoint gives it (hyperv:<endpoint id>:<VM GUID>, proxmox:<endpoint id>:<uuid>) |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
list_vm_restore_targets
Hosts for a VM restore. Portée read. Lecture seule : oui. Destructif : non.
The machines of the same client as the source that can take the restore of a VM or container restore point as a new guest, the source first. Each has unavailable: null when it can, else the reason. Hyper-V: the Windows machines, with their virtual switches and free space (reasons: not a Hyper-V host, configuration version not supported, agent too old, deletion pending, offline). Proxmox: the Linux machines, with node: the Proxmox VE version, the storages with free space and the bridges, for target_storage and bridge of preview_vm_restore (reasons: not a Proxmox VE node, Proxmox VE version out of support, machine version of the VM not supported, agent too old, deletion pending, offline). A machine of another client is never listed.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
source_endpoint_id | string (uuid) | oui | Id of the Hyper-V host or Proxmox node that made the backup. |
snapshot_id | string | oui | Restore point of the VM or container (list_vm_snapshots) |
list_enrollment_tokens
List enrollment tokens. Portée read. Lecture seule : oui. Destructif : non.
Lists the enrollment tokens of the MSP, newest first: client, status (pending, used, revoked, expired), origin (portal or mcp), dates, the machine a used token enrolled. Never the value of a token: it is shown once, at its creation.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | non | Id of the client. |
status | "pending" | "used" | "revoked" | "expired" | non | |
origin | "portal" | "mcp" | non | mcp: made by an AI assistant. |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
deployment_tokens_list
List deployment tokens. Portée read. Lecture seule : oui. Destructif : non.
Lists the deployment tokens of the MSP, newest first. A deployment token enrolls many machines of one client (GPO, Intune, Ansible...). For each: client, status (active, exhausted: its most machines are enrolled, revoked, expired), machines enrolled and the most it may enroll (null: no limit), expiry, tags, allowed networks. Never the value of a token. A person creates and revokes a deployment token in the portal only: no tool does it.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
client_id | string (uuid) | non | Id of the client. |
status | "active" | "exhausted" | "revoked" | "expired" | non | |
limit | integer | non | Number of items, 1 to 100. Défaut : 20. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |
deployment_token_endpoints
List the machines of a deployment token. Portée read. Lecture seule : oui. Destructif : non.
The deployment token (state and counter) and the machines it enrolled, newest first: name, status, enrollment date, source address. Deleted machines too.
Paramètres
| Nom | Type | Obligatoire | Description |
|---|---|---|---|
deployment_token_id | string (uuid) | oui | Id of the deployment token. |
limit | integer | non | Number of items, 1 to 100. Défaut : 50. Valeur : 1 à 100. |
cursor | string | non | next_cursor of the previous answer, for the next page. Longueur : 1 à 512. |