agent+server: P2R-11 pre/post hook execution for backup jobs

Agent: new runner.BackupHooks struct + runHook helper invoked via
/bin/sh -c (cmd.exe /C on Windows). pre_hook non-zero exit aborts
the backup; post_hook always runs with RM_JOB_STATUS=succeeded|failed
in env. Output streamed as 'hook(<phase>): …' log.stream lines.
Hooks only run for kind=backup (other kinds skip both phases).

Server: resolveBackupHooks resolves group → host default → empty,
decrypts via crypto.AEAD with per-slot ad bytes, plumbs plaintext
into CommandRunPayload for both schedule.fire and per-group
Run-now dispatch sites. Decrypt failures degrade silently to no
hook so a malformed blob can't poison every backup.
This commit is contained in:
2026-05-04 10:57:28 +01:00
parent 18b0bf976d
commit 7b1990cf11
11 changed files with 379 additions and 52 deletions
+75
View File
@@ -0,0 +1,75 @@
// hooks_resolve.go — server-side resolution of pre/post hooks for a
// backup dispatch (P2R-11). The agent receives plaintext hook bodies
// in CommandRunPayload; this file is where the AEAD blob on the
// source group (or the host's default) gets decrypted into the
// strings the wire payload carries.
//
// Resolution order:
// 1. source_group.<phase>_hook (per-group override)
// 2. host.<phase>_hook_default (host-wide default)
// 3. "" (no hook → agent skips that phase)
//
// Decrypt errors are logged and treated as "no hook configured" so
// a malformed blob can't poison every backup. The audit trail
// captures the underlying state regardless.
package http
import (
"log/slog"
"gitea.dcglab.co.uk/steve/restic-manager/internal/store"
)
// resolveBackupHooks returns the (pre, post) plaintext hook strings
// the agent should run around the backup. Both are empty when no
// hook is configured at either level.
func (s *Server) resolveBackupHooks(host *store.Host, g *store.SourceGroup) (pre, post string) {
if s.deps.AEAD == nil {
return "", ""
}
pre = s.decryptHookOrFallback(g.PreHook, host.PreHookDefault, host.ID, "pre")
post = s.decryptHookOrFallback(g.PostHook, host.PostHookDefault, host.ID, "post")
return pre, post
}
// decryptHookOrFallback returns the per-group hook decrypted, or
// (when that's empty) the host default decrypted, or "" if neither
// is configured. Decrypt failures log and degrade to empty.
func (s *Server) decryptHookOrFallback(group, hostDefault, hostID, phase string) string {
tryDecrypt := func(blob, slot string) (string, bool) {
if blob == "" {
return "", false
}
plain, err := s.deps.AEAD.Decrypt(blob, []byte("hook:"+hostID+":"+slot+":"+phase))
if err != nil {
slog.Error("decrypt hook", "host_id", hostID, "phase", phase, "slot", slot, "err", err)
return "", false
}
return string(plain), true
}
if v, ok := tryDecrypt(group, "group"); ok {
return v
}
if v, ok := tryDecrypt(hostDefault, "host"); ok {
return v
}
return ""
}
// EncryptHookForGroup encrypts a hook body for storage on a source
// group. Caller passes the plaintext from a UI form; an empty body
// returns "" so the store persists NULL (cleared).
func (s *Server) EncryptHookForGroup(hostID, phase, body string) (string, error) {
if body == "" {
return "", nil
}
return s.deps.AEAD.Encrypt([]byte(body), []byte("hook:"+hostID+":group:"+phase))
}
// EncryptHookForHost is the host-default twin of EncryptHookForGroup.
func (s *Server) EncryptHookForHost(hostID, phase, body string) (string, error) {
if body == "" {
return "", nil
}
return s.deps.AEAD.Encrypt([]byte(body), []byte("hook:"+hostID+":host:"+phase))
}
+9
View File
@@ -79,6 +79,13 @@ func (s *Server) handleRunSourceGroup(w stdhttp.ResponseWriter, r *stdhttp.Reque
return
}
// Resolve hooks (group → host default → empty). Best-effort host
// lookup; failure proceeds with no hook rather than block the run.
var preHook, postHook string
if host, herr := s.deps.Store.GetHost(r.Context(), hostID); herr == nil {
preHook, postHook = s.resolveBackupHooks(host, g)
}
// Backup invocations don't consume RetentionPolicy — that lives on
// forget. Sending the resolved set here would just be dead weight.
res, status, code, msg := s.dispatchJobWithPayload(r.Context(), user, hostID, api.JobBackup,
@@ -88,6 +95,8 @@ func (s *Server) handleRunSourceGroup(w stdhttp.ResponseWriter, r *stdhttp.Reque
Tag: g.Name,
BandwidthUpKBps: upOverride,
BandwidthDownKBps: downOverride,
PreHook: preHook,
PostHook: postHook,
})
if code != "" {
s.runGroupError(w, r, status, code, msg)
+14
View File
@@ -192,6 +192,18 @@ func (s *Server) dispatchBackupForGroupCore(ctx context.Context, conn *ws.Conn,
"schedule_id", scheduleID, "group", g.Name, "err", err)
return "", err
}
// Resolve pre/post hooks (group → host default → empty) so they
// ride on the backup payload as plaintext. The host lookup is
// cheap; failure here is non-fatal (we proceed without hooks
// rather than block the backup).
var preHook, postHook string
if host, herr := s.deps.Store.GetHost(ctx, hostID); herr == nil {
preHook, postHook = s.resolveBackupHooks(host, g)
} else {
slog.Warn("schedule.fire: load host for hook resolve",
"host_id", hostID, "err", herr)
}
// Backup ignores RetentionPolicy — the forget cadence lives on
// host_repo_maintenance and is driven by the server-side ticker
// (P2R-06). Don't ship the field on backup dispatches.
@@ -201,6 +213,8 @@ func (s *Server) dispatchBackupForGroupCore(ctx context.Context, conn *ws.Conn,
Includes: g.Includes,
Excludes: g.Excludes,
Tag: g.Name,
PreHook: preHook,
PostHook: postHook,
})
if err != nil {
slog.Warn("schedule.fire: marshal command.run",