Files
stalwart-migrator/internal/stalwartapi/management.go
T
jcoffey-dev 7e04351b0f Add cutover; drop rollback in favour of operator-provided recovery
Two changes that arrived together: the cutover phase (ARCHITECTURE.md 4.5)
is implemented, and the rollback phase is deleted. Recovery from a failed
migration is now explicitly the operator's own snapshot or backup, and out
of scope for this tool.

internal/cutover implements 4.5 as seven checkpointed steps: verify the
staged binary's version, install it, preserve and rewrite the service
definition, reload, start, wait for a healthy JMAP session, recalculate
quotas.

The unit is rewritten in place rather than generated from a template. An
operator's unit carries hardening options, limits and dependencies this
tool has no business having an opinion about, and regenerating it would
silently drop them. It repoints ExecStart (preserving systemd's -@:+!
prefix characters and every argument after the executable), updates
--config, and strips recovery-mode Environment lines - leaving
STALWART_RECOVERY_MODE=1 set would recovery-boot the service on every
restart, forever. It refuses on a unit with no ExecStart, and on an
Environment line mixing a recovery variable with others: a line it only
partly understands is one it must not edit.

Quota recalculation is the one step allowed to fail without failing the
phase. Its wire format is grounded in Stalwart's x:Task schema reference -
Task/set creating one AccountMaintenance per account with maintenanceType
recalculateQuota - but the upgrade guide only documents the WebUI path, so
two details remain inferred and are called out in stalwartapi/task.go:
whether the schema's "read-only" annotation on accountId/maintenanceType
means "immutable after creation", and whether a finished task simply leaves
the queue (TaskStatus documents Pending/Retry/Failed with no success
state). Warning rather than failing is the honest response to that
uncertainty, and stale counters are an accounting problem next to calling
for a restore of a machine that is otherwise migrated and serving mail.

Docker deployments are refused outright: cutting a container over means
pulling an image and recreating it, not swapping a binary.

On removing rollback. The implementation worked and was tested, and it was
removed because restoring bytes correctly is not the hard part. It copied
file contents and permissions and verified every restored file against a
manifest - and did not preserve ownership. Run as root, as this tool
requires, it would have produced a byte-perfect, checksum-verified,
root-owned data directory that Stalwart, running as its own user, could not
open, and it would have reported success. The PostgreSQL path was worse:
pg_dump without --clean emits CREATE TABLE + COPY, which fails replaying
into a database whose tables still exist, and the ON_ERROR_STOP=1 added so
a half-applied restore couldn't be reported as success turned that into a
hard failure. None of it had ever run against a real server. A filesystem
snapshot has none of these failure modes, because it never lost the
metadata to begin with.

So cutover's gate is no longer rollback.CanRollBack but an explicit
RecoveryPointConfirmed acknowledgement. That is an assertion, not a check -
this tool cannot verify someone else's snapshot - and its only value is
that nobody migrates a production mail server having never been asked the
question. Two consequences are accepted deliberately: restoring any
pre-migration recovery point discards mail delivered since, and a failed
migration now stops and reports rather than undoing itself.

What the tool still does to make a manual restore easier: the old binary is
preserved and never deleted, the original service definition is preserved
before the rewrite, the settings and principals dumps stay on disk, and
every artifact path and checksum stays in the checkpoint where `status
<run-id>` can print it.

Also removed: the `confirm` command stub and RollbackWindowClosed, whose
only purpose was closing a rollback window that no longer exists, and
checkpoint.PhaseRollback. Old state.json files still load - JSON ignores
the now-unknown field.

Still open, and recorded in 8: cutover ignores systemd drop-ins, so an
ExecStart or Environment override in stalwart.service.d/*.conf is invisible
to the rewrite - including the recovery variable it exists to strip;
nothing prevents concurrent runs on the same run-id; and nothing in this
repo has ever run against a real Stalwart, real systemd, or a real store.
2026-08-23 17:52:47 -07:00

210 lines
7.2 KiB
Go

package stalwartapi
import (
"bytes"
"context"
"encoding/json"
"fmt"
"io"
"net/http"
"sort"
"strings"
)
// managementCapabilities are the JMAP capability URNs Stalwart requires for
// its management object calls (x:Account/*): standard JMAP core plus its
// own urn:stalwart:jmap extension. Confirmed against
// docs/ref/object/account.md and crates/jmap-proto/src/request/capability.rs
// in stalwartlabs/stalwart.
var managementCapabilities = []string{"urn:ietf:params:jmap:core", "urn:stalwart:jmap"}
type jmapRequest struct {
Using []string `json:"using"`
MethodCalls []any `json:"methodCalls"`
}
type jmapRawResponse struct {
MethodResponses []json.RawMessage `json:"methodResponses"`
}
// methodResponse is one [name, args, callId] triple from a JMAP response,
// per RFC 8620 §3.2 - Stalwart's management API follows the same envelope
// shape as its regular JMAP methods.
type methodResponse struct {
Name string
Args json.RawMessage
CallID string
}
// call POSTs one JMAP-style request to the management API (Stalwart's /api
// endpoint - see docs/ref/object/account.md, which is distinct from /jmap)
// using this Client's own credentials, and returns its parsed method
// responses in order.
func (c *Client) call(ctx context.Context, using []string, methodCalls []any) ([]methodResponse, error) {
return c.callAs(ctx, c.Username, c.Password, strings.TrimRight(c.BaseURL, "/")+"/api", using, methodCalls)
}
// callAs is call's underlying primitive: it accepts an explicit
// username/password/URL rather than always using this Client's own
// credentials and the management endpoint. MailboxSnapshot uses this to
// call standard JMAP methods (not Stalwart's x: management objects) against
// the URL a JMAP session document says to use, authenticated as an
// impersonated identity rather than this Client's own.
func (c *Client) callAs(ctx context.Context, username, password, url string, using []string, methodCalls []any) ([]methodResponse, error) {
reqBody, err := json.Marshal(jmapRequest{Using: using, MethodCalls: methodCalls})
if err != nil {
return nil, fmt.Errorf("stalwartapi: encode request: %w", err)
}
req, err := http.NewRequestWithContext(ctx, http.MethodPost, url, bytes.NewReader(reqBody))
if err != nil {
return nil, err
}
req.Header.Set("Content-Type", "application/json")
req.SetBasicAuth(username, password)
resp, err := c.httpClient().Do(req)
if err != nil {
return nil, fmt.Errorf("stalwartapi: call %s: %w", url, err)
}
defer resp.Body.Close()
if resp.StatusCode != http.StatusOK {
body, _ := io.ReadAll(io.LimitReader(resp.Body, 4096))
return nil, fmt.Errorf("stalwartapi: %s returned %s: %s", url, resp.Status, strings.TrimSpace(string(body)))
}
var raw jmapRawResponse
if err := json.NewDecoder(resp.Body).Decode(&raw); err != nil {
return nil, fmt.Errorf("stalwartapi: decode response from %s: %w", url, err)
}
responses := make([]methodResponse, 0, len(raw.MethodResponses))
for _, r := range raw.MethodResponses {
var triple [3]json.RawMessage
if err := json.Unmarshal(r, &triple); err != nil {
return nil, fmt.Errorf("stalwartapi: parse method response envelope: %w", err)
}
var name, callID string
if err := json.Unmarshal(triple[0], &name); err != nil {
return nil, fmt.Errorf("stalwartapi: parse method response name: %w", err)
}
if err := json.Unmarshal(triple[2], &callID); err != nil {
return nil, fmt.Errorf("stalwartapi: parse method response call id: %w", err)
}
responses = append(responses, methodResponse{Name: name, Args: triple[1], CallID: callID})
}
return responses, nil
}
// account is the subset of x:Account/get's response fields this tool needs,
// confirmed against Stalwart's own docs/ref/object/account.md (whose
// stalwart-cli example is `query Account --fields id,name,domainId,usedDiskQuota`).
type account struct {
ID string `json:"id"`
Name string `json:"name"`
DomainID string `json:"domainId"`
}
// AccountSnapshot enumerates every account on the instance via Stalwart's
// management API - x:Account/query to list ids, then x:Account/get to fetch
// their name/domainId - and returns the account count, set of domains in
// use, and (via MailboxSnapshot, per account) every mailbox's message
// count. This is what preflight's snapshot (ARCHITECTURE.md §4.1) and
// validate's directory-integrity and content-integrity checks (§4.7)
// compare before and after migration - the latter being the actual
// no-data-loss guarantee.
//
// A per-account mailbox-count failure (most likely: this Client's Username
// lacks the `impersonate` permission MailboxSnapshot depends on) does not
// fail the whole snapshot - the account/domain enumeration above is already
// useful on its own, and one account's failure shouldn't hide a working
// result for every other account. Instead it's recorded in
// Snapshot.MailboxErrors, keyed by account email, so callers can report
// exactly what's missing rather than silently treating an unreachable
// account's mailboxes as having zero messages.
func (c *Client) AccountSnapshot(ctx context.Context) (*Snapshot, error) {
ids, err := c.AccountIDs(ctx)
if err != nil {
return nil, err
}
if len(ids) == 0 {
return &Snapshot{}, nil
}
getResp, err := c.call(ctx, managementCapabilities, []any{
[]any{"x:Account/get", map[string]any{"ids": ids, "properties": []string{"id", "name", "domainId"}}, "g"},
})
if err != nil {
return nil, fmt.Errorf("stalwartapi: Account/get: %w", err)
}
accounts, err := accountGetList(getResp)
if err != nil {
return nil, err
}
mailboxCounts := map[string][]MailboxCount{}
mailboxErrors := map[string]string{}
for _, a := range accounts {
if a.Name == "" {
continue // no login/email to impersonate against
}
counts, err := c.MailboxSnapshot(ctx, a.Name)
if err != nil {
mailboxErrors[a.Name] = err.Error()
continue
}
mailboxCounts[a.Name] = counts
}
domainSet := map[string]bool{}
for _, a := range accounts {
if a.DomainID != "" {
domainSet[a.DomainID] = true
}
}
domains := make([]string, 0, len(domainSet))
for d := range domainSet {
domains = append(domains, d)
}
sort.Strings(domains)
return &Snapshot{
AccountCount: len(accounts),
Domains: domains,
MailboxCounts: mailboxCounts,
MailboxErrors: mailboxErrors,
}, nil
}
func accountQueryIDs(responses []methodResponse) ([]string, error) {
if len(responses) == 0 {
return nil, fmt.Errorf("stalwartapi: Account/query returned no method responses")
}
r := responses[0]
if r.Name == "error" {
return nil, fmt.Errorf("stalwartapi: Account/query error: %s", r.Args)
}
var result struct {
IDs []string `json:"ids"`
}
if err := json.Unmarshal(r.Args, &result); err != nil {
return nil, fmt.Errorf("stalwartapi: parse Account/query response: %w", err)
}
return result.IDs, nil
}
func accountGetList(responses []methodResponse) ([]account, error) {
if len(responses) == 0 {
return nil, fmt.Errorf("stalwartapi: Account/get returned no method responses")
}
r := responses[0]
if r.Name == "error" {
return nil, fmt.Errorf("stalwartapi: Account/get error: %s", r.Args)
}
var result struct {
List []account `json:"list"`
}
if err := json.Unmarshal(r.Args, &result); err != nil {
return nil, fmt.Errorf("stalwartapi: parse Account/get response: %w", err)
}
return result.List, nil
}