Run deletion compaction also in non-clustered mode

ANTIETCD.md

@@ -0,0 +1,469 @@
+# AntiEtcd
+
+Simplistic miniature etcd replacement based on [TinyRaft](https://git.yourcmc.ru/vitalif/tinyraft/)
+
+- Embeddable
+- REST API only, gRPC is shit and will never be supported
+- [TinyRaft](https://git.yourcmc.ru/vitalif/tinyraft/)-based leader election
+- Websocket-based cluster communication
+- Supports a limited subset of etcd REST APIs
+- With optional persistence
+
+(c) Vitaliy Filippov, 2024
+
+License: Mozilla Public License 2.0 or [VNPL-1.1](https://git.yourcmc.ru/vitalif/vitastor/src/branch/master/VNPL-1.1.txt)
+
+# Contents
+
+- [CLI Usage](#cli-usage)
+- [Options](#options)
+  - [HTTP](#http)
+  - [Persistence](#persistence)
+  - [Clustering](#clustering)
+- [Embedded Usage](#embedded-usage)
+- [About Persistence](#about-persistence)
+- [Supported etcd APIs](#supported-etcd-apis)
+  - [/v3/kv/txn](#v3-kv-txn)
+  - [/v3/kv/put](#v3-kv-put)
+  - [/v3/kv/range](#v3-kv-range)
+  - [/v3/kv/deleterange](#v3-kv-deleterange)
+  - [/v3/lease/grant](#v3-lease-grant)
+  - [/v3/lease/keepalive](#v3-lease-keepalive)
+  - [/v3/lease/revoke or /v3/kv/lease/revoke](#v3-lease-revoke-or-v3-kv-lease-revoke)
+  - [Websocket-based watch APIs](#websocket-based-watch-apis)
+- [HTTP Error Codes](#http-error-codes)
+
+## CLI Usage
+
+```
+npm install antietcd
+
+node_modules/.bin/antietcd \
+    [--cert ssl.crt] [--key ssl.key] [--port 12379] \
+    [--data data.gz] [--persist_interval 500] \
+    [--node_id node1 --cluster_key abcdef --cluster node1=http://localhost:12379,node2=http://localhost:12380,node3=http://localhost:12381]
+    [other options]
+```
+
+Antietcd doesn't background itself, so use systemd or start-stop-daemon to run it as a background service.
+
+## Options
+
+### HTTP
+
+<dl>
+
+<dt>--port 2379</dt>
+<dd>Listen port</dd>
+
+<dt>--cert &lt;cert&gt;</dt>
+<dd>Use TLS with this certificate file (PEM format)</dd>
+
+<dt>--key &lt;key&gt;</dt>
+<dd>Use TLS with this key file (PEM format)</dd>
+
+<dt>--ca &lt;ca&gt;</dt>
+<dd>Use trusted root certificates from this file.
+Specify &lt;ca&gt; = &lt;cert&gt; if your certificate is self-signed.</dd>
+
+<dt>--client_cert_auth 1</dt>
+<dd>Require TLS client certificates signed by <ca> or by default CA to connect.</dd>
+
+<dt>--ws_keepalive_interval 30000</dt>
+<dd>Client websocket ping (keepalive) interval in milliseconds</dd>
+
+</dl>
+
+### Persistence
+
+<dl>
+
+<dt>--data &lt;filename&gt;</dt>
+<dd>Store persistent data in &lt;filename&gt;</dd>
+
+<dt>--persist_interval &lt;milliseconds&gt;</dt>
+<dd>Persist data on disk after this interval, not immediately after change</dd>
+
+<dt>--persist_filter ./filter.js</dt>
+<dd>Use persistence filter from ./filter.js (or a module). <br />
+Persistence filter is a function(cfg) returning function(key, value) ran
+for every change and returning a new value or undefined to skip persistence.</dd>
+
+<dt>--compact_revisions 1000</dt>
+<dd>Number of previous revisions to keep deletion information in memory</dd>
+
+</dl>
+
+### Clustering
+
+<dl>
+
+<dt>--node_id &lt;id></dt>
+<dd>ID of this cluster node</dd>
+
+<dt>--cluster &lt;id1&gt;=&lt;url1&gt;,&lt;id2&gt;=&lt;url2&gt;,...</dt>
+<dd>All other cluster nodes</dd>
+
+<dt>--cluster_key &lt;key&gt;</dt>
+<dd>Shared cluster key for identification</dd>
+
+<dt>--election_timeout 5000</dt>
+<dd>Raft election timeout</dd>
+
+<dt>--heartbeat_timeout 1000</dt>
+<dd>Raft leader heartbeat timeout</dd>
+
+<dt>--wait_quorum_timeout 30000</dt>
+<dd>Timeout for requests to wait for quorum to come up</dd>
+
+<dt>--leader_priority &lt;number&gt;</dt>
+<dd>Raft leader priority for this node (optional)</dd>
+
+<dt>--stale_read 1</dt>
+<dd>Allow to serve reads from followers. Specify 0 to disallow</dd>
+
+<dt>--reconnect_interval 1000</dt>
+<dd>Unavailable peer connection retry interval</dd>
+
+<dt>--dump_timeout 5000</dt>
+<dd>Timeout for dump command in milliseconds</dd>
+
+<dt>--load_timeout 5000</dt>
+<dd>Timeout for load command in milliseconds</dd>
+
+<dt>--forward_timeout 1000</dt>
+<dd>Timeout for forwarding requests from follower to leader in milliseconds</dd>
+
+<dt>--replication_timeout 1000</dt>
+<dd>Timeout for replicating requests from leader to follower in milliseconds</dd>
+
+<dt>--compact_timeout 1000</dt>
+<dd>Timeout for compaction requests from leader to follower in milliseconds</dd>
+
+</dl>
+
+## Embedded Usage
+
+```js
+const AntiEtcd = require('antietcd');
+
+// Configuration may contain all the same options like in CLI, without "--"
+// Except that persist_filter should be a callback (key, value) => newValue
+const srv = new AntiEtcd({ ...configuration });
+
+// Start server
+srv.start();
+
+// Make a local API call in generic style:
+let res = await srv.api('kv_txn'|'kv_range'|'kv_put'|'kv_deleterange'|'lease_grant'|'lease_revoke'|'lease_keepalive', { ...params });
+
+// Or function-style:
+res = await srv.txn(params);
+res = await srv.range(params);
+res = await srv.put(params);
+res = await srv.deleterange(params);
+res = await srv.lease_grant(params);
+res = await srv.lease_revoke(params);
+res = await srv.lease_keepalive(params);
+
+// Error handling:
+try
+{
+    res = await srv.txn(params);
+}
+catch (e)
+{
+    if (e instanceof AntiEtcd.RequestError)
+    {
+        // e.code is HTTP code
+        // e.message is error message
+    }
+}
+
+// Watch API:
+const watch_id = await srv.create_watch(params, (message) => console.log(message));
+await srv.cancel_watch(watch_id);
+
+// Stop server
+srv.stop();
+```
+
+## About Persistence
+
+Persistence is very simple: full database is dumped into JSON, gzipped and saved as file.
+
+By default, it is written and fsynced on disk on every change, but it can be configured
+to dump DB on disk at fixed intervals, for example, at most every 500 ms - of course,
+at expense of slightly reduced crash resiliency (example: `--persist_interval 500`).
+
+You can also specify a filter to exclude some data from persistence by using the option
+`--persist_filter ./filter.js`. Persistence filter code example:
+
+```js
+function example_filter(cfg)
+{
+    // <cfg> contains all command-line options
+    const prefix = cfg.exclude_keys;
+    if (!prefix)
+    {
+        return null;
+    }
+    return (key, value) =>
+    {
+        if (key.substr(0, prefix.length) == prefix)
+        {
+            // Skip all keys with prefix from persistence
+            return undefined;
+        }
+        if (key === '/statistics')
+        {
+            // Return <unneeded_key> from inside value
+            const decoded = JSON.parse(value);
+            return JSON.stringify({ ...decoded, unneeded_key: undefined });
+        }
+        return value;
+    };
+}
+
+module.exports = example_filter;
+```
+
+## Supported etcd APIs
+
+NOTE: `key`, `value` and `range_end` are always encoded in base64, like in original etcd.
+
+Range requests are only supported across "directories" separated by `/`.
+
+It means that in range requests `key` must always end with `/` and `range_end` must always
+end with `0`, and that such request will return a whole subtree of keys.
+
+### /v3/kv/txn
+
+Request:
+
+```ts
+type TxnRequest = {
+  compare?: (
+    { key: string, target: "MOD", mod_revision: number, result?: "LESS" }
+    | { key: string, target: "CREATE", create_revision: number, result?: "LESS" }
+    | { key: string, target: "VERSION", version: number, result?: "LESS" }
+    | { key: string, target: "LEASE", lease: string, result?: "LESS" }
+    | { key: string, target: "VALUE", value: string }
+  )[],
+  success?: (
+    { request_put: PutRequest }
+    | { request_range: RangeRequest }
+    | { request_delete_range: DeleteRangeRequest }
+  )[],
+  failure?: (
+    { request_put: PutRequest }
+    | { request_range: RangeRequest }
+    | { request_delete_range: DeleteRangeRequest }
+  )[],
+  serializable?: boolean,
+}
+```
+
+`serializable` allows to serve read-only requests from follower even if `stale_read` is not enabled.
+
+Response:
+
+```ts
+type TxnResponse = {
+  header: { revision: number },
+  succeeded: boolean,
+  responses: (
+    { response_put: PutResponse }
+    | { response_range: RangeResponse }
+    | { response_delete_range: DeleteRangeResponse }
+  )[],
+}
+```
+
+### /v3/kv/put
+
+Request:
+
+```ts
+type PutRequest = {
+  key: string,
+  value: string,
+  lease?: string,
+}
+```
+
+Other parameters are not supported: prev_kv, ignore_value, ignore_lease.
+
+Response:
+
+```ts
+type PutResponse = {
+  header: { revision: number },
+}
+```
+
+### /v3/kv/range
+
+Request:
+
+```ts
+type RangeRequest = {
+  key: string,
+  range_end?: string,
+  keys_only?: boolean,
+  serializable?: boolean,
+}
+```
+
+`serializable` allows to serve read-only requests from follower even if `stale_read` is not enabled.
+
+Other parameters are not supported: revision, limit, sort_order, sort_target,
+count_only, min_mod_revision, max_mod_revision, min_create_revision, max_create_revision.
+
+Response:
+
+```ts
+type RangeResponse = {
+  header: { revision: number },
+  kvs: { key: string }[] | {
+    key: string,
+    value: string,
+    lease?: string,
+    mod_revision: number,
+  }[],
+}
+```
+
+### /v3/kv/deleterange
+
+Request:
+
+```ts
+type DeleteRangeRequest = {
+  key: string,
+  range_end?: string,
+}
+```
+
+Other parameters are not supported: prev_kv.
+
+Response:
+
+```ts
+type DeleteRangeResponse = {
+  header: { revision: number },
+  // number of deleted keys
+  deleted: number,
+}
+```
+
+### /v3/lease/grant
+
+Request:
+
+```ts
+type LeaseGrantRequest = {
+  ID?: string,
+  TTL: number,
+}
+```
+
+Response:
+
+```ts
+type LeaseGrantResponse = {
+  header: { revision: number },
+  ID: string,
+  TTL: number,
+}
+```
+
+### /v3/lease/keepalive
+
+Request:
+
+```ts
+type LeaseKeepaliveRequest = {
+  ID: string,
+}
+```
+
+Response:
+
+```ts
+type LeaseKeepaliveResponse = {
+  result: {
+    header: { revision: number },
+    ID: string,
+    TTL: number,
+  }
+}
+```
+
+### /v3/lease/revoke or /v3/kv/lease/revoke
+
+Request:
+
+```ts
+type LeaseRevokeRequest = {
+  ID: string,
+}
+```
+
+Response:
+
+```ts
+type LeaseRevokeResponse = {
+  header: { revision: number },
+}
+```
+
+### Websocket-based watch APIs
+
+Client-to-server message format:
+
+```ts
+type ClientMessage =
+  { create_request: {
+    key: string,
+    range_end?: string,
+    start_revision?: number,
+    watch_id?: string,
+  } }
+  | { cancel_request: {
+    watch_id: string,
+  } }
+  | { progress_request: {} }
+```
+
+Server-to-client message format:
+
+```ts
+type ServerMessage = {
+  result: {
+    header: { revision: number },
+    watch_id: string,
+    created?: boolean,
+    canceled?: boolean,
+    compact_revision?: number,
+    events?: {
+      type: 'PUT'|'DELETE',
+      kv: {
+        key: string,
+        value: string,
+        lease?: string,
+        mod_revision: number,
+      },
+    }[],
+  }
+} | { error: 'bad-json' } | { error: 'empty-message' }
+```
+
+### HTTP Error Codes
+
+- 400 for invalid requests
+- 404 for unsupported API / URL not found
+- 405 for non-POST request method
+- 501 for unsupported API feature - non-directory range queries and so on
+- 502 for server is stopping
+- 503 for quorum-related errors - quorum not available and so on

antietcd-app.js

@@ -48,6 +48,8 @@ Persistence:
     Use persistence filter from ./filter.js (or a module).
     Persistence filter is a function(cfg) returning function(key, value) ran
     for every change and returning a new value or undefined to skip persistence.
+--compact_revisions 1000
+    Number of previous revisions to keep deletion information in memory
 
 Clustering:
 
@@ -77,8 +79,6 @@ Clustering:
     Timeout for forwarding requests from follower to leader in milliseconds
 --replication_timeout 1000
     Timeout for replicating requests from leader to follower in milliseconds
---compact_revisions 1000
-    Number of previous revisions to keep deletion information in memory
 --compact_timeout 1000
     Timeout for compaction requests from leader to follower in milliseconds
 `;

antietcd.js

@@ -29,6 +29,7 @@ class AntiEtcd extends EventEmitter
         this.stopped = false;
         this.inflight = 0;
         this.wait_inflight = [];
+        this.api_watches = {};
     }
 
     async start()
@@ -124,6 +125,16 @@ class AntiEtcd extends EventEmitter
                 }));
             });
         }
+        if (!this.cluster)
+        {
+            // Run deletion compaction without followers
+            const mod_revision = this.antietcd.etctree.mod_revision;
+            if (mod_revision - this.antietcd.etctree.compact_revision > (this.cfg.compact_revisions||1000)*2)
+            {
+                const revision = mod_revision - (this.cfg.compact_revisions||1000);
+                this.antietcd.etctree.compact(revision);
+            }
+        }
     }
 
     _handleRequest(req, res)
@@ -393,6 +404,30 @@ class AntiEtcd extends EventEmitter
         return await this.api('lease_keepalive', params);
     }
 
+    // public watch API
+    async create_watch(params, callback)
+    {
+        const watch = this.etctree.api_create_watch({ ...params, watch_id: null }, callback);
+        if (!watch.created)
+        {
+            throw new RequestError(400, 'Requested watch revision is compacted', { compact_revision: watch.compact_revision });
+        }
+        const watch_id = params.watch_id || watch.watch_id;
+        this.api_watches[watch_id] = watch.watch_id;
+        return watch_id;
+    }
+
+    async cancel_watch(watch_id)
+    {
+        const mapped_id = this.api_watches[watch_id];
+        if (!mapped_id)
+        {
+            throw new RequestError(400, 'Watch not found');
+        }
+        this.etctree.api_cancel_watch({ watch_id: mapped_id });
+        delete this.api_watches[watch_id];
+    }
+
     // internal handlers
     async _handle_kv_txn(data)
     {

common.js

@@ -1,9 +1,10 @@
 class RequestError
 {
-    constructor(code, text)
+    constructor(code, text, details)
     {
         this.code = code;
         this.message = text;
+        this.details = details;
     }
 }