404e919494
refactor(dsrepl): make shard allocator more robust and consistent
...
Co-Authored-By: Thales Macedo Garitezi <thalesmg@gmail.com>
2024-03-20 13:20:24 +01:00
0e18bd6e80
fix(dsrepl): increase replication site id bitsize back
...
In order to minimize chances of site id collision to practically zero.
2024-03-20 13:20:24 +01:00
ac9700dd28
fix(dsrepl): split shard allocator into a separate module
2024-03-20 13:20:23 +01:00
1b647035d0
chore(dsrepl): make dialyzer a bit happier
2024-03-20 13:20:23 +01:00
611b3f0e07
feat(dsrepl): use more straightforward way to drop ra shards
2024-03-20 13:20:23 +01:00
74881e8706
feat(dsrepl): make storage layer unaware of granularity of time
...
Storage also becomes a bit more pure, depending on the upper layer to
provide the timestamps, which also makes it possible to handle more
operations idempotently.
2024-03-20 13:20:23 +01:00
3cb36a5619
feat(ds-lts): extract timestamp from storage key itself
...
1. This avoids the need to deserialize the message to get the timestamp.
2. It also makes possible to decouple the storage key timestamp from the
message timestamp, which might be useful for replication purposes.
2024-03-19 20:21:56 +01:00
5cc0246351
feat(dsrepl): allow to tune select ra options
2024-03-19 20:21:55 +01:00
54b5adf868
feat(dsrepl): allocate shards predictably
...
To ensure strictly optimal and fair shard allocation across
cluster. Before this commit it was quite easy to end up with
an allocation significantly skewed towards some node, because
of the nature of randomness and relatively small number of
shards.
2024-03-19 20:21:55 +01:00
887e151be5
fix(dsrepl): handle errors gracefully in shard egress process
...
Also add cooldown on timeout / unavailability.
2024-03-19 20:21:53 +01:00
e16aee99b4
chore(dsrepl): clarify how to perform leadership transfer in runtime
2024-03-19 20:21:18 +01:00
00d509f27b
feat(dsrepl): prefer local replica in read path
...
To optimize out any unnecessary RPCs. Given the load should be
smoothed evenly across the cluster, choosing non-leader node is
not a priority.
2024-03-19 20:11:42 +01:00
19305c223c
fix(dsrepl): require majority for replication-related tables
2024-03-19 20:11:42 +01:00
f89909f60c
fix(dsrepl): tolerate trigger election timeouts for existing servers
2024-03-19 20:11:42 +01:00
3b59cf2ebf
feat(dsrepl): move shard allocation to separate process
...
That starts shard and egress processes only when shards are fully
allocated.
2024-03-19 20:11:41 +01:00
4dafbf21f6
fix(dsrepl): make db + shard part of machine state
...
It doesn't feel right, but right now is the easiest way to have it
in the scope of `apply/3`, because `init/1` doesn't actually invoked
for ra machines recovered from the existing log / snapshot.
2024-03-19 20:11:41 +01:00
d19128ed65
feat(dsrepl): cache shard metadata in persistent terms
2024-03-19 20:11:41 +01:00
e6c2c2fb07
feat(dsrepl): manage generations / db config through ra machine
2024-03-19 20:11:39 +01:00
5e94bdb932
feat(dsrepl): allocate shards once predefined number of sites online
...
Before this commit the most likely shard allocation outcome was
that all shard are allocated to just one node.
2024-03-19 20:11:03 +01:00
be793e4735
fix(dsrepl): reassign timestamp at the time of submission
...
This is needed to ensure total message order for a shard, and
guarantee that no messages will be written "in the past". which
may break replay consistency.
2024-03-19 20:11:01 +01:00
146f082fdc
feat(dsrepl): implement raft-based replication
...
Still very rough but mostly working.
2024-03-19 20:09:44 +01:00
f2dc940436
Merge remote-tracking branch 'upstream/release-56' into 0319-sync-release56
2024-03-19 15:20:08 +01:00
2ebc8dcc55
fix(ds): use `infinity` timeout when storing batches
2024-03-14 10:17:18 -03:00
6af01b916e
feat(ds): implement `get_delete_streams`, `make_delete_iterator` and `delete_next` callbacks for builtin storage
...
Part of https://emqx.atlassian.net/browse/EMQX-11841
2024-03-08 09:56:46 -03:00
09905d78cd
chore(ds): make error handling slightly simpler
...
Co-Authored-By: Thales Macedo Garitezi <thalesmg@gmail.com>
2024-03-07 12:59:57 +01:00
b39c710ec2
fix(ds): tidy up few typespecs
2024-03-07 12:59:57 +01:00
2146d9e1fe
feat(ds): introduce error classes in critical API functions
...
For now, only recoverable / unrecoverable errors are introduced.
2024-03-07 12:59:57 +01:00
5d87d400f4
feat(ds): add atomic store API
...
Part of https://emqx.atlassian.net/browse/EMQX-11841
2024-03-06 15:24:14 -03:00
06334798a5
fix(ds): fix `drop_generation` typespec
...
This typespec fix will be used downstream by other backends.
2024-03-04 14:15:59 -03:00
b706caf294
feat(ds): export types
2024-02-29 14:27:18 +03:00
d5ae0e5c53
feat(ds): update delete/count interface
2024-02-28 22:51:24 +03:00
b010d34640
chore(ds): add delete callbacks
2024-02-26 17:35:13 +03:00
46877e979b
chore: update copyright-year
2024-02-23 08:21:06 +01:00
d469f4158e
chore: bump app vsns
2024-02-20 16:53:57 -03:00
8cfb22f0b8
fix(ds): Retry getting the shard leader
2024-02-16 12:42:48 +01:00
280fcd8c52
Merge pull request #12437 from ieQu1/dev/optimize_make_filter
...
Optimize emqx_ds_bitmask_keymapper:make_filter function.
2024-02-05 17:32:28 +01:00
4665837cf0
fix(ds): Apply review remarks
2024-02-05 16:52:06 +01:00
c7888ad1f1
Merge pull request #12475 from ieQu1/dev/lean-stream
...
Use a more compact data structure to represent streams
2024-02-05 13:55:24 +01:00
698ba3f271
fix(ds): Optimize emqx_ds_bitmask_keymapper:make_filter
...
This optimization makes idle polling faster
2024-02-05 10:54:19 +01:00
8edbec5929
refactor(ds): Clarify the language used in ds_bitmapper
2024-02-05 10:54:18 +01:00
75023f2ca3
Merge pull request #12442 from ieQu1/dev/ds-license-apache
...
Relicense apps/emqx_durable_storage under Apache 2.0
2024-02-05 10:16:52 +01:00
2e56810ea2
refactor(ds): Use a simple improper list to represent the streams
2024-02-03 21:15:54 +01:00
3d2ac97c61
feat(ds): Add a CLI interface to inspect status of DS databases
2024-02-02 10:11:01 +01:00
b50d6bf1fd
chore(ds): Change the license to Apache 2.0
...
Due to technicalities parts of the original code were licensed under
BSL.
In preparations for the public release of the feature, the license has
been changed to Apache 2.0
2024-02-01 00:10:48 +01:00
d51deac222
fix(ds): use configured data dir for site storage
2024-01-31 15:42:26 -03:00
2479e1189a
fix(ds): Remove unused module
2024-01-29 00:36:13 +01:00
eec56b0d6b
fix(sessds): Improve comments
2024-01-26 17:49:33 +01:00
8e8d3af096
fix(sessds): Refactor emqx_persistent_session_ds to use CRUD module
2024-01-26 17:49:33 +01:00
8e31afe6c2
fix(ds): don't make data dir part of the schema
...
The data directory was ending up being persisted in the database schema. This led to
issues when opening the DB on different nodes.
2024-01-25 14:44:06 -03:00
305a54f646
chore(ds): Update BPAPI version
2024-01-24 19:33:30 +01:00
eee221f1d0
feat(ds): Make egress batching configurable
2024-01-24 19:33:30 +01:00
137535a821
feat(ds): Introduce egress process for the builtin backend
2024-01-24 19:33:30 +01:00
9b7df302e8
fix(ds): Cache database metadata in RAM
2024-01-24 18:43:48 +01:00
1eb47d0c16
perf(ds): inherit only LTS paths containing wildcards when adding a new generation
...
Fixes https://github.com/emqx/emqx/pull/12338#discussion_r1462139499
2024-01-23 09:20:28 -03:00
d122340c13
Merge pull request #12338 from thalesmg/ds-message-gc-20240115
...
feat(ds): add message GC
2024-01-22 16:57:26 -03:00
db710c4be5
feat(lts): inherit previous generation's lts when possible
2024-01-22 14:53:17 -03:00
75b08b525b
feat(ds): add `list_generations` and `drop_generation` APIs
2024-01-22 14:53:17 -03:00
57074015c6
feat(ds): allow customizing the data directory
...
The storage expectations for the RocksDB DB may be different from our usual data
directory. Also, it may consume a lot more storage than other data.
This allows customizing the data directory for the builtin DS storage backend.
Note: if the cluster was already initialized using a directory path, changing that config
will have no effect. This path is currently persisted in mnesia and used when reopening
the DB.
2024-01-19 13:07:24 -03:00
7763b0fd34
chore: bump app vsns
2024-01-08 17:42:56 -03:00
756980837c
Merge branch 'release-54' into sync-r54-m-20240108
2024-01-08 17:39:42 -03:00
caf461fdf6
fix(ds): Don't start the supervision tree when feature is not in use
2024-01-08 17:59:19 +01:00
9e0f3ce53b
feat(ds): restore original add_generation/update_db_config callback semantics
2023-12-29 13:12:15 +03:00
23ded313ec
chore: update app versions
2023-12-22 15:29:22 +01:00
20543d55ef
chore: bump app vsn
2023-12-22 13:13:30 +01:00
322b7bb7d2
chore: bump app vsn
2023-12-22 13:00:37 +01:00
612be8e280
fix(ds): make dialyzer happy
2023-12-20 22:09:03 +08:00
ed38ca67d5
fix(ds): Unified the names of the `add_generation` API
2023-12-20 18:58:03 +08:00
31060733a5
feat(ds): add an API for making new generations
2023-12-15 16:08:52 +08:00
c22a3686ae
fix(emqx_durable_storage): fix type specs
2023-12-14 17:35:32 +01:00
c1f2287b86
Merge remote-tracking branch 'origin/release-54'
2023-12-14 15:26:49 +01:00
dbc8141930
Merge pull request #12125 from thalesmg/ds-cache-m-20231206
...
chore(ds): return DS key from `next` and add `update_iterator` callback
2023-12-14 10:23:25 -03:00
50f4aba5cd
fix(dialyzer): batch 3
2023-12-09 15:50:09 +01:00
66d043becd
feat(ds): introduce `update_iterator` callback
2023-12-08 15:04:18 -03:00
ddbb8560fa
fix(dialyzer): batch 2
2023-12-08 17:59:55 +01:00
2a6d72878f
chore(ds): return DS message key along with batch
2023-12-07 11:36:08 -03:00
c43b3eb535
fix(sessds): Add debug logs for the session garbage collection
2023-12-06 15:37:23 +01:00
130a5a5442
fix(ds): pass topics to `emqx_topic:words/1` before feeding LTS tree
...
So that empty levels in topics will be properly mapped into `''` atoms.
2023-12-04 13:39:01 +03:00
0ae618d010
fix(ds): Use emqx_rpc for calls that work with large binaries
2023-12-01 08:27:05 +01:00
9fd2fa95a8
chore: bump apps versions
2023-11-30 20:01:12 +01:00
05e47254e2
fix(ds): Fixes related to the shards table
2023-11-28 12:59:09 +01:00
e8d18b0e09
Merge pull request #12014 from ieQu1/ds-qos0
...
Support QoS0 messages in persistent_session_ds
2023-11-23 23:43:18 +01:00
449bafc27e
fix(ds): LTS trie handles empty topic levels
2023-11-23 22:16:38 +01:00
c5bb86db67
feat(ds): Support QoS 0
2023-11-23 22:16:38 +01:00
1da9ec1d7d
Merge pull request #12015 from ieQu1/ds-lts-dont-store-message
...
feat(ds): Don't store #message record in the DB
2023-11-23 18:56:45 +01:00
c69b82455e
feat(ds): Don't store #message record in the DB
2023-11-23 17:56:35 +01:00
29d42506d5
fix(ds): impose more flexible types for ds implementations
2023-11-23 17:16:11 +03:00
a1b9a14fa1
feat(ds): allow fdb implementation for durable storage
2023-11-22 14:49:38 +03:00
3165b4f645
fix(ds): Abort application startup when rocksdb is not avialable
2023-11-21 20:27:07 +01:00
3d823beb11
fix(ds): Apply review remarks
2023-11-21 19:37:01 +01:00
f5c71e8068
refactor(ds): Add a wrapper to the store batch API
2023-11-21 19:37:01 +01:00
2a1f7d946a
feat(ds): Shard messages by publisher client ID
2023-11-21 19:37:01 +01:00
62542e5844
feat(ds): Metadata storage for the replication layer
2023-11-21 19:37:01 +01:00
28a577ad09
chore: bump apps versions
2023-11-14 11:02:26 +01:00
088de9476c
fix(ds): use integer for tag values
...
Follow up to https://github.com/emqx/emqx/pull/11906#discussion_r1389115973
2023-11-10 11:26:32 -03:00
2ade6da7a6
fix(ds): Remove redundand type
2023-11-10 12:53:30 +01:00
feef23fc08
feat(ds): Pass store_batch through RPC
2023-11-10 12:16:39 +01:00
8dc8237331
feat(ds): Introduce emqx_ds behavior
2023-11-10 12:16:39 +01:00
42536bb5f7
chore(ds): avoid using records in persistence / rpc
...
Fixes https://emqx.atlassian.net/browse/EMQX-11279
2023-11-09 12:00:25 -03:00
a1cdbaa76d
fix(ds): Address code review remarks
2023-11-08 04:31:27 +01:00
7cb0322856
fix(emqx): Move bpapi and emqx message record to emqx_utils app
2023-11-08 03:04:19 +01:00
74cb43f8b1
fix(ds): Add unique ID to the key
2023-11-08 03:04:19 +01:00
54951c273f
feat(ds): mix safe cutoff into keymapper filter during iteration
2023-11-08 03:04:19 +01:00
7a94db25c3
fix(ds): don't iterate over incomplete epoch in bitmask lts storage
2023-11-08 03:04:19 +01:00
38b7f516bc
fix(ds): Change the directory where message database is stored
2023-11-08 03:04:19 +01:00
8e5dda40be
fix(ds): Fix static checks
2023-11-08 03:04:19 +01:00
99329e1243
refactor(ds): Address review remarks
2023-11-08 03:04:19 +01:00
2de79dd9ac
feat(ds): Replay QoS1 messages
2023-11-08 03:04:19 +01:00
e745e42093
test(ds): Explore full range of keys when testing ratchet function
2023-11-08 03:04:19 +01:00
87689890ff
chore(ds): Fix linter and compilation warnings
2023-11-08 03:04:19 +01:00
ef46c09caf
feat(ds): Implement ratchet function for bitmask keymapper
2023-11-08 03:04:19 +01:00
164ae9e94a
feat(ds): LTS bitfield storage passes all tests
2023-11-08 03:04:19 +01:00
56b6b176c2
fix(ds): LTS shall keeps the concrete topic indexes
2023-11-08 03:04:19 +01:00
7428e7037b
feat(ds): Bitfield + Learned Topic Structure
2023-11-08 03:04:19 +01:00
ac91dbc58f
feat(ds): Restore LTS trie from a dump
2023-11-08 03:04:19 +01:00
f1ab7c8a7c
feat(ds): Add persist callback to LTS trie
2023-11-08 03:04:19 +01:00
c149e0e2df
fix(ds): Pass topic filter to emqx_ds:make_iterator call
2023-11-08 03:04:19 +01:00
51a6f623fd
refactor(ds): Split out bitfield keymapper to a different module
2023-11-08 03:04:19 +01:00
903b3863d1
chore(ps_ds): make persistent session module use new `emqx_ds` APIs
2023-11-08 03:04:19 +01:00
2972bf14ee
refactor(ds): Implement create_generation gen_rpc storage layer call
2023-11-08 03:04:19 +01:00
6d65707d41
refactor(ds): Implement drop_db function
2023-11-08 03:04:19 +01:00
c6a721a7eb
refactor(ds): Passthrough open_db and get_channels to storage layer
2023-11-08 03:04:19 +01:00
59d01dc823
refactor(ds): Implement emqx_ds:open_db
2023-11-08 03:04:19 +01:00
7095cb8583
refactor(ds): Refactor storage layer
2023-11-08 03:04:19 +01:00
c91df2f5cd
refactor(ds): Create a prototype of replication layer
2023-11-08 03:04:19 +01:00
f1454bb57e
feat(ds): learned topic structure
2023-11-08 03:04:19 +01:00
7ab57824dc
chore(ds): change return type of `storage_layer:next/{1,2}`
...
Part of https://emqx.atlassian.net/browse/EMQX-10942
The goal is to help make it clear to the caller of `next` what to do next: if the iterator
should still be used or if no new messages will ever come out of it.
From:
```erlang
-spec next(iterator()) -> {value, binary(), iterator()} | none | {error, closed}.
```
To:
```erlang
-spec next(iterator()) -> {ok, iterator(), [binary()]} | end_of_stream.
-spec next(iterator(), pos_integer()) -> {ok, iterator(), [binary()]} | end_of_stream.
```
2023-11-08 03:04:19 +01:00
ca8da5723a
Merge pull request #11704 from zmstone/0928-sync-release-53
...
0928 sync release 53
2023-10-02 21:57:32 +02:00
3f6e0e890b
chore: bump emqx_durable_storage and emqx_license app vsn
2023-10-02 20:13:09 +02:00
2358d67908
refactor: move session stuff from `emqx_ds` to `emqx_persistent_session_ds`
...
Part of https://emqx.atlassian.net/browse/EMQX-10942
2023-09-29 18:00:24 -03:00
dafd7c6085
chore: bump apps versions
2023-09-21 10:58:42 +02:00
105bebc250
chore: merge release-52 into master
2023-09-21 10:22:47 +02:00
a2ddd9d5f5
fix(session): respect existing session even if expiry interval = 0
...
If the original connection had Session-Expiry-Interval > 0, and the
new connection set Session-Expiry-Interval = 0, the MQTTv5 spec says
that (supposedly) we still have to continue with the existing session
(if it hasn't expired yet).
Co-Authored-By: Thales Macedo Garitezi <thalesmg@gmail.com>
2023-09-20 22:55:54 +04:00
c1583f7f9d
fix(ds): refine `topic()` type to describe parsed topics
...
And separate it from `topic_filter()` type, which describes parsed
topic filters.
2023-09-20 22:55:52 +04:00
7a9916c84d
fix(sessds): convert ds iterator topics upon opening ds session
2023-09-20 22:55:25 +04:00
045d8b7f10
refactor(ds): reorder functions to improve readability
2023-09-20 22:55:24 +04:00
7326ef550b
fix(sessds): make existing parts of persistent session impl work
2023-09-20 22:55:22 +04:00
5ed5ac48ee
refactor: combine shard id and keyspace into a single value
2023-09-19 12:31:32 -03:00
a511088fd4
refactor: address review comments
2023-09-18 17:47:56 -03:00
b30bcf32bd
feat(ds): introduce keyspace concept
...
Fixes https://emqx.atlassian.net/browse/EMQX-10579
This introduces the concept of "keyspaces" to our durable storage (DS) implementation, and
also refactors some places where "shard" and "keyspace" would be mixed up.
We might want to tune the storage options differently for distinct sets of topics, the
keyspaces. The keyspace is composed by one or more shards.
- Keyspaces are identified simply by binary strings.
- DS configuration is scoped by keyspaces instead of shards.
- Starting a new DS shard requires definining to which keyspace the shard belongs.
2023-09-14 16:39:00 -03:00
cae9ae1fab
fix(ps_router): use `disc_copies` for storing persistent session routes
2023-09-13 10:02:39 -03:00
ea73d674b9
chore: bump version to e5.2.1-alpha.2
2023-09-13 14:55:17 +02:00
f1294736b7
feat(ds): add filter for message persistence
...
Fixes https://emqx.atlassian.net/browse/EMQX-10520
2023-09-11 16:25:34 -03:00
fe4640922d
feat(ds): add delete callback
2023-09-11 16:25:33 -03:00
23d63f5e01
refactor: fork clientid type for `emqx_ds:session_id/0`
2023-09-04 15:38:25 -03:00
922ca5e141
feat(ds): close iterators when handling `UNSUBSCRIBE` packets
...
Fixes https://emqx.atlassian.net/browse/EMQX-9742
2023-08-30 11:14:28 -03:00
33a0048155
refactor: move logic to `ensure_iterator`
2023-08-25 15:24:53 -03:00
dbfacae283
fix: reinstate transactions
2023-08-22 13:29:15 -03:00
3344bfb0bd
refactor: rm `emqx_ds_replay`
2023-08-22 13:29:15 -03:00
021755b82b
refactor: rm iterators from DS `#session{}` record
2023-08-22 13:29:15 -03:00
e4e88ebf36
test: add scenario for node stopping midway during subscribe
2023-08-22 13:29:15 -03:00
Andrew Mayorov