66ec2e6ad0
fix(dsrepl): Retry sending ra commands to the leader
2024-05-24 19:08:05 +02:00
53620e8439
fix(dsrepl): Treat all exceptions from storage layer as recoverable
2024-05-24 19:08:05 +02:00
a7259bc35d
fix(dsstor): Delete generation metadata before dropping it
2024-05-24 19:08:05 +02:00
ba6382adae
fix(dsrepl): trigger "last-resort" pending transitions handler when idle
...
This is a hack to work around the unintended issues causing shard
allocator to become idle even when there are pending transitions.
2024-05-23 14:54:01 +02:00
bf326acd7b
fix(dsrepl): handle stopping non-yet-running shard supervisor
2024-05-23 14:34:27 +02:00
398dc97ed6
Merge pull request #13092 from keynslug/fix/dsrepl/site-autoleave
...
fix(dsrepl): properly handle transaction abort during forget site
2024-05-23 10:56:23 +03:00
6eb04f90a3
fix(ds): Allow to write batches to older generations
2024-05-22 20:28:16 +02:00
59a09fb86f
fix(ds): Apply review remarks
2024-05-22 18:01:33 +02:00
29345aaa30
fix(ds): Fix idle event generation in bitfield_lts layout
2024-05-22 18:01:33 +02:00
b3ded7edce
fix(ds): Fix code review remark
2024-05-22 18:01:33 +02:00
60edf5e9b8
fix(ds): Move responsibility of returning end_of_stream to the CBM
2024-05-22 18:01:33 +02:00
0ff307e789
fix(ds): Include generation ID in the storage events
...
Make sure storage events originating from generation X are handled in
the context of the same generation.
2024-05-22 18:01:33 +02:00
acdae4fba3
fix(ds): Workaround for the idempotency error when dropping gens
2024-05-22 18:01:33 +02:00
eb7c43ee9d
fix(ds): Always store messages in the current generation
2024-05-22 18:01:33 +02:00
e4a73f003a
feat(ds): Implement format_status callback
...
Reduce volume of logs and crash reports from DS
2024-05-22 18:01:32 +02:00
1526c527d0
fix(ds): Log generation operations
2024-05-22 18:01:32 +02:00
aca2d9586c
fix(ds): Fix return type of drop_generation
2024-05-22 18:01:32 +02:00
c6fc76e335
fix(ds): Perform read operations on the leader.
2024-05-22 18:01:32 +02:00
4580906405
fix(ds): Use erpc instead of gen_rpc for `delete_next'
2024-05-22 18:01:32 +02:00
e6c5c1b598
chore(dsrepl): provide more information in rebalancing log messages
2024-05-22 17:24:08 +02:00
c355c9ad50
fix(dsrepl): properly handle transaction abort during forget site
2024-05-22 17:22:55 +02:00
86f99959b0
Merge pull request #13054 from keynslug/fix/EMQX-12365/node-leave
...
fix(dsrepl): anticipate and handle nodes leaving the cluster
2024-05-17 09:43:15 +02:00
ee6e7174cf
fix(sessds): Rename the durable messages DB to `messages`
2024-05-16 21:31:32 +02:00
5157e61418
fix(dsrepl): verify if shards already allocated first
2024-05-16 18:56:54 +02:00
0119728d45
feat(dsrepl): also reflect pending transitions in ds status
2024-05-16 18:56:21 +02:00
26c4a4f597
feat(dsrepl): reflect conflicts and inconsistencies in ds status
2024-05-16 18:32:08 +02:00
7e86e3e61c
fix(dsrepl): anticipate and handle nodes leaving the cluster
...
Also make `claim_site/2` safer by refusing to claim a site for a node
that is already there.
2024-05-16 18:32:07 +02:00
35e360fcbe
feat(api-ds): provide more information on nonexistent site leave
2024-05-14 15:57:41 +02:00
525e4dac95
Merge pull request #13036 from ieQu1/dev/reduce-log-spam
...
tests: Reduce log spam in the failed test suites
2024-05-14 10:53:30 +02:00
ac3f5a083d
test: Reduce log spam in the failed test suites
2024-05-13 22:00:33 +02:00
8506ca7919
Merge pull request #12998 from ieQu1/dev/improve-latency
...
Use leader's clock when calculating LTS cutoff timestamp
2024-05-13 21:54:06 +02:00
9f7ef9f34f
fix(ds): Apply review remarks
2024-05-13 19:35:24 +02:00
a0a3977043
feat(ds): Assign latest timestamp deterministically
2024-05-08 23:17:57 +02:00
2236af84ba
feat(ds): two-stage storage commit on the storage level
2024-05-08 23:17:57 +02:00
1ddbbca90e
feat(ds): Allow incremental update of the LTS trie
2024-05-08 23:17:57 +02:00
68ca891f41
test(ds): Use streams to create traffic
2024-05-08 23:17:57 +02:00
d84c180ccb
feat(dsrepl): avoid contacting unreachable ra servers
...
Assuming estabilished Erlang distribution channel is a reliable way to
tell whether a remote node is reachable.
2024-05-08 18:12:13 +02:00
3642bcd1b6
docs(ds): Fix comment for the builtin DS metrics
2024-05-06 11:21:32 +02:00
b2a633aca1
fix(ds): Use leader's clock for computing LTS safe cutoff time
2024-05-06 11:21:32 +02:00
1ff2e02fd9
feat(ds): Pass current time to the storage layer via argument
2024-05-06 11:21:32 +02:00
8ac9700aab
feat(ds): Add an API for DB-global variables
2024-05-06 11:21:32 +02:00
86d45522e3
fix(dsrepl): Don't reverse elements of batches
2024-05-06 11:21:32 +02:00
bcfa7b2209
fix(ds): Destroy LTS tries when the generation is dropped
2024-05-06 11:21:32 +02:00
9999ccd36c
feat(ds): Ignore safe cutoff time for streams without varying levels
2024-05-06 11:21:32 +02:00
4c76a2574d
fix(ds): Fix egress flush condition
2024-04-21 21:51:31 +02:00
43f8346c00
fix(dssnap): ensure idempotent write of empty chunks
2024-04-19 18:52:33 +02:00
5d7b2e2ce6
fix(dsrepl): attempt leadership transfer on terminate
...
In addition to on removal. The reasoning is basically the same: try to
avoid situations when log entries are replicated (or will be considered
replicated when the new leader is elected) but the leader terminates
before replying to the client.
To be clear: this is a stupid solution. Something much more robust is
needed.
2024-04-15 22:05:24 +02:00
89f42f1171
fix(dsrepl): make placeholder shard process permanent under supervisor
2024-04-15 16:43:52 +02:00
c4d1360b96
fix(dsrepl): trigger election for new ra servers unconditionallly
...
Otherwise we might end up in a situation when there's no member online
yet at the time of the election trigger, and the election will never
happen.
2024-04-15 16:42:29 +02:00
d12e907209
fix(dsrepl): correctly handle ra membership change command results
...
Before this change, results similar to `{error, {no_more_servers_to_try,
[{error, nodedown}, {error, not_member}]}}` were considered retryable
failures, which is incorrect.
2024-04-08 22:44:34 +02:00
ieQu1