Commit Graph

569 Commits

Author SHA1 Message Date
tamo
dcf29e1081
fix the error handling in case there is a panic while creating a dump 2021-05-24 17:33:42 +02:00
tamo
529f7962f4
handle parallel requests for the dump actor 2021-05-24 15:42:12 +02:00
tamo
8a11c6c429
Implements the legacy behaviour of the dump
When asked if a dump exists we check if it's the current dump, and if
it's not then we check on the filesystem for any file matching our
`uid.dump`
2021-05-24 12:35:46 +02:00
tamo
4cbf866821
merge with main 2021-05-12 18:12:37 +02:00
tamo
e0e23636c6
fix the serializer + reformat the file 2021-05-12 17:04:24 +02:00
Marin Postma
295f496e8a
atomic index dump load 2021-05-12 16:21:37 +02:00
bors[bot]
47a1bc34de
Merge #189
189: Fix snapshots r=irevoire a=MarinPostma



Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-05-12 09:28:50 +00:00
tamo
6d837e3e07
the route to create a dump must return a 202 2021-05-11 17:34:34 +02:00
Marin Postma
1b671d4302
fix-snapshot 2021-05-11 13:57:18 +02:00
tamo
c30b32e173
add the criterion attribute when importing dumps from the v1 2021-05-11 13:21:36 +02:00
tamo
9e798fea75
fix the import of dump without unprocessing updates 2021-05-11 13:03:47 +02:00
tamo
384afb3455
fix the way we return the settings 2021-05-11 11:47:04 +02:00
tamo
92a7c8cd17
make clippy happy 2021-05-11 00:27:22 +02:00
tamo
8b7735c20a
move the import of the updates in the v2 and ignore the v1 for now 2021-05-11 00:20:55 +02:00
tamo
7d748fa384
integrate the new Settings in the dumps 2021-05-10 20:48:06 +02:00
tamo
d767990424
fix the import of the updates in the dump 2021-05-10 20:25:12 +02:00
tamo
ef438852cd
fix the v1 2021-05-10 20:25:12 +02:00
tamo
40ced3ff8d
first working version 2021-05-10 20:25:12 +02:00
tamo
5f5402a3ab
provide a way to access the internal content path of all processing State 2021-05-10 20:25:12 +02:00
tamo
26dcb9e66d
bump milli version and fix a performance issue for large dumps 2021-05-10 20:25:12 +02:00
Marin Postma
956012da95
fix dump lock 2021-05-10 20:25:12 +02:00
tamo
24192fc550
fix tests 2021-05-10 20:25:12 +02:00
tamo
efca63f9ce
[WIP] rebase on main 2021-05-10 20:25:09 +02:00
tamo
c3552cecdf
WIP rebase on main 2021-05-10 20:24:18 +02:00
tamo
0f94ef8abc
WIP: dump 2021-05-10 20:24:18 +02:00
tamo
0275b36fb0
[WIP] rebase on main 2021-05-10 20:24:14 +02:00
tamo
1b5fc61eb6
[WIP] rebase on main 2021-05-10 20:23:12 +02:00
tamo
0fee81678e
[WIP] rebase on main 2021-05-10 20:22:18 +02:00
tamo
c4d898a265
split the dumps between v1 and v2 2021-05-10 20:20:57 +02:00
tamo
e389c088eb
WIP: rebasing on master 2021-05-10 20:20:57 +02:00
bors[bot]
ceb8d6e1c9
Merge #186
186: settings fix r=MarinPostma a=MarinPostma

add type checked settigns validation. For now it only transform the settings accepting wildcard


Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-05-10 16:42:12 +00:00
Marin Postma
0cc79d414f
add test 2021-05-10 18:34:25 +02:00
Marin Postma
8d11b368d1
implement check 2021-05-10 18:22:41 +02:00
Marin Postma
706643dfed
type setting struct 2021-05-10 17:30:09 +02:00
Marin Postma
b192cb9c1f
enable string syntax for the filters 2021-05-06 12:48:31 +02:00
bors[bot]
998d5ead34
Merge #182
182: remove facet setting r=MarinPostma a=MarinPostma

remove useless code


Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-05-05 11:22:12 +00:00
Marin Postma
ec7eb7798f
remove facet setting 2021-05-04 22:36:31 +02:00
Marin Postma
a717925caa
remove filters, rename facet_filters to filter 2021-05-04 18:20:56 +02:00
bors[bot]
88ae02f8d9
Merge #174
174: Upgrade Tokenizer r=MarinPostma a=curquiza



Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
2021-05-04 15:57:07 +00:00
Clémentine Urquizar
eb03a3ccb1
Upgrade Milli and Tokenizer 2021-05-04 17:56:19 +02:00
bors[bot]
77740829bd
Merge #177
177: bump milli r=MarinPostma a=MarinPostma



Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-05-04 13:49:37 +00:00
Marin Postma
928fb34eff
bump milli and fix tests 2021-05-04 15:10:22 +02:00
bors[bot]
1e6b40a24b
Merge #172
172: Fix cors authentication issue r=MarinPostma a=MarinPostma

The error was due to the middleware returning an error, instead of a response containing the error.

close #110


Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-05-03 08:38:42 +00:00
Marin Postma
78217bcf18
Fix cors authentication issue 2021-04-29 16:28:12 +02:00
bors[bot]
53c88d9fa3
Merge #170
170: Improve CI r=MarinPostma a=curquiza

Checked with @Kerollmops to improve (a little bit) the CI execution time.

Co-authored-by: Clémentine Urquizar <clementine@meilisearch.com>
2021-04-29 14:08:33 +00:00
bors[bot]
b14fdb1163
Merge #171
171: Update mini-dashboard with version 0.1.2 r=MarinPostma a=mdubus

Update of the mini-dashboard sha1 & assets-url, due to a new release

Co-authored-by: Morgane Dubus <morgane.d@meilisearch.com>
2021-04-29 13:48:54 +00:00
Morgane Dubus
3d5fba94c2 Update mini-dashboard with version 0.1.2 2021-04-29 15:22:41 +02:00
Clémentine Urquizar
3ee2b07918
Improve CI 2021-04-29 15:19:48 +02:00
bors[bot]
8bc7dd8b03
Merge #143
143: Shared update store r=irevoire a=MarinPostma

This PR changes the updates process so that only one instance of an update store is shared among indexes.

This allows updates to always be processed sequentially without additional synchronization, and fixes the bug where all the first pending update for each index were reported as processing whereas only one was.

EDIT:

I ended having to rewrite the whole `UpdateStore` to allow updates being really queued and processed sequentially in the ordered they were added. For that purpose I created a `pending_queue` that orders the updates by a global update id.

To find the next `update_id` to use, both globally and for each index, I have created another database that contains the next id to use.

Finally, all updates that have been processed (with success or otherwise) are all stores in an `updates` database.

The layout for the keys of these databases are such that it is easy to iterate over the elements for a particular index, and greatly reduces the amount of code to do so, compared to the former implementation.

I have also simplified the locking mechanism for the update store, thanks to the StateLock data structure, that allow both an arbitrary number of readers and a single writer to concurrently access the state. The current state can be either Idle, Processing, or Snapshotting. When an update or snapshotting is ongoing, the process holds the state lock until it is done processing its task. When it is done, it sets bask the state to Idle.

I have made other small improvements here and there, and have let some other for work, such as:
- When creating an update file to hold a request's content, it would be preferable to first create a temporary file, and then atomically persist it when we have written to it. This would simplify the case when there is no data to be written to the file, since we wouldn't have to take care about cleaning after ourselves.
- The logic for content validation must be factored.
- Some more tests related to error handling in the process_pending_update function.
- The issue #159

close #114


Co-authored-by: Marin Postma <postma.marin@protonmail.com>
2021-04-27 18:41:55 +00:00
Thomas Payet
e6fd1afc3d
Merge pull request #163 from meilisearch/curquiza-patch-1
Update README.md
2021-04-27 18:51:04 +02:00