2afd2c81ba
Change hash function to blake2 for partition keys based on strings
2021-03-10 16:23:57 +01:00
f319a7d374
Refactor model stuff, including cleaner CRDTs
2021-03-10 16:21:56 +01:00
6a3dcf3974
Rename n_tokens into capacity
2021-03-10 14:52:03 +01:00
7cda917b6b
update condition
2021-03-05 17:08:03 +01:00
d7e005251d
Not fully tested: new multi-dc MagLev
2021-03-05 16:22:29 +01:00
3882d5ba36
Remove epidemic propagation for fully replicated stuff: write directly to all nodes
2021-03-05 15:09:18 +01:00
49c25a1509
Simulate stuff moving around
2021-02-25 10:53:33 +01:00
5fe95ebae7
fix tracing
2021-02-24 12:18:01 +01:00
13e2eda0c2
Arrange block manager
2021-02-24 11:58:03 +01:00
09fd6ea7f0
I was tired yesterday
2021-02-24 11:05:59 +01:00
a52ab69640
fix misuse of sled transactions
2021-02-23 22:45:36 +01:00
20e6e9fa20
Update sled & try to debug deadlock (but its in sled...)
2021-02-23 21:27:28 +01:00
bf25c95fe2
Make updated() be a sync function that doesn't fail
2021-02-23 20:25:15 +01:00
28bc967c83
Handle correctly deletion dues to offloading
2021-02-23 19:59:43 +01:00
55156cca9d
Several changes in table_sync:
...
- separate path for case of offloading a partition we don't store
- use sync::Mutex instead of tokio::Mutex, make less fn's async
2021-02-23 19:11:02 +01:00
40763fd749
Cargo fmt
2021-02-23 18:46:25 +01:00
6e6f7e8555
Replace some checksums where it makes sense
2021-02-23 18:14:37 +01:00
e8e4418ca7
Add blake2 and xxhash hash functions
2021-02-23 17:52:28 +01:00
b1b640ae8b
rename hash() to sha256sum(), we might want to change it at some places
2021-02-21 15:24:30 +01:00
80892df8cc
Some refactoring
2021-02-21 13:11:10 +01:00
3bcbbe1e31
More precise logging (warn only when returning a 500)
2021-02-20 00:30:39 +01:00
10b983b8e7
Add verification of part numbers in CompleteMultipartUpload (WIP #30 )
2021-02-20 00:13:07 +01:00
1de96248e0
add application/xml header and missing xml escapes
2021-02-19 23:40:18 +01:00
5d1fa591d9
Add compatibility list
2021-02-19 19:10:23 +01:00
e64ecbdccd
S3 compatibility: return 404 instead of 400 on some multipart commands
2021-02-19 18:51:05 +01:00
55a2a636ca
Implement ListObjectsV2
2021-02-19 16:44:06 +01:00
02d512f3fd
Fix #28 , extra headers being ignored (because of profound stupidity)
2021-02-19 12:38:22 +01:00
76390085ef
Small improvements in the S3 put workflow
2021-02-19 12:11:02 +01:00
36814be447
Fix S3 ListObjects result and replace println!s by debug!s
2021-01-16 16:05:54 +01:00
e818f51073
Forgot a bump
2021-01-15 18:36:51 +01:00
ceeb0732a2
Use 0.1.0b instead of 0.1.0 (for compatibility with new Error type)
2021-01-15 18:27:58 +01:00
1d1d497e2b
Bump everything to 0.1.1
2021-01-15 17:54:48 +01:00
851893a3f2
Do not accept domains such as [hello
2021-01-15 17:49:10 +01:00
f8a40e8c4f
Explicitly set code path unreachable
2021-01-15 17:11:15 +01:00
fad7bc405b
Behavior problem: do not panic anymore + add tests
2021-01-15 17:03:54 +01:00
1e10c6a61c
Doc tests that do not compile/work must be tagged with ignore
2021-01-15 17:03:38 +01:00
11a79a95dd
Simplify Error file
2021-01-15 16:25:44 +01:00
c441a358cd
Remove unused dependencies
2021-01-15 16:16:32 +01:00
f496e41ef4
Replace an already done check by unreachable!()
2021-01-15 15:44:44 +01:00
2f4378a9c4
Fix formatting
2020-12-17 22:51:44 +01:00
ccda9ab1ca
Merge branch 'master' into feature/website
2020-12-17 21:09:50 +01:00
3132deca58
Web server access control
2020-12-17 20:43:14 +01:00
011ff87b5f
Push update
2020-12-15 13:23:22 +01:00
3bc4d57a0f
First implementation of the CLI
2020-12-15 12:48:24 +01:00
a3566e49da
Start to implement Website CLI
2020-12-14 21:50:40 +01:00
d0eb6a457f
Migrate RPC to new schema
2020-12-14 21:46:49 +01:00
96388acf23
Implement migration
2020-12-12 21:35:29 +01:00
8956db2a81
Make less things public
2020-12-12 17:58:19 +01:00
5c6c067b0c
More documentation on CRDTs (we should probably extract this to a
...
standalone crate!)
2020-12-12 17:06:40 +01:00
e1ce2b228a
WIP table migration
2020-12-12 17:00:31 +01:00
0b3084ca5f
Merge branch 'master' into doc/model
2020-12-12 16:05:28 +01:00
e8c12072ce
Merge branch 'master' into feature/website
2020-12-10 20:12:56 +01:00
51d0c14e44
CLI structure
2020-12-10 18:13:32 +01:00
022b386a50
Improved compatibility on list API call
2020-12-06 15:39:03 +01:00
986e15459a
Merge branch 'master' into feature/website
2020-12-06 15:21:09 +01:00
4a5bbbb810
Propose ETag fix
2020-12-05 19:23:46 +01:00
76b489f3d3
Reformulate patch
2020-12-05 19:20:07 +01:00
bd7e3d1bd1
Fix Content-Length
2020-12-05 18:57:22 +01:00
9f46fb699a
Content-range fix
2020-12-05 16:37:59 +01:00
15f409d404
Merge branch 'master' into feature/website
2020-11-29 17:19:55 +01:00
d54f15b2c6
Small optimisation
2020-11-29 17:07:14 +01:00
fed97f37e1
ETag patch
2020-11-29 16:38:01 +01:00
601ae25ad2
Small refactorings
2020-11-29 16:21:28 +01:00
cbd10c1b0a
Add some doc on LWW
2020-11-23 18:17:48 +01:00
8722e27600
CRDT doc
2020-11-23 17:49:21 +01:00
aa320aa04a
Merge branch 'master' into feature/website
2020-11-22 19:54:47 +01:00
17dc610f8a
Also check hash for < 3KB files
2020-11-22 11:14:46 +01:00
435d5f9205
Fix base64/hex checksum comparison
2020-11-22 11:04:33 +01:00
fb18f5e17a
Fix wrong http status code
2020-11-21 18:14:02 +01:00
28efe341cb
Merge branch 'master' into feature/website
2020-11-21 18:01:50 +01:00
b7a377308b
Handle HEAD
2020-11-21 17:58:14 +01:00
a88fd49f71
Use handle_get
2020-11-21 17:50:19 +01:00
0f33231ee6
We are able to serve a file
2020-11-21 15:15:25 +01:00
d4c7f4e374
Fix host to key
2020-11-21 12:01:02 +01:00
2f6eca4ef3
Merge remote-tracking branch 'origin/master' into feature/website
2020-11-21 10:52:27 +01:00
a8b3c8fd58
data hexdump in warning
2020-11-20 23:53:54 +01:00
f9be964c3f
Warning when cannot decode entry (data format incompatibilities)
2020-11-20 23:53:54 +01:00
4f7f1d1cb3
less type bounds
2020-11-20 23:53:54 +01:00
68be5072e5
simplify
2020-11-20 23:53:54 +01:00
173f0dbac9
oops
2020-11-20 23:53:54 +01:00
f8a04852a2
Convert bucket table to better CRDT representation
2020-11-20 23:53:54 +01:00
e02e9e035e
Begin improve model to use better CRDTs
2020-11-20 23:53:54 +01:00
5b363626f4
Support punnycode
2020-11-20 21:23:32 +01:00
2dc9a48022
Fix reconciliation logic
2020-11-20 20:12:32 +01:00
e9fd265ce6
Slight refactoring to make things clearer with DeletedFilter
2020-11-20 20:11:04 +01:00
04f455ff7f
Make it compile again
2020-11-19 14:56:00 +01:00
fc427b0b66
Merge branch 'master' into feature/website
2020-11-19 14:39:30 +01:00
6076d869b1
Build error
2020-11-11 21:17:34 +01:00
2765291796
Build path correctly
2020-11-11 19:48:01 +01:00
7d7b9e95a9
Simplify and_then(Some) as map() and remove move
2020-11-11 16:36:48 +01:00
5a5592c176
Replace with option syntaxic sugar
2020-11-11 16:12:42 +01:00
d445c4ef9c
WIP fetch object
2020-11-11 15:24:25 +01:00
3cb3994cd2
Add documentation to host_to_bucket
2020-11-10 17:05:10 +01:00
cacf8ddf2d
Panic when it is a logical error
2020-11-10 15:52:20 +01:00
d1b2fcc1e7
Rewrite for clarity
2020-11-10 15:48:40 +01:00
ab62c59acb
Fix indent again
2020-11-10 15:40:33 +01:00
8797eed0ab
Fixes due to integration tests
2020-11-10 15:32:04 +01:00
1e52ee9f5b
Rewrite authority to host while staying on stack
2020-11-10 15:26:48 +01:00
27795a390c
Fix formatting
2020-11-10 09:59:52 +01:00
4093833ae8
Extract bucket
2020-11-10 09:57:07 +01:00
09137fd6b5
Log host
2020-11-08 16:06:52 +01:00
c78df603d7
Add some documentation
2020-11-08 16:02:16 +01:00
71721f5bcf
Merge branch 'master' into feature/website
2020-11-08 15:53:33 +01:00
0791e7164e
Parse host header
2020-11-08 15:47:25 +01:00
a50fa70d45
Refactor error management in API part
2020-11-08 15:05:28 +01:00
6a8b972f3a
Modif parse_bucket_key to fail when bucket is ""
2020-11-08 13:39:44 +01:00
e1415f5976
Add a second test
2020-11-07 15:34:53 +01:00
9a50ce12a8
Fix formatting
2020-11-07 13:59:30 +01:00
8f4ada1965
Add a test for parse_bucket_key
2020-11-07 13:53:32 +01:00
0d3bc169ee
It compiles!
2020-11-03 12:37:16 +01:00
b3caa3628d
Fix description of the crate
2020-11-02 15:57:23 +01:00
cea871d944
Skeleton to the new web API
2020-11-02 15:48:39 +01:00
104e2ce0a2
Add "web" configuration entry
2020-10-31 17:28:56 +01:00
5faf069e33
trace test
2020-07-15 15:41:49 +02:00
27a0d0d859
pretty
2020-07-15 15:37:35 +02:00
1c70552f95
Validate content MD5 and SHA256 sums for PutObject and UploadPart
2020-07-15 15:31:13 +02:00
6c7f9704ea
Implement correct ETag for objects created with PutObject
2020-07-13 16:51:30 +02:00
9305e5e87f
More headers taken into account
2020-07-09 17:04:43 +02:00
44dba0e53c
cargo fmt
2020-07-08 17:34:37 +02:00
64a6eda0d2
Migrate S3 api to use new model
2020-07-08 17:33:24 +02:00
84bbbfaa7b
Add multiple headers to object model
2020-07-08 17:33:14 +02:00
a5fa2a136b
(WIP) New object table model, TODO: update API calls to use it
2020-07-08 16:46:47 +02:00
86bf4dedac
Add support for model migrations
2020-07-08 16:10:53 +02:00
86fb7bbba5
Apply cargo fmt; add trace output when request signature is bad
2020-07-08 13:33:02 +02:00
f22ecb60a8
Update to Hyper 0.13.6 that accepts non-Sync streams in wrap_stream.
...
Simplifies code and makes it possible to publish on crates.io
2020-07-07 17:15:53 +02:00
3b0b11085e
Add versions to dependencies
2020-07-07 14:18:47 +02:00
cc65cdc0fe
Add license, description and repository to .toml files
2020-07-07 14:14:58 +02:00
bec26a1312
Rename garage_core to garage_model
2020-07-07 13:59:22 +02:00
fbe8fe81f2
Add automatic peer discovery from Consul
2020-06-30 18:33:14 +02:00
db6f1f35a8
Rename epidemic_factor to epidemic_fanout (that's what it is); complete conf example in readme
2020-06-30 15:03:34 +02:00
fa13cf6996
Repair: do not mark deleted when upstream object is not found
...
With the previous behaviour, repairing could see some data as absent
and decide that the object or version was deleted,
thus going on to delete the version and blocks.
In the case where read_quorum + write_quorum <= replication_factor
however, entries may not yet be returned by the get, thus data would
have been deleted that should hot have been. The new behavior is more
cautious and just skips the entry when the warning is emitted.
2020-05-04 13:30:42 +00:00
b46a7788d1
Implement HTTP ranges in get
2020-05-04 13:09:23 +00:00
16fbb32fd3
Rate limit requests a bit more seriously
...
droping the slot later (after reading the request response)
means that we aren't freeing our quota slot,
so the maximum number of simultaneous requests now also counts the
response reading phase
TODO next: quotas per rpc destination node, or maybe per datacenter (?)
2020-05-01 19:18:54 +00:00
d867bbcfb5
Implement DeleteObjects
2020-05-01 15:52:35 +00:00
3324971701
Slightly improved S3 compatibility
...
- ListBucket does not require any of the parameters (delimiter,
prefix, max-keys, etc)
- URLs are properly percent_decoded
- PutObject and DeleteObject calls now answer correctly
(empty body, version id in the x-amz-version-id: header)
2020-05-01 14:30:50 +00:00
3686f100b7
Compatibility fixes
2020-04-28 10:35:04 +00:00
0957d0fdfa
Work on API
2020-04-28 10:18:14 +00:00
0877a5500c
Abort multipart upload
2020-04-26 20:46:33 +00:00
81ecc4999e
Implement multipart uploads
2020-04-26 20:39:32 +00:00
1999c0ae51
Update delete code
2020-04-26 19:11:19 +00:00
0a283e4e70
Fix deletion propagation
2020-04-26 18:59:17 +00:00
9cb870f950
Prepare for multipart uploads
2020-04-26 18:55:13 +00:00
ea7e4748ed
S3 compatibility: fix bucket listing and HEAD and PUT on bucket
2020-04-26 16:22:33 +00:00
0e49e0c8b5
Add key table to repair procedure
2020-04-26 16:22:22 +00:00
e3203f998b
Remove leading / in keys; better delimiter handling
2020-04-24 22:28:15 +02:00
be4831d768
Less verbosity
2020-04-24 19:27:27 +00:00
a52db67954
xml escape
2020-04-24 18:56:00 +00:00
91b2d1fcc1
Some basic S3 functionnality
2020-04-24 18:47:11 +00:00
f2e05986c4
Starting to be S3 compatible
2020-04-24 17:46:52 +00:00
d8f5e643bc
Split code for modular compilation
2020-04-24 10:10:01 +00:00
51fb3799a1
Key management admin commands
2020-04-23 20:25:45 +00:00
4ef84a0558
Move repair to separate file
2020-04-23 18:36:12 +00:00
44a1089d95
Make table objects slightly more fool-proof; add key table
2020-04-23 18:16:52 +00:00
c9c6b0dbd4
Reorganize code
2020-04-23 17:05:46 +00:00
01a8acdeec
Better error reporting
2020-04-23 16:23:06 +00:00
82f4cd8719
Continue pinging nodes when they are down ; overall better handling of down nodes
2020-04-23 16:06:11 +00:00
2fe82be3bc
RPC to ourself do not pass through serialization + HTTPS
2020-04-23 14:40:59 +00:00
37f880bc09
RequestStrategy with possible interruption or not
2020-04-23 13:37:10 +00:00
73574ab43e
Fix in rpc_client (see comment in code)
2020-04-22 20:42:23 +00:00
897fafa8db
Improvements to block resync queue & worker
2020-04-22 20:32:58 +00:00
2556a1e383
I'm stupid though
2020-04-22 20:06:12 +00:00
231cb32955
Do not delete block if just a single replication error. Write TODO stuff.
2020-04-22 19:25:15 +00:00
8971f34c81
Well they still have to exit when we're exiting though
2020-04-22 17:04:33 +00:00
e8214cb180
Better concurrency:
...
Use Notify instead of stupid sleep in background worker
Use Semaphore to limit concurrent requests in rpc_client
Make more background tasks cancellable
2020-04-22 16:51:52 +00:00
ec59e896c6
Make UUID & Hash Copy and remove some .clone() noise
2020-04-21 17:08:42 +00:00
8915224966
Return BadRequest codes for some admin_rpc failure cases
2020-04-21 16:45:32 +00:00
b1ddb933b0
Make the repair command accept subcommands to not do everything all the time
2020-04-21 16:40:17 +00:00
a04218047e
Do full sync on node (re)start
2020-04-21 16:15:32 +00:00
2a84d965ab
Improve table sync
2020-04-21 16:05:55 +00:00
0226561035
Do not insert deletion marker if there is no object to delete
2020-04-21 14:33:12 +00:00
be0a2bae81
Add node tags in configuration
2020-04-21 14:08:28 +00:00
cc4f2f1cfb
Pretty logging
2020-04-21 12:54:55 +00:00
53cf4d1baa
Log which workers are doing what
2020-04-19 21:33:38 +00:00
ec7f9f07e2
Implement repair object->version and version->block ref
2020-04-19 21:27:08 +00:00
04acaea231
Don't do version & block_ref updates in background on deletion
2020-04-19 20:52:20 +00:00
5ae32972ef
Implement repair command
2020-04-19 20:36:36 +00:00
a54f3158f1
Less output
2020-04-19 19:38:45 +00:00
ea75564851
More aggressive sync timings & improve other stuff
2020-04-19 17:59:59 +00:00
e325c7f47a
Add hostname to node info
2020-04-19 19:08:48 +02:00
a6129d8626
Begin implement bucket management & admin commands
2020-04-19 17:15:48 +02:00
302502f4c1
Add support for fully replicated tables with epidemic dissemination of updates
2020-04-19 15:14:23 +02:00
7131553c53
Refactor sharding logic; coming next: full replication with epidemic dissemination
2020-04-19 13:22:28 +02:00
4ba54ccfca
Reorder imports.
...
Trying to separate:
1. Stuff for handling the swarm of nodes and generic table data replication
2. Stuff for the object store core application: metadata tables and block management
3. Stuff for the S3 API
2020-04-18 19:39:57 +02:00
bd1618e78e
Remove proto.rs & move some definitions out of data.rs
2020-04-18 19:30:05 +02:00
f41583e1b7
Massive RPC refactoring
2020-04-18 19:21:34 +02:00
3f40ef149f
Fix sync: use max root checksum level
2020-04-17 21:59:07 +02:00
f62b54f1df
Fix add to resync on incref
2020-04-17 21:14:06 +02:00
ace07da7c1
Fix walk_ring_from
2020-04-17 21:08:43 +02:00
40c48e6a59
Several resync workers; add delay on retry resync
2020-04-17 20:58:10 +02:00
29a1e94f23
Implement missing handler for read_range
2020-04-17 19:38:47 +02:00
b4e96bdcf0
Fix paths :o
2020-04-17 19:20:17 +02:00
4abfb75509
Implement sending blocks to nodes that need them
2020-04-17 19:16:08 +02:00
db1c4222ce
Don't send items...
...
...if syncer doesn't need them because he's going to delete the partition anyway.
Also, fix block resync queue
2020-04-17 18:51:29 +02:00
4bacaaf53f
Resync block on read error
2020-04-17 18:38:11 +02:00
b780f6485d
Make sync send data both ways
2020-04-17 18:27:29 +02:00
69f1d8fef2
WIP
...
TODOs:
- ensure sync goes both way
- finish sending blocks to other nodes when they need them before deleting
2020-04-17 17:09:57 +02:00
e41ce4d815
Implement getting missing blocks when RC increases
...
Issue: RC increases also when the block ref entry is first put by the actual client.
At that point the client is probably already sending us the block content,
so we don't need to do a get...
We should add a delay before the task is added or find something to do.
2020-04-17 15:40:13 +02:00
867646093b
Table range deletion
2020-04-17 14:49:10 +02:00
6ce14e2c9e
Make all requests continue in the background even after we got enough responses.
2020-04-16 23:13:15 +02:00
768d22ccdb
Better prints again, and a great question
2020-04-16 19:57:13 +02:00
2a05fd135a
Change debug prints a bit
2020-04-16 19:37:08 +02:00
2f3b1a072f
WIP
2020-04-16 19:28:02 +02:00
2832be4396
WIP
2020-04-16 18:41:10 +02:00
e8d750175d
Implement ring comparison algorithm
2020-04-16 17:04:28 +02:00
f01c1e71b5
Begin work on sync...
2020-04-16 14:50:49 +02:00
43ce5e4ab4
Fix table RPC to not be interruptible
2020-04-12 23:05:53 +02:00
2bea76ce16
Small refactorings
2020-04-12 22:24:53 +02:00
9f8b3b5a18
TLS for command line client
2020-04-12 19:41:19 +02:00
c788fc9f9e
Cleanup
2020-04-12 19:18:31 +02:00
d2814b5c33
TLS works \o/
...
So, the issues were:
- webpki does not support IP addresses as DNS names in URLs,
so I hacked the HttpsConnector to always provide a fixed string
as the DNS name for server certificate validation
- the certificate requied a SAN section which was complicated to build
but eventually the solution is there in genkeys.sh
2020-04-12 19:00:30 +02:00
d1e8f78b2c
Trying to do TLS
2020-04-12 15:51:19 +02:00
5967c5a5af
Refactor a bit
2020-04-12 13:03:55 +02:00
419c70e506
fixes
...
- make block_put call uninterruptible by client
- used meta_replication_factor instead of data_replication_factor
- listen on ipv6
2020-04-12 12:10:33 +02:00
9c931f5eda
Keep network status & ring in a tokio::sync::watch
...
advantages
- reads don't prevent preparing writes
- can be followed from other parts of the system by cloning the receiver
2020-04-11 23:53:32 +02:00
5dd59e437d
Local refcounting of blocks
2020-04-11 23:00:26 +02:00
dcf58499a4
table::insert_many, version_table::updated
2020-04-11 19:43:29 +02:00
53289b69e5
Background task runner that replaces tokio::spawn
2020-04-11 18:51:11 +02:00
4a2624b76a
We don't want the debugging to fail, actually.
2020-04-10 23:43:35 +02:00
cf8fd948fc
Add block ref table
2020-04-10 23:11:52 +02:00
ff4fb97568
(Try to) disable LTO ?
2020-04-10 22:55:01 +02:00
a50f07dfdc
Refactor
2020-04-10 22:26:48 +02:00
3477864142
Fix the Sync issue. Details:
...
So the HTTP client future of Hyper is not Sync, thus the stream
that read blocks wasn't either. However Hyper's default Body type
requires a stream to be Sync for wrap_stream. Solution: reimplement
a custom HTTP body type.
2020-04-10 22:01:48 +02:00
d66c0d6833
Why is it not Sync??
2020-04-09 23:45:07 +02:00
a3eb88e601
Locally, transactions
2020-04-09 20:58:39 +02:00
1d786c2c66
Something works
2020-04-09 18:43:53 +02:00
101444abb3
Some progress
2020-04-09 17:32:28 +02:00
4c1aee42d5
Reorganize table API
2020-04-09 16:16:27 +02:00
a450103ed0
Work & TODO
2020-04-08 23:47:34 +02:00
cc580da0ae
Some work
2020-04-08 23:01:49 +02:00
bacc76a057
Some work in actually storing things
2020-04-08 22:00:41 +02:00
d50edcdb4f
Basic workflow for client PUT, next is do actual storage
2020-04-08 00:39:07 +02:00
90cdffb425
custom data type for hashes and identifiers
2020-04-07 18:10:20 +02:00
82b7fcd280
Config management & ring
2020-04-07 17:00:48 +02:00
061e676136
Refactor; ability to update network config
2020-04-07 16:26:22 +02:00
46d5b896e8
Fix pinging
2020-04-07 00:00:43 +02:00
a7b85146fe
Forgot to recalculate hash at strategic locations
2020-04-06 23:10:54 +02:00
a09f019cc5
Seems to be fixed
2020-04-06 22:54:03 +02:00
87f2b4d2fc
Ununderstandable error
2020-04-06 22:27:51 +02:00
3c36b449a3
Some work
2020-04-06 21:02:15 +02:00
1a5e6e39af
Some more basic work
2020-04-06 19:55:39 +02:00
7102db1d54
First commit: skeleton for something great
2020-04-05 23:33:42 +02:00
