Commit graph

507 commits

Author SHA1 Message Date
Pavel Karpy
486ec8a2e5 [#1318] metabase: Reuse comparing func from require
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
1caeb0b648 [#1318] metabase: Add missing error check in tests
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
2aec5736e0 [#1318] node/gc: Fix tombstones handling
Handle only expired tombstones.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
7799f8e4cf [#1318] engine: Change tombstone clear process
- Delete objects physically on tombstone's arrival;
- Store information about tombstones in the Graveyard;
- Clear Graveyard every epoch based on the information about TS in the
network.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
2583f608e8 [#1318] metabase: Add DropGraves method
Method clean up Graveyard and should be called by the routine that handles
tombstone expiration.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
daab30c391 [#1318] metabase: Add iteration offset
Add offset element to the iterations over deleted objects (both the
Graveyard and the Garbage buckets).

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
fe8076e60a [#1318] metabase: Add immediate object deletion
Mark objects with GC immediately after a Tombstone is received.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
8107c8d1a9 [#1318] metabase: Separate buckets with TS and GC marks
It allows storing information about object in both ways at the same time:
1. Metabase should know if an object is covered by a tombstone (that is
not expired yet);
2. It should be possible to physically delete objects covered by a
tombstone immediately (mark with GC) but keep tombstone knowledge.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-29 16:38:52 +03:00
Pavel Karpy
bd27837364 [#1321] node: Register GC event channel before shard init
Morph "NewEpoch" event handling was registered in a closure over
`addNewEpochNotificationHandler` func. That may lead to the data race:
if a shard was initialized before the event registration, everything works
as planned, but if registration was made earlier, it was not able to
include GC handlers since a shard has not called `eventChanInit` yet and,
therefore, it has not registered handler yet.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-04-25 10:39:37 +03:00
Evgenii Stratonikov
622ea4818f [#1250] *: Remove io/ioutil imports
It is deprecated starting from go1.16.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-04-07 17:23:28 +03:00
Leonard Lyubich
629a4f79cd [#1293] metabase: Fix out-of-range panic in freePotentialLocks
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-04-05 15:38:51 +03:00
Evgenii Stratonikov
48eb87d32c [#1149] writecache: Prevent corruption in Head
Also, remove optimization comments:
1. Having to maintain an execute the same logic for headers as for
   objects is quite inefficient, as it increases memory footprint.
2. Unmarshaling object is a cheap operation if data slice is in memory.
3. For unmarshaling header-only, I think we need SDK support.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:40:25 +03:00
Evgenii Stratonikov
68903c9fd9 [#1143] shard: Support degraded mode in Get and GetRange
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:33:22 +03:00
Evgenii Stratonikov
f058cead8f [#1143] shard: Handle some errors in degraded mode
If metabase is corrupted for some reason, failback to
checking blobstor directly.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:33:22 +03:00
Evgenii Stratonikov
08e7914729 [#1143] blobstor: Implement existsSmall check
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:33:22 +03:00
Evgenii Stratonikov
aa0cc1f824 [#1143] blobovnicza: Copy object data in Get
Data returned from `*bbolt.Bucket.Get()` is only valid for the lifetime
of the transaction.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:33:22 +03:00
Evgenii Stratonikov
6472a170eb [#1143] shard: Introduce explicit Degraded mode
`Degraded` mode is set automatically after error counter is over the
threshold. `ReadOnly` mode can still be set by an administrator.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-31 15:33:22 +03:00
Evgenii Stratonikov
de5a2f6574 [#1262] metabase: Remove list index in place Delete
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
43867a3093 [#1262] metabase: Do not allocate intermediate slices for indices
```
name              old alloc/op   new alloc/op   delta
Put/parallel-8       123kB ± 4%     119kB ± 3%  -2.72%  (p=0.006 n=10+9)
Put/sequential-8     170kB ± 1%     168kB ± 1%  -1.42%  (p=0.000 n=10+10)

name              old allocs/op  new allocs/op  delta
Put/parallel-8         473 ± 1%       469 ± 0%  -0.87%  (p=0.000 n=10+10)
Put/sequential-8       792 ± 0%       787 ± 0%  -0.58%  (p=0.000 n=10+10)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
0e9b6be3fd [#1262] metabase: Remove intermediate allocations in decodeList
```
name              old time/op    new time/op    delta
Put/parallel-8      1.57ms ±11%    1.51ms ± 3%   -4.06%  (p=0.043 n=9+10)
Put/sequential-8    5.16ms ± 2%    5.16ms ± 3%     ~     (p=1.000 n=9+10)

name              old alloc/op   new alloc/op   delta
Put/parallel-8       126kB ± 4%     123kB ± 4%   -2.54%  (p=0.016 n=8+10)
Put/sequential-8     171kB ± 1%     170kB ± 1%     ~     (p=0.182 n=9+10)

name              old allocs/op  new allocs/op  delta
Put/parallel-8         565 ± 2%       473 ± 1%  -16.18%  (p=0.000 n=9+10)
Put/sequential-8       819 ± 1%       792 ± 0%   -3.34%  (p=0.000 n=9+10)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
d45df614fb [#1262] metabase: Optimize decodeList
Prevent additional allocation during `append` in `Put`.

```
name              old alloc/op   new alloc/op   delta
Put/parallel-8       131kB ± 1%     126kB ± 4%  -3.87%  (p=0.005 n=8+8)
Put/sequential-8     172kB ± 1%     171kB ± 1%  -0.73%  (p=0.028 n=10+9)
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
456e1584d6 [#1262] metabase: Add benchmarks for Put
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
f1223b46df [#1262] blobovnicza: Make helper in Put function idempotent
`Batch` can execute the function multiple times leading to multiple
increases of a size approximation.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-28 17:08:11 +03:00
Evgenii Stratonikov
ad92493b86 [#1268] blobstor: Cleanup zstd encoders/decoders
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-25 14:05:06 +03:00
Leonard Lyubich
318639e5bf [#1247] *: Clarify docs about returned errors from apistatus package
`apistatus` package provides types which implement build-in `error`
interface. Add `error of type` pattern when documenting these errors in
order to clarify how these errors should be handled (e.g. `errors.Is` is
not good).

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-17 16:34:00 +03:00
Leonard Lyubich
70ffdf3478 [#1247] object: Return NOT_FOUND and ALREADY_REMOVED statuses
Replace `ErrNotFound`/`ErrAlreadyRemoved` error from
`pkg/core/object` package with `ObjectNotFound`/`ObjectAlreadyRemoved`
one from `apistatus` package. These errors are returned by storage
node's server as NeoFS API statuses.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-17 16:34:00 +03:00
Alex Vanin
0adb29c035 [#1214] Fix metabase put logging
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2022-03-16 17:17:18 +03:00
Alex Vanin
8f476f3c4d [#1238] Adopt neofs-node for non pointer slices in SDK
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2022-03-16 15:29:31 +03:00
Alex Vanin
7f3195b197 [#1238] Adopt neofs-cli for non pointer slices in SDK
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2022-03-16 15:29:31 +03:00
Leonard Lyubich
9feb5f9405 [#1175] engine: Fix AddShard implementation
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
9dff07200c [#1181] local storage: Process expired locks similar to tombstones
There is a need to process expired `LOCK` objects similar to `TOMBSTONE`
ones: we collect them on `Shard`, notify all other shards about
expiration so they could unlock the objects, and only after that mark
lockers as garbage.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
323dea95c6 [#1181] Do not use deprecated elements
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
e3f4e48d87 [#1175] Fix typo in error return docs
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
ced854bc2e [#1175] object/fmt: Handle LOCK objects
Make `FormatValidator.ValidateContent` to verify payload of `LOCK`
objects. Pass locked objects to `Locker` interface. Require from
`Locker.Lock` to return `apistatus.IrregularObjectLock` error on a
corresponding condition.

Also add error return to `DeleteHandler.DeleteObjects` method. Require
from method to return `apistatus.ObjectLocked` error on a corresponding
condition. Adopt implementations.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
d163008b63 [#1175] storage engine: Fix Lock implementation
`Lock` must return `apistatus.IrregularObjectLock` error instead of
`apistatus.ObjectLocked` one.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
07107c8bfd [#1175] metabase: Update locker list during Inhume
`Inhume` operation can potentially mark lockers as garbage. There is a
need to update locker list in locked bucket.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
26b8e33ed7 [#1175] metabase: Fix encoding of updated locker list
Even single locker must be encoded using `encodeList`, or subsequent
`decodeList` call will fail.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
88cefa7c5f [#1175] storage engine: Return status error on Delete of locked object
Make `StorageEngine.Delete` to forward first encountered
`apistatus.ObjectLocked` error during shard processing.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
97b6ec8150 [#1175] storage engine: Return status error on Inhume of locked object
Make `StorageEngine.Inhume` to forward first encountered
`apistatus.ObjectLocked` error during shard processing.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
545861d1bc [#1175] storage engine: Implement Lock operation
Implement `StorageEngine.Lock` method which works similar to `Inhume`
but calls `Lock` on the processing shards.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
6ed85ff1e1 [#1175] shard: Implement Lock operation
Implement `Shard.Lock` method which required rw mode and calls `Lock` on
underlying metabase.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
e21eedee06 [#1175] shard: Return status error on Inhume of locked object
Make `Shard.Inhume` to forward `apistatus.ObjectLocked` returned by
underlying metabase.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
b585791d6e [#1175] metabase: Return status error on Lock of irregular object
Make `DB.Lock` to return `apistatus.IrregularObjectLock` if at least one
of the locked objects is irregular (not of type REGULAR).

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
23fcacd3f2 [#1175] metabase: Return status error on Inhume of locked object
Make `DB.Inhume` to return `apistatus.ObjectLocked` if at least one of
the inhumed objects is locked.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
303eb2e078 [#1176] metabase: Ignore locked objs in IterateCoveredByTombstones
Make `DB.IterateCoveredByTombstones` to not pass locked objects to the
handler. The method is used by GC, therefore it will not consider locked
objects as candidates for deletion even if their tombstone is expired.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
dba3e58dc5 [#1176] metabase: Ignore locked objects in IterateExpired
Make `DB.IterateExpired` to not pass locked objects to the handler. The
method is used by GC, therefore it will not consider them as candidates
for deletion.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
9508633a7e [#1175] metabase: Work with LOCK objects
After introduction of LOCK objects (of type `TypeLock`) complicated
extended its behavior:
  * create `lockers` container bucket (LCB) during PUT;
  * remove object from LCB during DELETE;
  * look up object in LCB during EXISTS;
  * get object from LCB during GET;
  * list objects from LCB during LIST with cursor;
  * select objects from LCB during SELECT with '*'.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
9f13674a10 [#1175] metabase: Implement LOCK operation
Implement `DB.Lock` method which marks list of the objects as locked by
another object. Only regular objects can be locked.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
14d27455f3 [#1175] metabase: Add container bucket for LOCKs
Create class of container buckets with `LOCKED` suffix. Put identifiers
of the objects of type `LOCK` to these buckets during `DB.Put`.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
e90ac9c6f9 [#1181] metabase: Remove unused CleanUp method
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-15 13:03:23 +03:00
Leonard Lyubich
7ccd1625af [#1214] *: Use single Object type in whole project
Remove `Object` and `RawObject` types from `pkg/core/object` package.
Use `Object` type from NeoFS SDK Go library everywhere. Avoid using the
deprecated elements.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-03-04 17:45:00 +03:00
Evgenii Stratonikov
19ad349b27 [#1204] shard: Save ID in the metabase
`AddShard` must return shard id, so we temporarily open metabase
there.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 17:13:13 +03:00
Evgenii Stratonikov
1fe9cd4d36 [#1186] engine: Allow to skip metabase in GetRange
Similarly to `Get`. Also fix a bug where `ErrNotFound` is returned
instead of `ErrRangeOutOfBounds`.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 16:07:25 +03:00
Evgenii Stratonikov
2b5550ccf6 [#1186] blobstor: Fix comment
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 16:07:25 +03:00
Evgenii Stratonikov
b33fb0f739 [#1186] blobstor: Unify errors for Get and GetRangeBig
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 16:07:25 +03:00
Evgenii Stratonikov
69e1e6ca20 [#1186] engine: Read object directly from blobstor in case of conflicts
Metabase is expected to contain actual information about objects stored
in shard. If the object is present in metabase but is missing from
blobstor, peform an additional attempt to fetch it directly without
consulting metabase. Such a situation is unexpected, so error counter
is increased for the shard which has the object in the metabase. We
don't increase error counter for the shard which has the object in
blobstor, because some garbage can be expected there. In this
implementation there is no overhead for objects which are really
missing, i.e. are not present in any metabase.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 16:07:25 +03:00
Evgenii Stratonikov
6e6f3648d2 [#1188] metabase: log PUT / DELETE operations
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 15:21:35 +03:00
Evgenii Stratonikov
ce76325840 [#1146] engine: remove TODO and add a comment
Parallel check seems to be slower, so remove TODO.
```
name \ time/op      old          new           pool
Exists/2_shards-40  52.3µs ±23%   91.6µs ± 6%   84.2µs ± 1%
Exists/4_shards-40  72.7µs ±11%  121.8µs ± 8%  116.9µs ± 5%

name \ alloc/op     old          new           pool
Exists/2_shards-40  5.00kB ± 0%   5.03kB ± 0%   5.16kB ± 0%
Exists/4_shards-40  9.89kB ± 0%   9.93kB ± 1%  10.19kB ± 0%

name \ allocs/op    old          new           pool
Exists/2_shards-40     112 ± 1%      115 ± 1%      117 ± 1%
Exists/4_shards-40     207 ± 1%      211 ± 1%      216 ± 1%
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 12:31:41 +03:00
Evgenii Stratonikov
dab6bcabe4 [#1146] engine: Add benchmarks for exists
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-03-04 12:31:41 +03:00
Evgenii Stratonikov
050a4bb2b0 [#1115] *: link TODOs to corresponding issues
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-11 12:58:59 +03:00
Evgenii Stratonikov
f9d3111825 [#1115] blobovnicza: remove GetRange method
It is unused and incorrect as described in FIXME.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-11 12:58:59 +03:00
Evgenii Stratonikov
c1530dec5e [#1115] shard/gc: provide parsed addresses in HandleExpiredTombstones
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-11 12:58:59 +03:00
Pavel Karpy
1667ec9e6d [#1131] *: Adopt SDK changes
`object.Address` has been moved to `object/address`
`object.ID` has been moved to `object/id`

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-02-08 09:45:38 +03:00
Evgenii Stratonikov
ed50cf6207 [#1118] services/control: return error counter in ListShards
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-03 15:14:27 +03:00
Evgenii Stratonikov
7fb15fa1d0 [#1118] services/control: allow to reset error counter in SetShardMode
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-03 15:14:27 +03:00
Evgenii Stratonikov
6ad2624552 [#1118] engine: allow to set error threshold
There are certain errors which are not expected during usual node
operation and which tell us that something is wrong with the shard.
To prevent possible data corruption, move shard in read-only mode after
amount of errors exceeded some threshold. By default no actions are performed.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-03 15:14:27 +03:00
Evgenii Stratonikov
674f520da7 [#1132] *: Use path/filepath package when working with files
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-02-03 10:12:07 +03:00
Evgenii Stratonikov
5a03a14940 [#1094] metabase: move Select filters sanity check outside of transaction
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:40 +03:00
Evgenii Stratonikov
651adf46c6 [#1094] blobovnicza: calculate size in Init properly
If pre-existing blobovnicza is initialized, it's size should be updated
even if all buckets are in place.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:40 +03:00
Evgenii Stratonikov
418b553920 [#1094] shard: remove TODO for shard mode description
It was added back in 2fb379b7 when we had many shard modes. Now we have
only two and comments for constants are rather descriptive.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:40 +03:00
Evgenii Stratonikov
ed7c732676 [#1094] shard: unify collection of expired objects
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:40 +03:00
Evgenii Stratonikov
18cfd8b042 [#1086] services/control: implement RestoreShard RPC
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:14 +03:00
Evgenii Stratonikov
0e60b1d6c9 [#1086] services/control: implement DumpShard RPC
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:14 +03:00
Evgenii Stratonikov
adb80bebb2 [#1085] shard: rename Evacuate to Dump
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
d06425c852 [#1085] shard: allow to ignore errors in Evacuate
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
36eebb5932 [#1085] writecache: allow to ignore errors during iteration
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
cd75638ce3 [#1085] blobstor: allow to ignore errors during iteration
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
c954f0e71b [#1085] fstree: allow to ignore errors during iteration
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
e53ad2f468 [#1085] blobovnicza: allow to ignore errors during iteration
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
9c60ab893c [#1085] shard: allow to ignore errors during restore
We could also ignore errors during evacuate, but this requires
unmarshaling objects first which slowers the process considerably.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
0ef3d5ab03 [#1085] writecache: persist in-memory objects before shutdown
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
3215210d0c [#1085] shard: Close components in reverse order
Write-cache should be closed before blobstor and metabase.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
b432ec3a03 [#1085] shard/test: add object integrity check for Evacuate
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
c0f6c988f0 [#1085] shard: dump data from write-cache
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
ad01aaf8bf [#1085] writecache: add read-only mode
In read-only mode modifying operations are immediately returned with
error and all background operations are suspended.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
9f963e001b [#1085] writecache: set metaOnly flag for receiving from priority channel
This can lead to objects being flushed twice.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Evgenii Stratonikov
7eedb23eb7 [#1085] shard: allow to dump/restore objects
Dump contains magic and a list of objects prefixed by object size in bytes.
We can't use proto-marshaled list because this requires having all dump
in memory. Using TAR induces 512 byte overhead for each object which can
be a problem in some cases.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-26 15:28:13 +03:00
Pavel Karpy
fee2f5a330 [#1100] *: Adopt new SDK's owner.ID API
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-01-21 17:58:30 +03:00
ZhangTao1596
ac82899e85 [#1088] writecache: Fix small flushed when persist
Signed-off-by: ZhangTao1596 <zhangtao@ngd.neo.org>
2022-01-19 12:55:33 +03:00
Evgenii Stratonikov
486d5c2e86 [#1060] writecache: compress big object if needed
Small objects use `blobstor.Put`, so no changes are required.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-13 14:20:36 +03:00
Evgenii Stratonikov
0d969d7a06 [#1060] blobstor: allow to disable compression based on content-type
For some data compression makes little sense, as it is already compressed.
This commit allows to leave such data unchanged based on `Content-Type`
attribute. Currently exact, prefix and suffix matching are supported.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-13 14:20:36 +03:00
Pavel Karpy
375394dc99 [#1059] shard: Add shard mode to shard Info
Provide shard mode information via `DumpInfo()`. Delete atomic field from
Shard structure since it duplicates new field.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-12-30 14:14:48 +03:00
Pavel Karpy
93bd6be743 [#1057] shard: Do not allow memory change operations in "read-only"
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-12-29 18:06:15 +03:00
Pavel Karpy
4f756bf121 [#1057] shard: Make shard to have only two mode
Shard's mode was not used in the Node, so added only two modes whose roles
are clear. More modes will be added in the future.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-12-29 18:06:15 +03:00
Pavel Karpy
284188f8f9 [#1048] node: Fill shard's info with its components' infos
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-12-28 13:21:44 +03:00
ZhangTao1596
1be0c42726 [#1022] writecache: Fix db counter decrement and flushed big object
Signed-off-by: ZhangTao1596 <zhangtao@ngd.neo.org>
2021-12-15 14:23:49 +03:00
Leonard Lyubich
fc42844fbf [#1028] writecache: Evict flushed big objects
Make `flushBigObjects` routine to call `evictObjects` which was
successfully written to `BlobStor`.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-14 13:49:47 +03:00
Leonard Lyubich
d246e5193b [#1028] writecache: Mark big objects as flushed after write
Make `flushBigObjects` routine to mark objects which are written to
`BlobStor`. This prevents already flushed objects from being written on
the next iterator tick.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-14 13:49:47 +03:00
Leonard Lyubich
90fd883e32 [#915] blobovnicza: Use mean bucket size multiplier instead of min
For fullness estimation of `Blobovnicza` we use number of object stored
in each size bucket. In previous implementation we multiplied the number
by the difference in bucket boundaries. This expression rather
estimated the minimum volume (and for the smallest bucket, the maximum)
of objects in the bucket.

Multiply number of objects by mean bucket size.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-08 14:24:49 +03:00
Leonard Lyubich
2920c5203b [#915] blobovnicza: Re-use Bolt transaction in syncFullnessCounter
Make `syncFullnessCounter` to accept `bbolt.Tx` argument of Bolt
transaction within which counter should be synchronized. Pass
corresponding transaction during `Init`.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-08 14:24:49 +03:00
Leonard Lyubich
f9bcb6f5e3 [#1007] metabase: Copy BoltDB value returned by IsSmall
According to BoltDB documentation bucket `value is only valid for the
life of the transaction`.

Make `DB.IsSmall` copy value slice in order to prevent potential memory
corruptions (e.g. `runtime.stringtobyteslice` cast).

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-07 18:59:42 +03:00
Leonard Lyubich
6f41cd86ed [#979] Upgrade SDK NeoFS Go dependency
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-12-01 17:31:04 +03:00
Alex Vanin
1f911830a5 [#948] engine: Define EndOfListing error in component packages
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
c80c83b0b8 [#948] metabase: Simplify cursor structure
Use bbolt bucket keys to seek offset in root bucket
and container buckets.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
468caa83d9 [#948] Adopt sdk-go client
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
08bdd0d561 [#948] engine: Fix comments of object listing with cursor
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
aa9ce8a853 [#948] engine: Define cursor for object listing as a type
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
164cd10af8 [#948] engine: Hide shard IDs in ListWithCursor
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
0f6d8f6eea [#948] metabase: Use seek for faster cursor listing
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
8d471c7e36 [#948] core/object: Move listing objects error to core
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
c02c7bee5b [#948] engine: Add ListWithCursor method
Interface for shard.ListWithCursor invocations.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
5b6be7bc1c [#948] engine: Fix test file path
Before: `<t.Name()>.<id>.blobstor`
After:  `<t.Name()>/<id>.blobstor`
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
1950724a04 [#948] shard: Add ListWithCursor method
Interface for metabase.ListWithCursor method invocation.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
1a829a521f [#948] metabase: Add ListWithCursor method
ListWithCursor allows listing physically stored objects
from metabase with small chunks. Cursor tracks last
processed object, therefore new chunks are returned
on each request.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-19 16:57:19 +03:00
Alex Vanin
cb36f8b50e [#868] blobstor: Speed up compression test
Compression test does not require wide blobovnicza tree.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-16 13:35:34 +03:00
Alex Vanin
e4087893a1 [#966] engine: Fix test after neofs-api-go replacement
Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-11-15 13:06:10 +03:00
Evgenii Stratonikov
95893927aa *: replace neofs-api-go with neofs-sdk-go
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-11-12 17:29:09 +03:00
Leonard Lyubich
6c0b29e3e3 [#922] storage engine: Prevent any operations after first Close call
Make `BlockExecution` / `ResumeExecution` to not release per-shard worker
pools. Make `StorageEngine.Close` to block these methods and any
data-related operations. It is still releases the pools.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Leonard Lyubich
d6457ee485 [#922] shard: Allow multiple gc.stop() calls
Use `sync.Once` to prevent locks of stopping GC. It will also allow to
safely call `Shard.Close` multiple  times.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Leonard Lyubich
10f0bd91d6 [#922] engine: Change interface of container operations
Add `error` to return. Improve docs.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Leonard Lyubich
a537334f33 [#922] engine: Fix typos and improve naming related to exec blocks
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Leonard Lyubich
ec04e787aa [#922] storage engine: Support operation blocking
There is a need to disable execution of local data operation on storage
engine in runtime. If storage engine ops are blocked, node will act like
always but all local object operations will be denied.

Implement `BlockExecution` / `ResumeExecution` methods on `StorageEngine`
which blocks / resumes the execution of data ops. Wait for the completion of
all operations executed at the time of the call. Return error passed to
`BlockExecution` from all data-related methods until `ResumeExecution` call.
Make `Close` to block operations as well.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Leonard Lyubich
cea1de3a27 [#922] storage engine: Remove unused Exists method
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-11-12 17:28:38 +03:00
Evgenii Stratonikov
1462824ab8 [#947] writecache: refactor object persisting
a1696a8 introduced some logic which in some situations prevented big objects
to be persisted in FSTree. In this commit a refactoring is done with the
goal of simplifying the code and also checking #866 issue.

1. Split a monstrous function into multiple simple ones: memory objects
   can only be small and for writing through the cache we can do a dispatch
   in `Put` itself.
2. Determine objects to be put in database before the actual update
   as setting up a transaction has non-zero overhead.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-11-01 15:52:56 +03:00
Alex Vanin
2271944064 [#945] metabase: Use cleanup instead of defer in tests
Cleanup is more idiomatic way to clean test data

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-10-27 16:10:01 +03:00
Alex Vanin
e41aba610d [#945] metabase: Fix containers listing
Container listing should not ignore tombstone and
storage group objects which are not stored in
primary buckets.

Signed-off-by: Alex Vanin <alexey@nspcc.ru>
2021-10-27 16:10:01 +03:00
Evgenii Stratonikov
0057eeb0f7 [#901] *: release worker pools where possible
Some of the pools are initialized during config initialization,
so it isn't possible currently to release them in one place.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-25 18:53:36 +03:00
Evgenii Stratonikov
10afd26354 [#901] engine: release shard pools in Close
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-25 18:53:36 +03:00
Pavel Karpy
616013cb8a [#933] write-cache: Use real number of objects in counters
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-10-25 18:49:10 +03:00
Pavel Karpy
29955a3036 [#933] FSTree: Add NumberOfObjects method
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-10-25 18:49:10 +03:00
Evgenii Stratonikov
1a1435be3d [#791] neofs-lens: inspect objects by address
```
> neofs-lens list --path ./blob/blobovnicza/1/1/0
6ay4GfhR9RgN28d5ufg63toPetkYHGcpcW7G3b7QWSek/9ibXu6v4uTwLEcME5vyHev6Zi8LpxqiWTe1dahKpAbb6
H3VBttoLQoknzMDgnVNyLZ8EpkDnQjnaxDr9fnAWeEHA/Hw1titdGh7BrTe2yLotiYbVh9FQaRRNhoNzXTyetpFgt

> neofs-lens inspect --path ./blob/blobovnicza/1/1/0 \
    --address 6ay4GfhR9RgN28d5ufg63toPetkYHGcpcW7G3b7QWSek/9ibXu6v4uTwLEcME5vyHev6Zi8LpxqiWTe1dahKpAbb6 \
    --header --out payload
Version: v2.1
Type: REGULAR
CID: 6ay4GfhR9RgN28d5ufg63toPetkYHGcpcW7G3b7QWSek
ID: 9ibXu6v4uTwLEcME5vyHev6Zi8LpxqiWTe1dahKpAbb6
Owner: 2dokPzmmcLnnR21jQB3qPppTQRgwMNMKEWD
CreatedAt: 0
PayloadSize: 32
Attributes:
  foo: bar

> hexdump -C payload
00000000  ff 6c d4 71 c4 83 f1 5f  b9 0b ad b3 7c 58 21 b6  |.l.q..._....|X!.|
00000010  d9 55 26 a4 1a 95 04 68  0b 4e 7c 8b 76 3a 1b 1d  |.U&....h.N|.v:..|
00000020
```

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-18 11:55:02 +03:00
Evgenii Stratonikov
123328a2f4 blobstor: close all opened blobovniczas
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-18 10:44:50 +03:00
Leonard Lyubich
1b3637aad7 [#884] shard: Mark expired tombstones as garbage
Mark expired tombstones as garbage in `Shard.HandleExpiredTombstones`.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-15 16:09:44 +03:00
Leonard Lyubich
3e4f0c1eb9 [#674] object/put: Handle job submission errors
Close wait channel on shard pool error in order to prevent deadlock.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-14 10:20:39 +03:00
Leonard Lyubich
43ccab3294 [#674] storage engine: Fix unit test failed on shard pool NPE
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-14 10:20:39 +03:00
Leonard Lyubich
5b1975d52a [#674] storage engine: Use per-shard worker pools for PUT operation
Make `StorageEngine` to use non-blocking worker pools with the same
(configurable) size for PUT operation. This allows you to switch to using
more free shards when overloading others, thereby more evenly distributing
the write load.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-14 10:20:39 +03:00
Evgenii Stratonikov
dd678cd976 [#868] blobstor: allow to decompress objects on-the-fly
We should be able to read whatever we have written earlier.
Compression setting applies only to the new objects.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-08 16:01:45 +03:00
Evgenii Stratonikov
cc377b34d2 [#868] blobstor: initialize (de-)compressors in Init
Do not log in options constructors. Also failure to
initialize compression module (possibly due to invalid options) is
certainly an error deserving proper treatment.

Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2021-10-08 16:01:45 +03:00
Leonard Lyubich
8a1593fdcc [#791] writecache: Export OpenDB function
The function will be useful for the tool which works with write-cache parts.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-05 10:51:10 +03:00
Leonard Lyubich
ca4c9d4673 [#791] writecache: Implement iterator over small objects
There is a need to list addresses of the small objects stored in WriteCache
database.

Implement `IterateDB` function which accepts BoltDB instance and iterate
over all saved objects and passes their addresses to the hander.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-05 10:51:10 +03:00
Leonard Lyubich
f73c5c2259 [#791] blobovnicza: Support read-only mode
There is a need to open Blobovnicza instances in read-only mode in some
cases.

Add `ReadOnly` option. Do not create dir path in RO. Open underlying BoltDB
instance with ReadOnly flag. Document thal all writing operations should not
be called in ro (otherwise BoltDB txs fail).

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-05 10:51:10 +03:00
Leonard Lyubich
cc5d3288a1 [#791] blobovnicza: Don't iterate over range buckets in Iterate
`Blobovnicza` can be initialized with any number of range buckets, and
reconstructed with different size limit. In previous implementation
`Iterate` could miss some stored objects if we construct `Blobovnicza` with
smaller number of ranges.

Make `Iterate` to traverse all buckets regardless of current instance
bounds.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-05 10:51:10 +03:00
Leonard Lyubich
5e2ca0d04b [#791] blobovnicza: Implement method to iterate over object addresses
In previous implementation `Blobovnicza.Iterate` op decoded object data only
and passed it to the handler. There is a need to iterate over all addresses
of the stored objects.

Add `DecodeAddresses` and `WithoutData` methods of `IteratePrm` type. Add
`Address` method to `IterationElement` type. Make `Iterate` to decode object
addresses if `DecodeAddress` was called and not read the data if
`WithoutData` was called. Implement `IterateAddresses` helper function to
simplify the code.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-10-05 10:51:10 +03:00
Leonard Lyubich
c96c455125 [#840] shard: Fix metabase Inhume parameters for tombstones
Replace `WithAddresses` with `WithTombstoneAddress` on `InhumePrm`.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-27 11:27:41 +03:00
Leonard Lyubich
14329ab565 [#840] metabase: Distinguish objects with tombstones and GC marks
Each object from graveyard has tombstone or GC mark. If object has
tombstone, metabase should return `ErrAlreadyRemoved` on object requests.
This is the case when user clearly removed the object from container. GC
marks are used for physical removal which can appear even if object is still
presented in container (Control service, Policer job, etc.). In this case
metabase should return 404 error on object requests.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-27 11:27:41 +03:00
Leonard Lyubich
02e6df683a [#840] shard: Return only physically stored objects from List
`List` method of `Shard` must return only physically stored objects.

Use `AddPhyFilter` to select only phy objects.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-27 11:27:41 +03:00
Leonard Lyubich
504abdda06 [#789] shard: Exclude object.ErrAlreadyRemoved errors in refillMetabase
Tombstone and "alive" objects can be both stored in BlobStor. They can
appear during iterating in different order. Metabase returns
`ErrAlreadyRemoved` error if object is inhumed.

Ignore `object.ErrAlreadyRemoved` errors of `metabase.Put`in Shard's
`refillMetabase` operation.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-15 18:22:09 +03:00
Leonard Lyubich
6bf7a00cfe [#789] shard: Add option to refill metabase on initialization
Add `WithRefillMetabase` option constructor which allows to set flag to
refill metabase.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-15 18:22:09 +03:00
Leonard Lyubich
b9c22e21b1 [#789] shard: Add method to refill the metabase
There is a need to refill Metabase data with the objects from BlobStor.

Implement `refillMetabase` method which iterates over all objects from
BlobStor and saves them in Metabase.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-15 18:22:09 +03:00
Leonard Lyubich
8d016d2529 [#789] blobstor: Implement iterator
There is a need to be able to process all objects saved in `BlobStor`.

Implement `BlobStor.Iterate` method which iterates over all objects.
Implement `IterateBinaryObjects` and `IterateObjects` helper functions to
simplify the code.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-09-15 18:22:09 +03:00