Commit graph

33 commits

Author SHA1 Message Date
Pavel Karpy
2d7166f8d0 [] shard: Move NewEpoch event routing on SE level
It will allow dynamic shard management. Closing a shard does not allow
removing event handlers.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-10-04 10:08:55 +03:00
Pavel Karpy
431e331373 [] shard: Update metric counters
Use meta's operation results to change the metrics. Support typed object
counters.

Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-09-13 21:32:37 +04:00
Evgenii Stratonikov
339864b720 [] local_object_storage: Move shard.Mode to a separate package
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-07-21 17:56:06 +03:00
Evgenii Stratonikov
d8ba954aff [] shard: Use Set prefix for parameter setting
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-07-21 17:56:06 +03:00
Evgenii Stratonikov
34d319fed2 [] metabase: Use Set prefix for parameter setting
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-07-21 17:56:06 +03:00
Pavel Karpy
fed9e6679d [] node: Unlock locked object on its lock removal
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-07-18 11:42:25 +03:00
Pavel Karpy
6e752f36dc [] shard: Do not use pointers as parameters
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-06-03 07:35:17 +03:00
Pavel Karpy
e265ce2d52 [] meta: Do not use pointers as parameters
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2022-06-03 07:35:17 +03:00
Leonard Lyubich
1c30414a6c [] Upgrade NeoFS SDK Go module with new IDs
Core changes:
 * avoid package-colliding variable naming
 * avoid using pointers to IDs where unnecessary
 * avoid using `idSDK` import alias pattern
 * use `EncodeToString` for protocol string calculation and `String` for
  printing

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2022-06-01 17:41:45 +03:00
Pavel Karpy
2aec5736e0 [] 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 [] 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
daab30c391 [] 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
8107c8d1a9 [] 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 [] 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
6472a170eb [] 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
Leonard Lyubich
9dff07200c [] 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
Evgenii Stratonikov
c1530dec5e [] 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 [] *: 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
ed7c732676 [] shard: unify collection of expired objects
Signed-off-by: Evgenii Stratonikov <evgeniy@nspcc.ru>
2022-01-31 12:36:40 +03:00
Pavel Karpy
375394dc99 [] 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 [] 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
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
d6457ee485 [] 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
Evgenii Stratonikov
0057eeb0f7 [] *: 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
Leonard Lyubich
1b3637aad7 [] 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
Pavel Karpy
3d981320c6 [] gc: Stop GC work on Shard's Close
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
2021-08-05 14:44:05 +03:00
Leonard Lyubich
72f7df3339 [] shard/gc: Limit the accumulated batch of objects to be deleted
Change Shard's garbage remover to interrupt iterating over the metabase
graveyard when the buffer is full to the max size (`WithRemoverBatchSize`
Shard's option).

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
ff442e9559 [] shard/gc: Avoid usage of the range-loop vars in go-routines
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
717f2beb47 [] shard: Collect expired tombstones in GC every epoch
Add new epoch event handler to GC that finds all expired tombstones and
marks them and underlying objects to be removed. Shard uses callbacks
provided by the storage engine to mark underlying objects.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
dab45050b9 [] shard: Collect expired non-tombstone objects in GC every epoch
Add new epoch event handler to GC that finds all expired non-tombstone
objects and marks them to be removed.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
89a22450e5 [] shard: Define new epoch event
Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
a9a1acc880 [] shard: Control the completion of all handlers of the previous event
Group handlers of the particular event to a WaitGroup and wait for it before
the next event handling. This will ensure that all handlers complete and
prevent potential conflicts between past and present jobs.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00
Leonard Lyubich
13087dc3dd [] shard: Implement skeleton of internal GC
Shard's GC component consists of:
 * asynchronous remover that periodically wake up and removes all garbage
   objects from the shard, and goes to sleep for particular time interval;
 * external event listener that distributes jobs between workers;
 * group of workers that can handle a single job related to particular
   external event.

Remover and event listener represents go-routines which are started by
`init` method (calls from `Shard.Init`). In initial version all event
handlers are interrupted: this means that next event of the same type will
interrupt previous handling and start the new one.

GC is fully encapsulated in Shard. All GC configurations are reflected in
Shard's configuration.

Signed-off-by: Leonard Lyubich <leonard@nspcc.ru>
2021-02-19 11:56:32 +03:00