aboutsummaryrefslogtreecommitdiff
path: root/src/block.rs
Commit message (Collapse)AuthorAgeFilesLines
* Do not delete block if just a single replication error. Write TODO stuff.Alex Auvolat2020-04-221-10/+10
|
* Make UUID & Hash Copy and remove some .clone() noiseAlex Auvolat2020-04-211-8/+5
|
* Pretty loggingAlex Auvolat2020-04-211-9/+11
|
* Log which workers are doing whatAlex Auvolat2020-04-191-2/+4
|
* Implement repair object->version and version->block refAlex Auvolat2020-04-191-13/+1
|
* Implement repair commandAlex Auvolat2020-04-191-0/+84
|
* More aggressive sync timings & improve other stuffAlex Auvolat2020-04-191-2/+2
|
* Begin implement bucket management & admin commandsAlex Auvolat2020-04-191-2/+2
|
* Reorder imports.Alex Auvolat2020-04-181-0/+1
| | | | | | | 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
* Remove proto.rs & move some definitions out of data.rsAlex Auvolat2020-04-181-1/+11
|
* Massive RPC refactoringAlex Auvolat2020-04-181-51/+92
|
* Fix sync: use max root checksum levelAlex Auvolat2020-04-171-3/+3
|
* Fix add to resync on increfAlex Auvolat2020-04-171-5/+6
|
* Fix walk_ring_fromAlex Auvolat2020-04-171-1/+1
|
* Several resync workers; add delay on retry resyncAlex Auvolat2020-04-171-16/+15
|
* Fix paths :oAlex Auvolat2020-04-171-6/+8
|
* Implement sending blocks to nodes that need themAlex Auvolat2020-04-171-1/+60
|
* Don't send items...Alex Auvolat2020-04-171-7/+16
| | | | | ...if syncer doesn't need them because he's going to delete the partition anyway. Also, fix block resync queue
* Resync block on read errorAlex Auvolat2020-04-171-1/+8
|
* WIPAlex Auvolat2020-04-171-37/+63
| | | | | | TODOs: - ensure sync goes both way - finish sending blocks to other nodes when they need them before deleting
* Implement getting missing blocks when RC increasesAlex Auvolat2020-04-171-23/+140
| | | | | | | 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.
* Keep network status & ring in a tokio::sync::watchAlex Auvolat2020-04-111-12/+14
| | | | | | advantages - reads don't prevent preparing writes - can be followed from other parts of the system by cloning the receiver
* Local refcounting of blocksAlex Auvolat2020-04-111-29/+93
|
* Fix the Sync issue. Details:Alex Auvolat2020-04-101-5/+5
| | | | | | | 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.
* Why is it not Sync??Alex Auvolat2020-04-091-0/+49