Commit Graph

163 Commits

Author SHA1 Message Date
Hector Sanjuan
5fb0753aed Issue #202: Make sure the ipfs mock repo/stat endpoint behaves like ipfs
This should fix tests.

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-10-26 16:01:41 +02:00
dgrisham
2d2a9da793 FreeSpace metric impl (including descendalloc) refactored and tested. 2017-08-11 13:57:42 -06:00
Hector Sanjuan
bb82c27b25 Fix #87: Implement ipfs-cluster-ctl pin ls <cid>
I have updated API endpoints to be /allocations rather than /pinlinst

It's more self-explanatory.

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-04-06 21:12:16 +02:00
Hector Sanjuan
18034356fa Fix #57: Hijack /add requests and pin items in cluster after ipfs adds them.
License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-30 11:27:40 +02:00
Hector Sanjuan
2bbbea79cc Issue #49: Add disk informer
The disk informer uses "ipfs repo stat" to fetch the RepoSize value and
uses it as a metric.

The numpinalloc allocator is now a generalized ascendalloc which
sorts metrics in ascending order and return the ones with lowest
values.

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-27 20:40:49 +02:00
Hector Sanjuan
4bb30cd24a Fixes #16: trigger ipfs swarm connect to other ipfs nodes in the cluster.
License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-27 12:42:54 +02:00
Hector Sanjuan
37fab27ba6 Move MapPinTracker to its own submodule and add tests for it
License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-14 16:10:45 +01:00
Hector Sanjuan
b57fa40f47 Add a couple of consensus tests to keep coverage high.
The code path is well tested, but Go does not know about it
in submodules.

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-14 13:50:30 +01:00
Hector Sanjuan
9b652bcfb3 Rename CidArg to Pin.
CidArg used to be an internal name for an argument that carried a Cid.
Now it has surfaced to API level and makes no sense. It is a Pin. It
represents a Pin (Cid, Allocations, Replication Factor)

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-03-08 16:57:27 +01:00
Hector Sanjuan
37046dc925 go vet fixes
License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-02-15 15:46:51 +01:00
Hector Sanjuan
8e45ce64c2 Documentation: captain log, readme and golint fixes
License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-02-15 15:46:51 +01:00
Hector Sanjuan
2512ecb701 Issue #41: Add Replication factor
New PeerManager, Allocator, Informer components have been added along
with a new "replication_factor" configuration option.

First, cluster peers collect and push metrics (Informer) to the Cluster
leader regularly. The Informer is an interface that can be implemented
in custom wayts to support custom metrics.

Second, on a pin operation, using the information from the collected metrics,
an Allocator can provide a list of preferences as to where the new pin
should be assigned. The Allocator is an interface allowing to provide
different allocation strategies.

Both Allocator and Informer are Cluster Componenets, and have access
to the RPC API.

The allocations are kept in the shared state. Cluster peer failure
detection is still missing and re-allocation is still missing, although
re-pinning something when a node is down/metrics missing does re-allocate
the pin somewhere else.

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-02-14 19:13:08 +01:00
Hector Sanjuan
0e7091c6cb Move testing mocks to subpackage so they can be re-used
Related to #18

License: MIT
Signed-off-by: Hector Sanjuan <hector@protocol.ai>
2017-02-09 17:51:19 +01:00