GitLab

Really deal with the deleted tag issue.

dep: update go-libp2p

Deal with renamed tags.

dep: update go-libp2p

fix: reduce memory used by buffered writers

Allocate them as-needed and use a pool.

Work towards #322.

query: fix a goroutine leak when the routing table is empty

When the routing table is empty, `Run` would fail but _not_ close the
process (leaking some query goroutines). This patch fixes this in multiple
places by:

1. Not starting queries with no peers.
2. Failing queries with no peers earlier.

* Expose metrics views in a slice

* Rename Views to DefaultViews

This communicates that these aren't the only views that are possible, or that they should be considered mandatory.

query: fix error "leak"

Long-running queries can build up large error sets that we never actually use.
This is exacerbated by https://github.com/libp2p/go-libp2p-swarm/pull/115.

fixes https://github.com/libp2p/go-libp2p-swarm/issues/119

providers: run datastore GC concurrently

Motivation: Walking the datastore can take time and currently blocks
adding/removing providers.

We need to do this in the same goroutine to avoid some logical races.

fix(providers): gc

Otherwise, we'll delete everything.

Remove the old protocol from the defaults

Add initial set of metrics exposed via OpenCensus.

Fix some provider subsystem performance issues

batches deletes

1. Don't be n^2.
2. Don't bother walking the cache, just drop it.

We only access it from a single goroutine.

This is just extra work as we write through anyways.

gomod: update go-libp2p

go-libp2p-peerstore now has a TempAddrTTL of 2 minutes

gomod: update libp2p to v0.0.4

dialQueue: start the control loop later

Don't start the control loop till we start the queue. The _in_ queue will still
work, we just won't start processing peers from it until we start the control loop.

fixes #311