Refs #3419

In order to re-enable the test harness for the KMS (see
tendermint/kms#227), we need some marginally more realistic proposals
and votes. This is because the KMS does some additional sanity checks
now to ensure the height and round are increasing over time.

Fixes #3444

* comments on validator ordering

* NextValidatorsHash

* blockchain: update the maxHeight when a peer is removed

Refs #2699

* add a changelog entry

* make linter pass

* p2p: refactor Switch#Broadcast func

- call wg.Add only once
- do not call peers.List twice!
  * bad for perfomance
  * peers list can change in between calls!

Refs #3306

* p2p: use time.Ticker instead of RepeatTimer

no need in RepeatTimer since we don't Reset them

Refs #3306

* libs/common: remove RepeatTimer (also TimerMaker and Ticker interface)

"ancient code that’s caused no end of trouble" Ethan

I believe there's much simplier way to write a ticker than can be reset
https://medium.com/@arpith/resetting-a-ticker-in-go-63858a2c17ec

* Update proposer-selection.md

* Fixed typos

* fixed typos

* Attempt to address some comments

* Update proposer-selection.md

* Update proposer-selection.md

* Update proposer-selection.md

Added the normalization step.

* Addressed review comments

* New example for normalization section

Added a new example to better show the need for normalization
Added requirement for changing validator set
Addressed review comments

* Fixed problem with R2

* fixed the math for new validator

* test

* more small updates

* Moved the centering above the round-robin election

- the centering is now done before the actual round-robin block
- updated examples
- cleanup

* change to reflect new implementation for new validator

Merge master back to develop

Release/v0.31.0

* Make sure config.TimeoutBroadcastTxCommit < rpcserver.WriteTimeout()

* remove redundant comment

* libs/rpc/http_server: move Read/WriteTimeout into Config

* increase defaults for read/write timeouts

Based on this article
https://www.digitalocean.com/community/tutorials/how-to-optimize-nginx-configuration

* WriteTimeout should be larger than TimeoutBroadcastTxCommit

* set a deadline for subscribing to txs

* extract duration into const

* add two changelog entries

* Update CHANGELOG_PENDING.md

Co-Authored-By: melekes <anton.kalyaev@gmail.com>

* Update CHANGELOG_PENDING.md

Co-Authored-By: melekes <anton.kalyaev@gmail.com>

* 12 -> 10

* changelog

* changelog

Merge develop into release/0.31.0

See https://github.com/tendermint/tendermint/pull/3403/files#r266208947

In #3403 we unexposed BlockTimeIota from the ABCI, but it's still part
of the ConsensusParams struct, so we have to remember to add it back
after calling PB2TM.ConsensusParams. Instead, PB2TM.ConsensusParams
should take it as an argument

Fixes #3432

* changelog: fix formatting

* update release notes

* update changelog

* linkify

* update UPGRADING

* Update release summary

* Add pubsub config changes

* Add link to issue for pubsub changes

* bump ABCIVersion due to renaming BlockSizeParams -> BlockParams
(https://github.com/tendermint/tendermint/pull/3417#discussion_r264974791)

* Move changelog on consensus params entry to breaking

* Add @melekes' suggestion for breaking change in pubsub into upgrading.md

* Add changelog entry for #3351

* Add changelog entry for #3358 & #3359

* Add changelog entry for #3397

* remove changelog entry for #3397 (was already released in 0.30.2)

* move 3351 to improvements

* Update changelog comment

Fixes #3402

Fixes #3421

Before: it was creating a file named ${LOG:-tendermint.log} in .build/nodeX
After: it creates a file named tendermint.log

- update changelog, reset pending
 - bump versions
 - add external contributors (partly manually)

Merge master back to develop (do not squash)

* libs/db: close batch (#3397)

ClevelDB requires closing when WriteBatch is no longer needed, https://godoc.org/github.com/jmhodges/levigo#WriteBatch.Close

Fixes the memory leak in https://github.com/cosmos/cosmos-sdk/issues/3842

* update changelog and bump version to 0.30.2

there is no such file https://storage.googleapis.com/golang/go1.12.0.linux-amd64.tar.gz

Fixes #3405

Fixes #3408

* limit number of /subscribe clients and queries per client

Add the following config variables (under [rpc] section):
  * max_subscription_clients
  * max_subscriptions_per_client
  * timeout_broadcast_tx_commit

Fixes #2826

new HTTPClient interface for subscriptions

finalize HTTPClient events interface

remove EventSubscriber

fix data race

```
WARNING: DATA RACE
Read at 0x00c000a36060 by goroutine 129:
  github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe.func1()
      /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:168 +0x1f0

Previous write at 0x00c000a36060 by goroutine 132:
  github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe()
      /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:191 +0x4e0
  github.com/tendermint/tendermint/rpc/client.WaitForOneEvent()
      /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178
  github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1()
      /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298
  testing.tRunner()
      /usr/local/go/src/testing/testing.go:827 +0x162

Goroutine 129 (running) created at:
  github.com/tendermint/tendermint/rpc/client.(*Local).Subscribe()
      /go/src/github.com/tendermint/tendermint/rpc/client/localclient.go:164 +0x4b7
  github.com/tendermint/tendermint/rpc/client.WaitForOneEvent()
      /go/src/github.com/tendermint/tendermint/rpc/client/helpers.go:64 +0x178
  github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync.func1()
      /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:139 +0x298
  testing.tRunner()
      /usr/local/go/src/testing/testing.go:827 +0x162

Goroutine 132 (running) created at:
  testing.(*T).Run()
      /usr/local/go/src/testing/testing.go:878 +0x659
  github.com/tendermint/tendermint/rpc/client_test.TestTxEventsSentWithBroadcastTxSync()
      /go/src/github.com/tendermint/tendermint/rpc/client/event_test.go:119 +0x186
  testing.tRunner()
      /usr/local/go/src/testing/testing.go:827 +0x162
==================
```

lite client works (tested manually)

godoc comments

httpclient: do not close the out channel

use TimeoutBroadcastTxCommit

no timeout for unsubscribe

but 1s Local (5s HTTP) timeout for resubscribe

format code

change Subscribe#out cap to 1

and replace config vars with RPCConfig

TimeoutBroadcastTxCommit can't be greater than rpcserver.WriteTimeout

rpc: Context as first parameter to all functions

reformat code

fixes after my own review

fixes after Ethan's review

add test stubs

fix config.toml

* fixes after manual testing

- rpc: do not recommend to use BroadcastTxCommit because it's slow and wastes
Tendermint resources (pubsub)
- rpc: better error in Subscribe and BroadcastTxCommit
- HTTPClient: do not resubscribe if err = ErrAlreadySubscribed

* fixes after Ismail's review

* Update rpc/grpc/grpc_test.go

Co-Authored-By: melekes <anton.kalyaev@gmail.com>

Also

- init substructures to avoid panic in pb2tm.ConsensusParams
Before: if csp.Block is nil and we later try to access/write to it,
we'll panic.
After: if csp.Block is nil and we later try to access/write to it,
there'll be no panic.

When scaling and averaging is invoked, it is possible to have validators
with close priorities ending up with same priority. With the current code,
this  makes it impossible to verify the priority orders before and after updates.

Fixes #3383

Release/v0.30.2

Fixes #3369

ClevelDB requires closing when WriteBatch is no longer needed, https://godoc.org/github.com/jmhodges/levigo#WriteBatch.Close

Fixes the memory leak in https://github.com/cosmos/cosmos-sdk/issues/3842

ClevelDB requires closing when WriteBatch is no longer needed, https://godoc.org/github.com/jmhodges/levigo#WriteBatch.Close

Fixes the memory leak in https://github.com/cosmos/cosmos-sdk/issues/3842

Refs #3262

This fixes two small bugs:

1) lite/dbprovider: return `ok` instead of true in parse* functions. It's weird that we're ignoring `ok` value before.
2) consensus/state: previously because of the shadowing we almost never output "Error with msg". Now we declare both `added` and `err` in the beginning of the function, so there's no shadowing.

Update Gopkg.lock via dep ensure --update golang.org/x/crypto

see #3391 (comment) (nothing to review here really).

Although the version we were pinning to is from Nov. 2016 there were no substantial changes:

    jmhodges/levigo@2b8c778 added go-modules support (no code changes)
    jmhodges/levigo@853d788 added a badge to the readme

closes #3381

We're pinning repos without releases because it's very easy to upgrade all the dependencies by executing dep ensure --upgrade. Instead, we should just never run this command directly, only dep ensure --upgrade <some repo>. And we can defend that in PRs.

Refs #3374

The problem with pinning to exact revisions: people who import Tendermint as a library (e.g. abci/types) are stuck with these revisions even though the code they import may not even use them.

https://tools.ietf.org/html/rfc6962#section-2.1
"The largest power of two less than the number of items" is actually correct!

    For n > 1, let k be the largest power of two smaller than n (i.e.,
    k < n <= 2k).

Refs #3262

* fix failure in TestProposerFrequency

* Add test to check priority order after updates

* Changed applyRemovals() and removed Remove()

Changed applyRemovals() similar to applyUpdates()
Removed function Remove()
Updated comments

* review comments

* simplify applyRemovals and add more comments

* small correction in comment

* Fix check in test

* Fix priority check for centering, address review comments

* fix assert for priority centering

* review comments

* review comments

* cleanup and review comments

added upper limit check for validator voting power
moved check for empty validator set earlier
moved panic on potential negative set length in verifyRemovals
added more tests

* review comments

downstream

Signed-off-by: Silas Davis <silas@monax.io>

- update docker image on circleci
 - remove GOCACHE=off from Makefile (see:
 https://tip.golang.org/doc/go1.12#gocache)
 - update badge in readme
 - update in scripts/install
 - update Vagrantfile
 - update in networks/remote/integration.sh
 - tools/build/Makefile

Fixes: #3378

* Add stats to cleveldb implementation

* update changelog

* remote TODO

also
- sort keys
- preallocate memory

* fix const initializer []string literal is not a constant

* add test

* make BlockTimeIota a consensus parameter, not a locally configurable option

Refs #2920

* make TimeIota int64 ms

Refs #2920

* update Gopkg.toml

* fixes after Ethan's review

* fix TestRemoteSignerProposalSigningFailed

* update changelog

Before we're using it to get a round state in tests. Now it can be done
by calling csX.GetRoundState. We will need to rewrite
TestStateSlashingPrevotes and TestStateSlashingPrecommits, which are
commented right now, to not rely on EventDataRoundState#RoundState
field.

Refs #1527