* Start upgrade. Currently go test ./... hangs.
* (squash this) Fix staking tests
* wip
* note what changes need to be made to make this work on tm v0.23.0
* Fix addr -> pubkey map
* cleanup code
* Fix slashing test failures except for begin blocker
* fix all slashing tests
* fix lcd tests
* Address PR comments
* add link to changelog.
* (wip) start making addrToPubkey map persisted. Since amino can't handle maps,
we have to change from what this commit is doing.
* Use the correct method of storing a map
* (squash this) address PR comments
* Did you run 'make'?
* remove gaiadebug binary
Currently the crypto/keys tests take 2 minutes in circle CI.
A significant portion of this time is due to the bcrypt security
parameter. Changing it for these tests should reduce the time significantly.
* Update to tendermint v0.22.6-rc0
This is comprised of updating the crypto imports / API
* (squash this) switch to v0.22.6
If this passes tests, I'll squash this commit and update the PR.
* tools: Ensure Gopkg.lock is correct in linting
This adds dep status to the lint process. Also fixes linting errors
that existed earlier. (not sure why they didn't show up on CI)
Closes#1574
* Update dep, use the lock file new dep version creates
This is done so that the time spent on bcrypt during test cases
can be reduced. This change reduces the amount of time lcd tests
spend on bcrypt from 76% to 40%. (We need to reduce the number of
calls to bcrypt in a seperate PR, along with fixing other sources
of slowness)
Making the bcrypt security parameter a var shouldn't be a security issue:
One can't verify an invalid key by maliciously changing the bcrypt
parameter during a runtime vulnerability. The main security
threat this then exposes would be something that changes this during
runtime before the user creates their key. This vulnerability must
succeed to update this to that same value before every subsequent call
to gaiacli keys in future startups / or the attacker must get access
to the filesystem. However, with this same threat model (changing
variables in runtime), one can cause the user to sign a different tx
than what they see, which is a significantly cheaper attack then breaking
a bcrypt hash. (Recall that the nonce still exists to break rainbow
tables)
Merges the keybase and Ledger code from go-crypto (which is no more) into the SDK
Adds support for Ledger into gaiacli
Cherry-picks updated error handling from #1158
Aleksandr Bezobchuk
Dev Ojha
Alexander Bezobchuk
ValarDragon
Ethan Buchman
Christopher Goes