20180710 RNode community testing

Owned by Kelly Foster
Last updated: Jul 17, 2018 by Medha Parlikar (Unlicensed)
5 min read

Table of contents


Overview

Each week we invite community members to help test RNode.

  • When - Tuesdays at 14:00 UTC
  • Where - This meeting takes place online using Zoom
    • Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/134156866
    • Or iPhone one-tap :
          US: +16468769923,,134156866# or +16699006833,,134156866#
    • Or Telephone:
          Dial(for higher quality, dial a number based on your current location):
              US: +1 646 876 9923 or +1 669 900 6833
          Meeting ID: 134 156 866
          International numbers available: https://zoom.us/u/rDfBtpt
  • Additional communications are posted in #node-testing on the RChaing Discord.


What you can expect this week

We will test deploy and propose across the network. Testing will be similar to this demo.

Steps:

  1. Connect to the bootstrap and find your peers.

  2. Deploy a contract to your local node.

  3. Propose a block and observe the fork choice rule.


Configure your network

See RNode supported network configuration for information on how to configure your network to connect to other nodes.

Installation instructions

Updated 7/10/18, 12:53 UTC

Please install RNode prior to the start of the testing session. 

NOTE - At this time RNode does not run successfully in Docker on MacOSX. This is because you must run Docker with the host networking option.  By default, Docker runs in network bridge mode.  To use the host networking option, you must include the 'network=host' option  in your Docker run command. This flag does not work on MacOSX.

docker pull rchain/rnode:testing

Other requirements

Download and save the bonds.txt file

  • Download the file from https://repo.pyr8.io/rnode-test-sessions/bonds.txt

  • Save the file in a /genesis directory

  • Note - If you previously downloaded and save the bonds file for a different testing session, you need to make a change. Previously the system looked for the bonds file in the /validators directory. Changes in the software now supports looking for the bonds file in the /genesis directory. Please make sure the bonds file is stored in the correct directory.

Select a validator key

Empty the var/lib/rnode directory

  • If you have previously run RNode, we advise you empty this directory.

Connect to the bootstrap node 

rnode://c61769b39d368cbcbc9499634e030386c79d5b02@52.119.8.108:40400

Run instructions

rnode run -b rnode://c61769b39d368cbcbc9499634e030386c79d5b02@52.119.8.108:40400

Note the port is different than what is described in the network configuration dock.

Share your experience

You best support the improvement and development of the RChain platform when you file a bug to report challenge you faced or unsuccessful outcomes. This will help us collect all relevant information to better understand your setup and experience.



Testing session summary

  • Connections
    • Five peer connections
    • People who couldn't connect
      • Hang on key generation (known issue)
        • core-826
        •  Mateusz
          • about slow cert generation on linux. i "fix" this problem on my machine by install http://www.issihosts.com/haveged/ its additional enthropy source. on redhat/centos: `sudo yum install haveged -y && sudo systemctl enable haveged && sudo systemctl start haveged`ā€Øon ubuntu change yum to apt this is not safe. because random number generation will be far from ideal but ok for testing I think
          • Was able to connect after employing the fix described above
      • Node crash on startup java.lang.NullPointerException (core-821)
        • core-827
      • Paolo
        • my errors is node_1 | 14:35:36.172 [main] INFO coop.rchain.comm.connect.Connect$ - Bootstrapping from #{PeerNode c61769b39d368cbcbc9499634e030386c79d5b02}.ā€Ønode_1 | 14:35:36.190 [main] INFO coop.rchain.comm.connect.Connect$ - Initialize protocol handshake to #{PeerNode c61769b39d368cbcbc9499634e030386c79d5b02}ā€Ønode_1 | 14:35:36.442 [repl-io-32] INFO c.r.comm.transport.TcpTransportLayer - Current open connections: 1ā€Ønode_1 | 14:35:36.959 [grpc-default-executor-1] INFO c.r.comm.transport.TcpTransportLayer - Current open connections: 0ā€Ønode_1 | 14:35:36.965 [grpc-default-executor-1] WARN c.r.comm.transport.TcpTransportLayer - Failed to send a message to peer c61769b39d368cbcbc9499634e030386c79d5b02: UNAVAILABLE: io exception
          so then node_1 | Shutting down interpreter runtime ...ā€Ønode_1 | Shutting down Casper runtime ...ā€Ønode_1 | Goodbye.
      • Ed network port issues (ping not received for pong)
  • Deploy and propose
    • Jake received (success), deploy (success), show-blocks (success) propose (hang and no one else sees. See unable to propose blocks below.)
    • No response given, block not found messages (Henry and Lee)
      • core-828
      • From Will -
        • ubuntuāžœ ~ į… rnode proposeā€ØStarting with profile default
          It hang just there
          No response given since block not found
          "No response given since block not found"
        • Hangs on --show-blocks command too
        • core-823
      • From Piotr running on Digital Ocean
        • [grpc-default-executor-14] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block 4ed22df597... from #{PeerNode 98da91807d51cefafad93efed4843fa43d9c9cdf}. No response given since block not found. node_1 | 14:39:37.635 [grpc-default-executor-11] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block b87b9038f4... from #{PeerNode c61769b39d368cbcbc9499634e030386c79d5b02}. No response given since block not found. node_1 | 14:39:37.636 [grpc-default-executor-11] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block 4ed22df597... from #{PeerNode c61769b39d368cbcbc9499634e030386c79d5b02}. No response given since block not found. node_1 | 14:39:37.664 [grpc-default-executor-11] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block b87b9038f4... from #{PeerNode f7c786092ab60e56185fed57dcdb8bc81df9fe7f}. No response given sinc
      • From Mateusz
        • node_1 | 14:44:10.977 [grpc-default-executor-4] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block b87b9038f4... from #{PeerNode 98da91807d51cefafad93efed4843fa43d9c9cdf}. No response ā€Øgiven since block not found.ā€Ønode_1 | 14:44:10.978 [grpc-default-executor-4] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received request for block 4ed22df597... from #{PeerNode 98da91807d51cefafad93efed4843fa43d9c9cdf}. No response ā€Øgiven since block not found.
    • Lee
      • core-824
        • @Kelly 07:29:09.447 [grpc-default-executor-16] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Received Block with missing elements.ā€Ø07:29:09.447 [grpc-default-executor-16] WARN coop.rchain.casper.Validate$ - CASPER: Ignoring block b87b9038f4... because signature algorithm is unsupported.ā€Ø07:29:09.447 [grpc-default-executor-16] WARN coop.rchain.casper.Validate$ - CASPER: Ignoring block b87b9038f4... because block creator has 0 weight.ā€Ø07:29:09.448 [grpc-default-executor-16] INFO c.rchain.casper.util.comm.CommUtil$ - CASPER: Beginning request of missing block b87b9038f4... from peers...
      • hang
        • core-822
        • both the show-blocks command and the propose commands both cause my terminal to hang
        • when i use the deploy, propose and show-blocks commands I get errors starting with “WARNING: An illegal reflective access operation has occurred”
    • Unable to see proposed blocks
      • core-825 
      • everyone has a different genesis block hash
      • Medha's genesis block
        • blocks { blockHash: "b87b9038f42e4811de2d143efc583a36f98b6324a8c2f672ce2784ce5e05900a" blockSize: "11717" blockNumber: 0 version: 0 deployCount: 0 tupleSpaceHash: "" tupleSpaceDump: "Tuplespace hash not found!" timestamp: 0 faultTolerance: -1.0 mainParentHash: ""

      • Fix is clearing the data directory

      • Node assumes you're starting with a clear tuplespace. If you didn't shut down with an empty tuplespace, then node thinks the tuplespace is empty when it's not.

      • There may be another issues at play
      • It may also be a problem that the bootstrap was not cleared
      • Medha 
        • After 2 deploys and 2 proposes
        • "Tuplespace hash not found"




Bugs filed

key summary type created updated due assignee reporter priority status resolution
Refresh

Chat log

Participants

  • Henry Till
  • Lee Carter
  • Jake
  • Fror Dayan
  • Nuno Lopes
  • Will Qiu
  • Paolo DOnorio De Meo
  • Mateusz Probachta
  • Seth Bailey
  • Piotr Grzesik
  • Jerry
  • Ed Eykholt
  • Pawel Szulc
  • Sebastian Bach
  • Luigi Demeo
  • Deanna Duke
  • Alexander Guryev
  • Matthew Kaye
  • Stoica Daniel
  • Kent Shikama
  • Michael Birch