How to dfficiently re-start network

Niklaskkkk

Hi,
In lab1 one uses the bootstrap script to start. Later on in the Lab there are docker-compose yml up commands, with new features like couchDB, chaincode, new peers etc.

Sometimes you have to restart the network (turned computer off, or edited yml-files). What is the common way to re-start your network? Do you start back at the bootstrap script, or how is this done by you?

maniankara

Yes, I have also been wondering how to get back to the same state in a Hyperledger Fabric cluster? E.g. if you have couple of Chaincode instantiated, then there are some instances of containers running. How is this dealt in the production? I am sure there is easier way for this.

Niklaskkkk

For those wondering the same, I just tweaked the start.sh & stop.sh scrips in the startFiles folder. Here you can decide which containers to start up so you dont have to type these commands manually each time you reboot computer.

Niklaskkkk

However, as @maniankara asks, what about the created channels & chaincodes? Once I run docker-compose yml down & then up again (for instance if I needed to regenerate cryptomaterial), then suddenly all channels & chaincode needs to be re-installed again? If I run peer channel list the peers are not part of channels anymore....however, the artifacts are obviously still in the config folder.
So it seems this test network, the orderer, do not save its state?

ASAPBLOCKY

Yeah I asked myself the same question. Is there are special command to save the current state of Fabric? How are we supposed to do it in a production environment if the computer just shuts down? Do we have to start all over again?

Niklaskkkk

Currently I just edited the bootstrap.sh to at least quickly bring up different containers, channels & chaincode. It is not ideal, but at least better than doing it all over again:

!/bin/sh

#

Copyright IBM Corp All Rights Reserved

#

SPDX-License-Identifier: Apache-2.0

#
export PATH=$GOPATH/src/github.com/hyperledger/fabric/build/bin:${PWD}/../bin:${PWD}:$PATH
export FABRIC_CFG_PATH=${PWD}
CHANNEL_NAME=allarewelcome

remove previous crypto material and config transactions

rm -fr config/*

rm -fr crypto-config/*

generate crypto material

cryptogen generate --config=./crypto-config.yaml

if [ "$?" -ne 0 ]; then

echo "Failed to generate crypto material..."

exit 1

fi

generate genesis block for orderer

configtxgen -profile OneOrgOrdererGenesis -outputBlock ./config/genesis.block
if [ "$?" -ne 0 ]; then
echo "Failed to generate orderer genesis block..."
exit 1
fi

generate channel configuration transaction

configtxgen -profile OneOrgChannel -outputCreateChannelTx ./config/allarewelcome.tx -channelID allarewelcome
if [ "$?" -ne 0 ]; then
echo "Failed to generate channel configuration transaction..."
exit 1
fi

generate anchor peer transaction

configtxgen -profile OneOrgChannel -outputAnchorPeersUpdate ./config/Org1MSPanchors.tx -channelID allarewelcome -asOrg Org1MSP
if [ "$?" -ne 0 ]; then
echo "Failed to generate anchor peer update for Org1MSP..."
exit 1
fi

sleep 5s

set -ev

don't rewrite paths for Windows Git Bash users

export MSYS_NO_PATHCONV=1

docker-compose -f docker-compose.yml down

docker-compose -f docker-compose.yml up -d zookeeper0.example.com zookeeper1.example.com zookeeper2.example.com kafka0.example.com kafka1.example.com

sleep 60s

docker-compose -f docker-compose.yml up -d Org1ca.example.com Org2ca.example.com orderer.example.com peer1.org1.example.com couchdbOrg1Peer1 couchdbOrg1Peer0 couchdbOrg2Peer0 couchdbOrg2Peer1 peer0.org1.example.com peer0.org2.example.com peer1.org2.example.com cli

wait for Hyperledger Fabric to start

incase of errors when running later commands, issue export FABRIC_START_TIMEOUT=

export FABRIC_START_TIMEOUT=10

echo ${FABRIC_START_TIMEOUT}

sleep ${FABRIC_START_TIMEOUT}

Create the channel

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/etc/hyperledger/msp/users/[email protected]/msp" peer0.org1.example.com peer channel create -o orderer.example.com:7050 -c allarewelcome -f /etc/hyperledger/configtx/allarewelcome.tx

Join peer0.org1.example.com to the channel.

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/etc/hyperledger/msp/users/[email protected]/msp" peer0.org1.example.com peer channel join -b allarewelcome.block

Peer1.org1 fetch genesis block from the channel before joining

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/etc/hyperledger/msp/users/[email protected]/msp" peer1.org1.example.com peer channel fetch oldest allarewelcome.block -c allarewelcome --orderer orderer.example.com:7050

Join peer1.org1.example.com to the channel.

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/etc/hyperledger/msp/users/[email protected]/msp" peer1.org1.example.com peer channel join -b allarewelcome.block

Install chaincode for peer0.org1 version 1.1 through CLI

docker exec -e "CORE_PEER_ADDRESS=peer0.org1.example.com:7051" -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org1.example.com/users/[email protected]/msp" cli peer chaincode install -n ccForAll -v 1.1 -p github.com/sacc

Instantiate chaincode 1.1 on peer0.org1

docker exec -e "CORE_PEER_ADDRESS=peer0.org1.example.com:7051" -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org1.example.com/users/[email protected]/msp" cli peer chaincode instantiate -n ccForAll -v 1.1 -o orderer.example.com:7050 -C allarewelcome -c '{"Args":["Mach","50"]}' --policy "AND('Org1.peer', OR ('Org1.member'))"

Sleep 10 sec for cc to install

sleep 10s

Try to sign config block for Org2 to join network through CLI

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org1.example.com/users/[email protected]/msp" -e "CORE_PEER_ADDRESS=peer0.org1.example.com:7051" cli peer channel signconfigtx -f org2submitReady.pb

Send config changes to rest of network for approval

docker exec -e "CORE_PEER_LOCALMSPID=Org1MSP" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org1.example.com/users/[email protected]/msp" -e "CORE_PEER_ADDRESS=peer0.org1.example.com:7051" cli peer channel update -f org2submitReady.pb -c allarewelcome -o orderer.example.com:7050

Fetch config block allarewelcome for Peer0.org2

docker exec -e "CORE_PEER_LOCALMSPID=Org2MSP" -e "CORE_PEER_ADDRESS=peer0.org2.example.com:7051" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org2.example.com/users/[email protected]/msp" cli peer channel fetch config Org2AddedConfig.pb -o orderer.example.com:7050 -c allarewelcome

Join channel allarewelcome for peer0.org2

docker exec -e "CORE_PEER_LOCALMSPID=Org2MSP" -e "CORE_PEER_ADDRESS=peer0.org2.example.com:7051" -e "CORE_PEER_MSPCONFIGPATH=/opt/gopath/src/github.com/hyperledger/fabric/peer/crypto/peerOrganizations/org2.example.com/users/[email protected]/msp" cli peer channel join -b allarewelcome.block

kmyatt

Hi guys, So there seems to be a lot of variations of the same question. So I'm going to try and condense them all into one and then answer that. The question seems to be

"Why are we using docker compose to restart the network every time we make docker-compose.yml config updates"

While this is a good question, I think we are missing the point. Docker compose is not a production grade tool, and should not be used in production grade environments. It is an amazing tool for most things leading up to it, development, testing etc. So we teardown and reboot the network with little to no effort. It's also what's used in conjunction with the docker-compose.yml file. It is a must learn/have for creating an environment for experimenting and learning for students (as well as development purposes).

How are we supposed to do it in a production environment if the computer just shuts down? Do we have to start all over again?

This question conflicts with the concept of decentralization. Remember that Fabric is not just running on a local machine, yet it's a network of hosts. So one host disconnecting from the network should not have any dire effect on the actual network itself (at Prod level).

In regard to the Bootstrap file...
The bootstrap file has nothing to do with the actual course. It was created only to jumpstart you into an preexisting, running (yet extremely basic) network, in which we scale from there. Everything done in that file is minuscule compared to what you actually do and learn throughout the labs, so I recommend to stop focusing on that, and more on that labs, because if you pay close attention to the labs, you will learn all of that anyway.