阿里云ubuntu16.04运行Fabric1.2 e2e_cli

阿里云ubuntu16.04运行Fabric1.2 e2e_cli

  参加阿里云天池区块链比赛，在kubernetes运行fabric，但是总是遇到读不到块的问题，后打算，直接在阿里云服务器上先启动fabric网络，example中first network成功启动，但是运行e2e_cli时，遇到了一大堆坑
  不得不说，现在kubernetes和fabric都在快速变化，总是有各种问题，需要踩坑解决
  愿，坑越来越少

• Q1:阿里云CES服务器 WARNING: Connection pool is full, discarding connection: localhost

  • A:这是在阿里云CES服务器上遇到的，不影响运行结果，尚不清楚原因

• Q2:阿里云CES上运行e2e_cli，Ordering Service is not available

  • A:

    1. 对docker-compose file中orderer的文件挂在配置加上引号
    2. 注释阿里云服务器上/etc/resconv.conf中的option选项
    3. 对于所有的fabric容器，加上GODEBUG=netdns=go环境变量

  • ref:

    • 运行fabric第一个示例e2e_cli–genesis.block
    • 阿里云环境部署Hyperledger Fabric之SIGSEGV问题分析和解决经验分享
    • Hyperledger Fabric安装问题-阿里云环境问题

• Q3:org3查询出现Error: error endorsing query: rpc error

  • A: fabric1.2 bug,在e2e_cli/configtx.yaml org3配置下，添加’Org3MSP.member’
  • ref:
    
    • fabric1.2下e2e_cli问题记录
    • e2e_cli has failures when queried chaincode using idemix identity
      

• Q4:CRIT 06c Error creating configtx, Unexpected key Capabilities

  orderer.example.com    | 2018-09-11 12:25:00.079 UTC [orderer/multichain] newLedgerResources -> CRIT 06c Error creating configtx manager and handlers: Error deserializing key Capabilities for group /Channel: Unexpected key Capabilities
  orderer.example.com    | panic: Error creating configtx manager and handlers: Error deserializing key Capabilities for group /Channel: Unexpected key Capabilities
  orderer.example.com    |
  orderer.example.com    | goroutine 1 [running]:
  orderer.example.com    | panic(0xb31bc0, 0xc42000db00)

  • A: 之前因为Orderer服务失败的问题，重新拉取1.0.0版本的镜像，但是代码还是1.2，将镜像换回1.2，按Q2、Q3即可成功启动fabric1.2网络

---

以下是在阿里云CES ubuntu16.04上运行fabric1.2 e2e_cli网络遇到的各种问题及解决办法，最终all good

---

• Q.1:Ordering Service is not available

  Error: failed to create deliver client: orderer client failed to connect
  to orderer.example.com:7050: failed to create new connection: context 
  deadline exceeded
  !!!!!!!!!!!!!!! Ordering Service is not available, Please try again ... !!!!!!!!!!!!!!!!

  • 查询运行的容器发现没有orderer容器，但orderer单独拿出来作为orderer-test.yaml，查找问题

    • orderer-test.yaml

     #orderer-test.yaml
    version: '2'
    
    services:
      orderer.example.com:
        extends:
          file:   base/docker-compose-base.yaml
          service: orderer.example.com
       container_name: orderer.example.com

    • 运行orderer，查看日志

    root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli#  docker-compose -f order-test.yaml up
    ...
    orderer.example.com    | 2018-09-11 13:14:07.238 UTC [orderer/common/server] initializeLocalMsp -> CRIT 008 Failed to initialize local MSP: could not load a valid signer certificate from directory /var/hyperledger/orderer/msp/signcerts: stat /var/hyperledger/orderer/msp/signcerts: no such file or directory
    orderer.example.com exited with code 1

• Q.2:发现是因为genesis.block文件没有挂载上，查资料，解决方案是挂载目录加上引号

   root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli# vi base/docker-compose-base.yaml 
  ...
  volumes:
  - "../channel-artifacts/genesis.block:/var/hyperledger/orderer/orderer.genesis.block"
  - ../crypto-config/ordererOrganizations/example.com/orderers/orderer.example.com/msp:/var/hyperledger/orderer/msp
  - ../crypto-config/ordererOrganizations/example.com/orderers/orderer.example.com/tls/:/var/hyperledger/orderer/tls
  

• Q.3:再次运行orderer，查看日志，找不到签名证书的问题解决，但是出现了段信号错误，表现在runtime.cgocall

  orderer.example.com    | [signal SIGSEGV: segmentation violation code=0x1 addr=0x63 pc=0x7fa35c480259]
  orderer.example.com    | 
  orderer.example.com    | runtime stack:
  orderer.example.com    | runtime.throw(0xd89f2e, 0x2a)
  orderer.example.com    |    /opt/go/src/runtime/panic.go:619 +0x81
  orderer.example.com    | runtime.sigpanic()
  orderer.example.com    |    /opt/go/src/runtime/signal_unix.go:372 +0x28e
  orderer.example.com    | 
  orderer.example.com    | goroutine 56 [syscall]:
  orderer.example.com    | runtime.cgocall(0xa9bf80, 0xc4202715f8, 0x29)
  orderer.example.com    |    /opt/go/src/runtime/cgocall.go:128 +0x64 fp=0xc4202715b8 sp=0xc420271580 pc=0x402124
  orderer.example.com    | net._C2func_getaddrinfo(0xc420177d30, 0x0, 0xc4200a74d0, 0xc42000e040, 0x0, 0x0, 0x0)
  orderer.example.com 

  • 再次查找资料，发现这是因为阿里云新服务器域名解析配置/etc/resonv.conf多了一项，解决办法就是注释掉option这一项

    
    #之前在配置coredns时遇到过这个文件，并注意到多了一项，没想到这竟是服务失败的一个原因
    
    root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli# vi /etc/resolv.conf 
    ...
    nameserver 100.100.2.138
    nameserver 100.100.2.136
    
    #options timeout:2 attempts:3 rotate single-request-reopen
    
    

  • 详细原因及解决办法参考:阿里云环境部署Hyperledger Fabric之SIGSEGV问题分析和解决经验分享
  • 这样，在运行orderer，正常运行,并尝试链接kafka

• Q.4:启动e2e_cli网络，orderer服务正常，但是又出现连接不上域名的问题，即出现了与行解析问题

  ction error: desc = "transport: Error while dialing dial tcp: lookup peer0.org1.example.com on 127.0.0.11:53: no such host". Reconnecting...
  2018-09-11 13:42:39.574 UTC [grpc] Printf -> DEBU 043 pickfirstBalancer: HandleSubConnStateChange: 0xc4205f9710, TRANSIENT_FAILURE
  Error: error getting endorser client for install: endorser client failed to connect to peer0.org1.example.com:7051: failed to create new connection: context deadline exceeded
  !!!!!!!!!!!!!!! Chaincode installation on peer peer0.org1 has Failed !!!!!!!!!!!!!!!!
  ================== ERROR !!! FAILED to execute End-2-End Scenario ==================

  • 根据之前查找的资料，orderer异常是因为pure Go resolver解析方案不支持/etc/resconv.conf的option项，使fabric启动了cgo resolver解析方案，因cgo问题引起了段异常

  • 但是注释掉该项，出现了 no such host问题，根据资料对所有的fabric(cli、ca、orderer、peer)容器环境变量加上 GODEBUG=netdns=go，强制使用pure go resolver解析dns

    root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli# vi base/docker-compose-base.yaml 
    ...
    orderer.example.com:
    container_name: orderer.example.com
    image: hyperledger/fabric-orderer
    environment:
    - GODEBUG=netdns=go
    ...
    peer0.org1.example.com:
    container_name: peer0.org1.example.com
    extends:
    file: peer-base.yaml
    service: peer-base
    environment:
    - GODEBUG=netdns=go
    ...
    peer1.org1.example.com:
    container_name: peer1.org1.example.com
    extends:
    file: peer-base.yaml
    service: peer-base
    environment:
    - GODEBUG=netdns=go
    ...
    peer0.org2.example.com:
    container_name: peer0.org2.example.com
    extends:
    file: peer-base.yaml
    service: peer-base
    environment:
    - GODEBUG=netdns=go
    ...
    peer1.org2.example.com:
    container_name: peer1.org2.example.com
    extends:
    file: peer-base.yaml
    service: peer-base
    environment:
    - GODEBUG=netdns=go
    
    root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli# vi docker-compose-cli.yaml 
    ...
    cli:
    container_name: cli
    image: hyperledger/fabric-tools
    tty: true
    environment:
    - GODEBUG=netdns=go
    - GOPATH=/opt/gopath
    
    root@liu:~/work/go/src/github.com/hyperledger/fabric/examples/e2e_cli# vi docker-compose-e2e-template.yaml 
    ...
    ca0:
    image: hyperledger/fabric-ca
    environment:
    - GODEBUG=netdns=go
    ...
    ca1:
    image: hyperledger/fabric-ca
    environment:
    - GODEBUG=netdns=go
    

  • 启动网络./network_setup.sh up

    2018-09-11 14:05:14.773 UTC [bccsp] GetDefault -> WARN 02d Before using BCCSP, please call InitFactories(). Falling back to bootBCCSP.
    2018-09-11 14:05:14.773 UTC [msp] Sign -> DEBU 02e Idemix identity &{Org3MSP x: "\312\273\272\344u\344\270\3767\000\323\341H\016&\004\317Xj}\343\376rh\332X)qZ\335r\177"
    y: "\r\223SG\233\373\246\316\260\272\223\305\237\252\365\367\353T\244\323.\r3\353\373\177}\275S%\003\037"
    } is signing
    Error: error endorsing query: rpc error: code = Unknown desc = Failed evaluating policy on signed data during check policy on channel [mychannel] with policy [/Channel/Application/Writers]: [Failed to reach implicit threshold of 1 sub-policies, required 1 remaining] - proposal response: 
    ===================== Query successful on peer1.org3 on channel 'mychannel' ===================== 
    
    ===================== All GOOD, End-2-End execution completed ===================== 
     _____   _   _   ____            _____   ____    _____ 
    | ____| | \ | | |  _ \          | ____| |___ \  | ____|
    |  _|   |  | | | | | |  _____  |  _|     __) | |  _|  
    | |___  | |\  | | |_| | |_____| | |___   / __/  | |___ 
    |_____| |_| \_| |____/          |_____| |_____| |_____|
    

• Q.5:网络启动了，但是出现了Error: error endorsing query: rpc error

  Error: error endorsing query: rpc error: code = Unknown desc = Failed 
  evaluating policy on signed data during check policy on channel 
  [mychannel] with policy [/Channel/Application/Writers]: [Failed to reach
  implicit threshold of 1 sub-policies, required 1 remaining] - proposal
  response: 
  ===================== Query successful on peer1.org3 on channel 
  'mychannel' ===================== 

  • 这是fabric1.2的bug，新版本已经fix，解决办法是修改configtx.yaml，为org3添加成员信息签名信息Org3MSP.member

    - &Org3
     # Name of the organization
    Name: Org3MSP
    
     # ID to load the MSP definition as
    ID: Org3MSP
    
     # Type of MSP - this org uses idemix for its MSP implementation
    MSPType: idemix
    
    MSPDir: crypto-config/idemix/idemix-config
    
     # Policies defines the set of policies at this level of the config tree
     # For organization policies, their canonical path is usually
     #   /Channel///
    Policies:
        Readers:
            Type: Signature
            Rule: "OR('Org3MSP.admin', 'Org3MSP.peer', 'Org3MSP.client','Org3MSP.member')"
        Writers:
            Type: Signature
            Rule: "OR('Org3MSP.admin', 'Org3MSP.client','Org3MSP.member')"
        Admins:
            Type: Signature
            Rule: "OR('Org3MSP.admin')"
    

  • 再次启动网络，查询正确

    2018-09-11 12:51:09.444 UTC [msp] Sign -> DEBU 02e Idemix identity &{Org3MSP x: "\304]\266ds\225u\003\r\337\306\312\335Q\230\025\232C-\337\3123\021t\262\275\263\323Hx\206\004"
    y: "\\\023\3004\033;\314\340\216s\350\307C\257\204\317[\226Ls\013\365\335'\211\322P\221\300\361\336\224"
    } is signing
    90
    ===================== Query successful on peer1.org3 on channel 'mychannel' ===================== 

• Ref

  • 运行fabric第一个示例e2e_cli
  • Hyperledger Fabric安装问题-阿里云环境问题
  • 阿里云环境部署Hyperledger Fabric之SIGSEGV问题分析和解决经验分享
  • fabric1.2下e2e_cli问题记录
  • e2e_cli has failures when queried chaincode using idemix identity
  • Hyperledger Fabric安装问题-证书生成时签名key不兼容问题
  • 搭建HyperLedger Fabric1.2.0开发环境