Hyperledger 2.2 Peer Out of Sync issue #fabric #fabric-kubernetes #fabric-sdk-node #fabric-endorser #fabric-peer


Avinash
 

Hi Fabric Experts,

Currently I am upgrading our network from HLF 1.4 to 2.2.
As our current HLF network using docker and kubernetes we decided to use below approach for the migration.

  1. Identify the last block using peer channel fetch
  2. Clone all the transactions blocks using peer channel fetch
  3. Convert these blocks to JSON
  4. Use the JSON objects to execute transactions using a fabric network.

We are running both HLF 1.4 and 2.2 network during the migration.
While transferring the transaction blocks network handling up to 65K transactions after that both peers going out of sync and maintaining different balance.
65K max I can reach so far sometime peer going out of sync as early as 5K transactions.

HLF 2.2 network Configuration:

  • One Org
  • 2 Peer
  • 3 Orderer

Fabric Network 2.2.5
fabric-ca-client 2.2.5
 
Please let me know if any one faced the similar issue and best way to solve the issue.


David Enyeart
 

Do you mean the existing v1.4 peers are out of sync?
Or you are re-processing all transactions into a new network and the new network peers are out of sync with each other?
Or the new peers are out of sync relative to the old peers?
Which state database?
Why don't you do an inline upgrade of the peers rather than re-process the tranactions?
Can you identify the first transaction that is out of sync and look at the logs from that block height.


Dave Enyeart


avinash.sachdewani2010---07/28/2021 11:48:00 PM---Hi Fabric Experts, Currently I am upgrading our network from HLF 1.4 to 2.2.

From: avinash.sachdewani2010@...
To: fabric@...
Date: 07/28/2021 11:48 PM
Subject: [EXTERNAL] [Hyperledger Fabric] Hyperledger 2.2 Peer Out of Sync issue #fabric #fabric-peer #fabric-sdk-node #fabric-endorser #fabric-kubernetes
Sent by: fabric@...





Hi Fabric Experts, Currently I am upgrading our network from HLF 1.4 to 2.2. As our current HLF network using docker and kubernetes we decided to use below approach for the migration. Identify the last block using peer channel fetch ‍‍‍‍‍‍‍ZjQcmQRYFpfptBannerStart 
This Message Is From an External Sender
This message came from outside your organization.
ZjQcmQRYFpfptBannerEnd

Hi Fabric Experts,

Currently I am upgrading our network from HLF 1.4 to 2.2.
As our current HLF network using docker and kubernetes we decided to use below approach for the migration.

1. Identify the last block using peer channel fetch
2. Clone all the transactions blocks using peer channel fetch
3. Convert these blocks to JSON
4. Use the JSON objects to execute transactions using a fabric network.
We are running both HLF 1.4 and 2.2 network during the migration.
While transferring the transaction blocks network handling up to 65K transactions after that both peers going out of sync and maintaining different balance.
65K max I can reach so far sometime peer going out of sync as early as 5K transactions.

HLF 2.2 network Configuration:

  • One Org
  • 2 Peer
  • 3 Orderer
    Fabric Network 2.2.5
    fabric-ca-client 2.2.5

    Please let me know if any one faced the similar issue and best way to solve the issue.






  • David Enyeart
     

    To close the loop on this one... we found that two peers were configured to use a single CouchDB database, and therefore the peers were stepping on each other and getting inconsistency between peer data and CouchDB data. We will make it more clear in the docs that each peer must have their own dedicated CouchDB database.


    Dave Enyeart


    David Enyeart---07/29/2021 07:27:07 AM---Do you mean the existing v1.4 peers are out of sync? Or you are re-processing all transactions into

    From: David Enyeart/Durham/IBM
    To: avinash.sachdewani2010@...
    Cc: fabric@...
    Date: 07/29/2021 07:27 AM
    Subject: Re: [Hyperledger Fabric] Hyperledger 2.2 Peer Out of Sync issue #fabric #fabric-peer #fabric-sdk-node #fabric-endorser #fabric-kubernetes




    Do you mean the existing v1.4 peers are out of sync?
    Or you are re-processing all transactions into a new network and the new network peers are out of sync with each other?
    Or the new peers are out of sync relative to the old peers?
    Which state database?
    Why don't you do an inline upgrade of the peers rather than re-process the tranactions?
    Can you identify the first transaction that is out of sync and look at the logs from that block height.


    Dave Enyeart



    avinash.sachdewani2010---07/28/2021 11:48:00 PM---Hi Fabric Experts, Currently I am upgrading our network from HLF 1.4 to 2.2.

    From: avinash.sachdewani2010@...
    To: fabric@...
    Date: 07/28/2021 11:48 PM
    Subject: [EXTERNAL] [Hyperledger Fabric] Hyperledger 2.2 Peer Out of Sync issue #fabric #fabric-peer #fabric-sdk-node #fabric-endorser #fabric-kubernetes
    Sent by: fabric@...




    Hi Fabric Experts, Currently I am upgrading our network from HLF 1.4 to 2.2. As our current HLF network using docker and kubernetes we decided to use below approach for the migration. Identify the last block using peer channel fetch ‍‍‍‍‍‍‍ZjQcmQRYFpfptBannerStart 
    This Message Is From an External Sender
    This message came from outside your organization.
    ZjQcmQRYFpfptBannerEnd

    Hi Fabric Experts,

    Currently I am upgrading our network from HLF 1.4 to 2.2.
    As our current HLF network using docker and kubernetes we decided to use below approach for the migration.

    1. Identify the last block using peer channel fetch
    2. Clone all the transactions blocks using peer channel fetch
    3. Convert these blocks to JSON
    4. Use the JSON objects to execute transactions using a fabric network.
    We are running both HLF 1.4 and 2.2 network during the migration.
    While transferring the transaction blocks network handling up to 65K transactions after that both peers going out of sync and maintaining different balance.
    65K max I can reach so far sometime peer going out of sync as early as 5K transactions.

    HLF 2.2 network Configuration:

  • One Org
  • 2 Peer
  • 3 Orderer
    Fabric Network 2.2.5
    fabric-ca-client 2.2.5

    Please let me know if any one faced the similar issue and best way to solve the issue.