Uploaded image for project: 'Fabric'
  1. Fabric
  2. FAB-17823

txValidationCode is not the same for all the peers for the same block

    XMLWordPrintable

    Details

    • Type: Bug
    • Status: Closed (View Workflow)
    • Priority: Medium
    • Resolution: Invalid: Environment Issue
    • Affects Version/s: v1.4.0, v1.4.2
    • Fix Version/s: None
    • Component/s: fabric-ledger
    • Labels:
      None
    • Steps to Reproduce:
      Hide
      1. Create a HF 1.4.2 Network with 2 Orgs and atleast one peer per Org
      2. Load the network with transactions (around a 100)
      3. Check the txValidation Code of a random transaction block. Should be Valid.
      4. Change minor peer configuration for one of the peers, such as debugging level or timeout value.
      5. Rebuild the peer container which configuration was changed. (Make sure disk mounting is disabled) (I used the command docker-compose up -d --build)
      6. Rejoin Channel. (Since all channel data was lost when rebuilding the container because disk mounting was not enabled.)
      7. Check the validity of the same block in step 3, the txValidationCode should change from valid to invalid now. Only for the peer that was rebuild.
      8. Stop the peer container that was rebuilt, execute peer node reset and started the peer container up again. The blocks still has an invalid txValidationCode and is different from other peers even though the peer was reset from the genesis block of the channel.
      Show
      1. Create a HF 1.4.2 Network with 2 Orgs and atleast one peer per Org 2. Load the network with transactions (around a 100) 3. Check the txValidation Code of a random transaction block. Should be Valid. 4. Change minor peer configuration for one of the peers, such as debugging level or timeout value. 5. Rebuild the peer container which configuration was changed. (Make sure disk mounting is disabled) (I used the command docker-compose up -d --build) 6. Rejoin Channel. (Since all channel data was lost when rebuilding the container because disk mounting was not enabled.) 7. Check the validity of the same block in step 3, the txValidationCode should change from valid to invalid now. Only for the peer that was rebuild. 8. Stop the peer container that was rebuilt, execute peer node reset and started the peer container up again. The blocks still has an invalid txValidationCode and is different from other peers even though the peer was reset from the genesis block of the channel.

      Description

      Noticed that when a peer leaves an application channel (peer container is rebuild with no disk mounting enabled, so channel data on the peer gets wiped out) and when the peer rejoins the same channel again the prior blocks become invalid only on the peer that was rebuilt. 

      Tried peer node reset on the peer that left the channel but prior blocks are still invalid. 

      Network setup: 

      2 Org - 4 Peers (2 peers per Org)

      HF: 1.4.2

       

       

        Attachments

          Activity

            People

            Assignee:
            manish-sethi Manish Sethi
            Reporter:
            KoushKilla Koushik Gavini
            Votes:
            2 Vote for this issue
            Watchers:
            6 Start watching this issue

              Dates

              Created:
              Updated:
              Resolved:

                Git Integration