1. Fabric
  2. Topics

peer pods and blockchain api service taking too much memory #fabric-chaincode #fabricsdk #java #kubernetes #production

amittripathi9887@...
 

Hi all,

I am trying to invoking 60000 records concurrently in hyperledger fabric network.... deployed in kubernetes environment.

after 200 transaction peers pod memory consumption going very high approx (10-12 gb) and peer pods is going in restart mode again and again.

I am using java sdk to put the transaction. I am taking the 60000 data from opencsv first and storing in arraylist. then looping through one by one into the blockchain. for 200 records it works well but when i try to send 60000 records one by one then fabric peers gets in a restart mode.
David Enyeart
 

Do you have 60000 individual Fabric transactions, or does each Fabric transaction represent a batch of N records?

 

What error is reported in peer logs?

 

Note that you can configure peer for concurrency limits to throttle the number of inbound requests to avoid excessive resource use on any single peer:

https://github.com/hyperledger/fabric/blob/main/sampleconfig/core.yaml#L479-L491

 

For large numbers of inbound transaction endorsements, the recommendation would be to spray the requests across multiple peers to avoid overwhelming any single peer.

 

 

From: fabric@... <fabric@...> on behalf of amittripathi9887@... <amittripathi9887@...>
Date: Friday, December 30, 2022 at 1:57 AM
To: fabric@... <fabric@...>
Subject: [EXTERNAL] [Hyperledger Fabric] peer pods and blockchain api service taking too much memory #fabric-chaincode #fabricsdk #java #kubernetes #production #fabric-sdk-java

Hi all, I am trying to invoking 60000 records concurrently in hyperledger fabric network. . . . deployed in kubernetes environment. after 200 transaction peers pod memory consumption going very high approx (10-12 gb) and peer pods is going in restart

ZjQcmQRYFpfptBannerStart

This Message Is From an External Sender

This message came from outside your organization.

ZjQcmQRYFpfptBannerEnd

Hi all,

I am trying to invoking 60000 records concurrently in hyperledger fabric network.... deployed in kubernetes environment.

after 200 transaction peers pod memory consumption going very high approx (10-12 gb) and peer pods is going in restart mode again and again.

I am using java sdk to put the transaction. I am taking the 60000 data from opencsv first and storing in arraylist. then looping through one by one into the blockchain. for 200 records it works well but when i try to send 60000 records one by one then fabric peers gets in a restart mode.

1 - 2 of 2