Date   

Peer/Orderer memory usage

Eryargi, Hakan
 

Hi,

 

During our performance tests, we noticed that, both peers’ and orderers’ memory usage is increasing under load and they do not release the claimed memory when the load is removed.

 

Is this expected? Anything we can do about it on the configuration side?

 

Peers: Up to 15G -> This is especially very high

Orderers: Up to 4G

 

Fabric version: 1.4.4

 

Thanks,

Hakan

 

 

 

 




This message is for the designated recipient only and may contain privileged, proprietary, or otherwise confidential information. If you have received it in error, please notify the sender immediately and delete the original. Any other use of the e-mail by you is prohibited. Where allowed by local law, electronic communications with Accenture and its affiliates, including e-mail and instant messaging (including content), may be scanned by our systems for the purposes of information security and assessment of internal compliance with Accenture policy. Your privacy is important to us. Accenture uses your personal data only in compliance with data protection laws. For further information on how Accenture processes your personal data, please see our privacy statement at https://www.accenture.com/us-en/privacy-policy.
______________________________________________________________________________________

www.accenture.com


Hyperledger Nexus Repository Sunset

Brett T Logan <brett.t.logan@...>
 

As we close out the migration of Hyperledger Fabric and its supporting projects off of Gerrit and Jenkins to GitHub and Azure Pipeline's, our final task was to retire the use of Nexus for serving non-release artifacts. We are replacing the self-hosted Nexus repo with an enterprise Artifactory instance hosted by JFrog.
 
With the migration to Artifactory complete the Linux Foundation will sunset nexus.hyperledger.org and nexus3.hyperledger.org effective FEBRUARY 1st
 
While the community should be largely unaffected by this change as the artifacts are mostly used in CI, there are a few pieces of chaincode that consume Maven libraries from Nexus that we modified to now pull from Artifactory in Fabric-Samples. If you are running Java chaincode from the Fabric-Samples repository that you cloned prior to January 20th, you should update your POM.xml or build.gradle files to point to https://hyperledger.jfrog.io/hyperledger/fabric-maven instead of Nexus, or reclone the Fabric-Samples repository.
 
As of today, we have also retired the Hyperledger Jenkins server and the Hyperledger Gerrit server.
 
Thank you,
 
Brett Logan
Software Engineer, IBM Blockchain
Phone: 1-984-242-6890
 


Pluggable DBMS

Trevor Lee Oakley <trevor@...>
 

I saw that DBMS is pluggable in the docs. I saw also couchdb is used for private db, but can we use any db, eg oracle or sql server?
 
 
Trevor
 
 


Re: Blockchain and Machine learning #fabric-chaincode #hyperledger-fabric #couchdb

David Enyeart
 

From the Fabric docs:

"If you want to build a dashboard or collect aggregate data as part of your application, you can query an off-chain database that replicates the data from your blockchain network. This will allow you to query and analyze the blockchain data in a data store optimized for your needs, without degrading the performance of your network or disrupting transactions. To achieve this, applications may use block or chaincode events to write transaction data to an off-chain database or analytics engine. For each block received, the block listener application would iterate through the block transactions and build a data store using the key/value writes from each valid transaction’s rwset. The Peer channel-based event services provide replayable events to ensure the integrity of downstream data stores."

Related tutorial and sample:

https://hyperledger.github.io/fabric-sdk-node/release-1.4/tutorial-channel-events.html

https://github.com/hyperledger/fabric-samples/tree/master/off_chain_data



Dave Enyeart

"Shabana Basharat via Lists.Hyperledger.Org" ---01/22/2020 11:34:48 PM---Hi everyone, "Blockchain and Machine learning Integration" is the hot topic today. can we integrate

From: "Shabana Basharat via Lists.Hyperledger.Org" <shabana.basharat=yahoo.com@...>
To: fabric@...
Cc: fabric@...
Date: 01/22/2020 11:34 PM
Subject: [EXTERNAL] [Hyperledger Fabric] Blockchain and Machine learning #fabric-chaincode #hyperledger-fabric #couchdb
Sent by: fabric@...





Hi everyone,
"Blockchain and Machine learning Integration" is the hot topic today. can we integrate machine learning libraries in fabric ??? can we deploy machine model inside chain code? As we all know fabric supports three languages and machine learning is mostly deployed in python so how we can integrate ??? what about data sets of assets on which we want to predict something ,how we can achieve?
I need guidance and discussion on this topic please elaborate .

Thanks




Blockchain and Machine learning #fabric-chaincode #hyperledger-fabric #couchdb

Shabana Basharat
 

Hi everyone,
 "Blockchain and Machine learning Integration" is the hot topic today. can we integrate machine learning libraries in fabric ??? can we deploy machine model inside chain code?  As we all know fabric supports three languages and machine learning is mostly deployed in python so how we can integrate ??? what about data sets of assets on which we want to predict something ,how we can achieve?
I need guidance and discussion on this topic please elaborate .

Thanks


Re: #fabric-questions Fabric Networking #fabric-questions

Nye Liu <nye@...>
 

You didn't mention if you are having NAT traversal issues or straight up firewall issues.

If the latter, opening ports should be sufficient.

If the former, put it in a DMZ with real public ip addresses and ditch NAT entirely.

NAT is a cancer, and never interacts well with p2p protocols.


Re: #fabric-questions Fabric Networking #fabric-questions

Yacov
 

I can only say that port forwarding anchor peers alone won't get you much, because peers try to connect to one another forming a full mesh of an undirected graph (peers p and qhave a single connection between them, either from p to qor from q to p).
Anchor peers are just for bootstrapping membership across organizations.





From:        "Cavell" <cavellt@...>
To:        fabric@...
Date:        01/23/2020 03:05 AM
Subject:        [EXTERNAL] [Hyperledger Fabric] #fabric-questions Fabric Networking
Sent by:        fabric@...




Hi,

I've been trying to setup a fabric network spread across multiple locations. One of the major issues has been locating and communicating with peer nodes behind firewalls. The only solutions I've come up with have been setting up a VPN and joining all the nodes to that or port forwarding some of the nodes (anchor peers and orderers) so that some communication can occur. The first one is usable now, but becomes a nightmare to manage if the network expands. The second is a security risk. Am I missing something obvious? Is there a better alternative to what I'm doing?

Thanks for any help given,
Cavell Teng




#fabric-questions Fabric Networking #fabric-questions

Cavell
 

Hi,

I've been trying to setup a fabric network spread across multiple locations. One of the major issues has been locating and communicating with peer nodes behind firewalls. The only solutions I've come up with have been setting up a VPN and joining all the nodes to that or port forwarding some of the nodes (anchor peers and orderers) so that some communication can occur. The first one is usable now, but becomes a nightmare to manage if the network expands. The second is a security risk. Am I missing something obvious? Is there a better alternative to what I'm doing? 

Thanks for any help given,
Cavell Teng


Re: JIRA Cleanup

Matthew Sykes
 

Items that were labeled on Jan 13 have now been closed out. Apologies for the JIRA spam.

If an issue was closed that should not have been, please feel free to reopen it.

Thanks.

On Mon, Jan 13, 2020 at 10:56 AM Matthew Sykes via Lists.Hyperledger.Org <matthew.sykes=gmail.com@...> wrote:
I have finished labeling JIRA items associated with the Fabric project. If you find an item with a `stale-item` label that you believe should not be closed out next week, please remove the label and add a comment indicating why it is still relevant.

Thanks.

On Thu, Jan 9, 2020 at 7:07 PM Matthew Sykes via Lists.Hyperledger.Org <matthew.sykes=gmail.com@...> wrote:
As part of our v2 shutdown (and my own new year resolution to be better about doing my chores), we'll be doing some work to close out old JIRA work items.

Over the next few days, open items that have not had a meaningful update in more than 9 months will be tagged with a 'stale-item' label. One week after that process has completed, any open items with that label will be closed out.

If a JIRA item you are interested in gets tagged and you want to stop it from being closed, please comment on the issue with information about why it is still relevant and remove the tag. This will indicate further discussion is warranted and we will defer closing the item pending that discussion.

Since the issues are simply being closed and not deleted, if something falls through the crack, we can always reopen as necessary.

Thanks.

--
Matthew Sykes



--
Matthew Sykes
matthew.sykes@...



--
Matthew Sykes
matthew.sykes@...


Re: Downtime for updates of WIKI & JIRA

Tim Johnson <tijohnson@...>
 

All updates have been completed and both servers are back on-line.

On 1/22/20 8:06 AM, Tim Johnson wrote:
The update of JIRA (jira.hyperledger.org) has been completed.

The update of WIKI (wiki.hyperledger.org) is still underway. We expect
to be complete within the hour.

Tim


hyperledger fabric-ca-client through javascript registering more than one user using the same admin identity but unable to access channel from User2 , User3 and so forth #fabric-ca

maruti praturi
 

Everything works fine when I have just "user1" who is registered using the following lines of code

const secret = await ca.register({ affiliation: 'org1.department1', enrollmentID: 'user1', role: 'client', extra: 'client1'}, adminIdentity);
const enrollment = await ca.enroll({ enrollmentID: 'user1', enrollmentSecret: secret });
const userIdentity = fabric_network_1.X509WalletMixin.createIdentity('Org1MSP', enrollment.certificate, enrollment.key.toBytes());
await wallet.import('user1', userIdentity);

When I add one more user "user2", it gets registered. But when I try to discover the channel, while running this below line of code

const network = await gateway.getNetwork('mychannel');

I get the below error

[Channel.js]: Channel:mychannel received discovery error:access denied

I have declared 2 users in count in crypto-config.yaml, still, I am getting this error.


Downtime for updates of WIKI & JIRA

Tim Johnson <tijohnson@...>
 

The update of JIRA (jira.hyperledger.org) has been completed.

The update of WIKI (wiki.hyperledger.org) is still underway. We expect
to be complete within the hour.

Tim


Re: core.yaml

Joe Alewine <joe.alewine@...>
 

In networks with multiple organizations (which should be every network), every organization should own its own peers. Often each organization will own several peers.
 
Regards,
 
Joe Alewine
IBM Blockchain, Raleigh
 
rocket chat: joe-alewine
slack: joe.alewine
 
 
 

----- Original message -----
From: Marina Wanis <marinamaged1996@...>
To: Joe Alewine <joe.alewine@...>
Cc:
Subject: [EXTERNAL] Re: [Hyperledger Fabric] core.yaml
Date: Wed, Jan 22, 2020 9:19 AM
 
Hi Joe, 
 
So, in case of multiple organizations how to setup the peers configurations ? 
 

From: fabric@... <fabric@...> on behalf of Joe Alewine <joe.alewine@...>
Sent: Wednesday, January 22, 2020 3:36 PM
To: hyperledger-fabric@... <hyperledger-fabric@...>
Subject: Re: [Hyperledger Fabric] core.yaml
 
Marina,
 
core.yaml is the file that governs the configuration of a single peer, which should be owned by a single organization, so wanting a sample core.yaml with multiple organizations doesn't make sense to me.
 
Regards,
 
Joe Alewine
IBM Blockchain, Raleigh
 
rocket chat: joe-alewine
slack: joe.alewine
 
 
 
----- Original message -----
From: "Marina Wanis" <marinamaged1996@...>
Sent by: fabric@...
To: "hyperledger-fabric@..." <hyperledger-fabric@...>
Cc:
Subject: [EXTERNAL] [Hyperledger Fabric] core.yaml
Date: Wed, Jan 22, 2020 1:07 AM
 

Hi,

 

I was trying to setup the peer configuration in core.yaml, does anyone have a code for core.yaml with multiple organizations? I wanted to take a look at a sample of the code in core.yaml with multiple organizations and each organization have at least one peer.

 

Thank you,

Marina

 

Sent from Mail for Windows 10

 

 
 
 


Re: Channel Policy

Jason Yellick <jyellick@...>
 

Your orderer processes must always satisfy the /Channel/Writers policy (or they will not be able to process config updates and channel creation transactions).  Based on the below, it looks like your policy restriction is excluding them.  Most likely, what you really want to modify is the /Channel/Application/Readers policy (in the Application section of your configtx.yaml), this will allow you to specify the subset of application orgs you wish to be able to submit transactions.

~Jason
 

----- Original message -----
From: "Nicholas Leonardi via Lists.Hyperledger.Org" <nlzanutim=yahoo.com@...>
Sent by: fabric@...
To: Fabric <fabric@...>
Cc: fabric@...
Subject: [EXTERNAL] [Hyperledger Fabric] Channel Policy
Date: Wed, Jan 22, 2020 8:08 AM
 
Hey guys,
 
Trying to change the channel writers policy but nothing seems to work.
 
Channel: &ChannelDefaults
Policies:
# Who may invoke the 'Deliver' API
Readers:
Type: ImplicitMeta
Rule: "ANY Readers"
# Who may invoke the 'Broadcast' API
Writers:
Type: Signature
Rule: "OR('Org1MSP.member')"
# By default, who may modify elements at this config level
Admins:
Type: ImplicitMeta
Rule: "ANY Admins"
 
 
I need the writers policy to be one organization Org1.
 
I've tried:
 
"OR('Org1MSP.member')"  -> got unexpected status: FORBIDDEN -- signature set did not satisfy policy: permission denied 
"OR('Org1MSP.admin')" -> got unexpected status: FORBIDDEN -- signature set did not satisfy policy: permission denied
 
 
Thanks in advance
 


Channel Policy

Nicholas Leonardi
 

Hey guys,

Trying to change the channel writers policy but nothing seems to work.

Channel: &ChannelDefaults
Policies:
# Who may invoke the 'Deliver' API
Readers:
Type: ImplicitMeta
Rule: "ANY Readers"
# Who may invoke the 'Broadcast' API
Writers:
Type: Signature
Rule: "OR('Org1MSP.member')"
# By default, who may modify elements at this config level
Admins:
Type: ImplicitMeta
Rule: "ANY Admins"


I need the writers policy to be one organization Org1.

I've tried:

"OR('Org1MSP.member')"  -> got unexpected status: FORBIDDEN -- signature set did not satisfy policy: permission denied 
"OR('Org1MSP.admin')" -> got unexpected status: FORBIDDEN -- signature set did not satisfy policy: permission denied


Thanks in advance


Re: core.yaml

Joe Alewine <joe.alewine@...>
 

Marina,
 
core.yaml is the file that governs the configuration of a single peer, which should be owned by a single organization, so wanting a sample core.yaml with multiple organizations doesn't make sense to me.
 
Regards,
 
Joe Alewine
IBM Blockchain, Raleigh
 
rocket chat: joe-alewine
slack: joe.alewine
 
 
 

----- Original message -----
From: "Marina Wanis" <marinamaged1996@...>
Sent by: fabric@...
To: "hyperledger-fabric@..." <hyperledger-fabric@...>
Cc:
Subject: [EXTERNAL] [Hyperledger Fabric] core.yaml
Date: Wed, Jan 22, 2020 1:07 AM
 

Hi,

 

I was trying to setup the peer configuration in core.yaml, does anyone have a code for core.yaml with multiple organizations? I wanted to take a look at a sample of the code in core.yaml with multiple organizations and each organization have at least one peer.

 

Thank you,

Marina

 

Sent from Mail for Windows 10

 

 


Got Error While Committing Block Connection Reset by Peer #fabric #couchdb #network

Faisal
 

This is the 2nd time the peer has crashed due to this error. Nework is on HLF 1.4.2 and the logs are attached in the file. Have anyone faced the same issue before or any ideas on resolving it? I have seen that this is common issue in GO as there are many questions posted on stackoverflow and issues opened on github related to it.

Thank you.


Re: core.yaml

Howin Ho
 

Hi Marina,

I have a working example in github. You are welcome to take a look and feedback comments.
https://github.com/hohowin/my-fabric

Cheers,

Howin.


core.yaml

Marina Wanis <marinamaged1996@...>
 

Hi,

 

I was trying to setup the peer configuration in core.yaml, does anyone have a code for core.yaml with multiple organizations? I wanted to take a look at a sample of the code in core.yaml with multiple organizations and each organization have at least one peer.

 

Thank you,

Marina

 

Sent from Mail for Windows 10

 


Re: [Hyperledger TSC] [Hyperledger Technical WG China] [i18n] Status report on translation of Fabric docs

Yang Cheng
 

Hi, Sara!

I agree with you, Transifex has good features like manage updates, but it's really inefficient, we have tried transifex for months and amlost finished Fabric 2.0 alpha translation, by the end we changed back to Github.

Let's work togethor and find a better way for translating. Keep connecting and if you have any idea please share with us.

And there is a i18n(https://chat.hyperledger.org/channel/i18n) channel in hyperledger chat, welcome to join the channel and discuss with us.


--
程阳
Yang Cheng
great_cy_ang@...

At 2020-01-20 17:36:14, "Sara Garifullina" <garifullina@...> wrote:

Hello everyone!

In Iroha, we were also trying to figure out a new way of translating our docs – we used POEditor before but it is so bad when it comes to automation. 
Ry helped us with getting access to Transifex. I believe its best feature when comparing to manual translation is that the files can be easily updated automatically. Although I agree on your points for sure, it is a tricky tool. 
Anyway, we might be writing some sort of a script anyway to compare the current docs with translations at some point – maybe we could combine our efforts somehow? We do not have much resources right now but still. Let's connect and share ideas! 

Sara Garifullina,
Community manager at Soramitsu
Contact me: garifullina@...



On Fri, Jan 17, 2020 at 7:07 AM Brian Behlendorf <bbehlendorf@...> wrote:
This has gone without a reply since it was posted so I thought I would add one,

It's terrific to see this energy for expanding the global footprint for Fabric! And for taking such a well researched and thoughtful approach to figuring out how to support the needs of translators efficiently. And your recommendations on bold at the bottom make sense for me. Thank you for writing up the recommendations and the rationale, that is valuable for future teams looking at this. An additional repo makes a ton of sense. I am sure here are good techniques to correlate updates to core docs to a need for updating their translated equivalents.

So far, the TSC seems like it has been happy leaving these questions up to individual projects rather than setting a site-wide standard. But the TSC and others in the community might still want to weigh in on this, and if it looks good, consider adopting it as a common standard across projects, so that it's even easier for volunteers for translations on any project to know how and where to plug in.

One last question: would it make sense for translation bundles for in-app localizations to be done in this -i18n repo, or to be done in the main code repo? I'm guessing the former so that a distribution can easily bundle them all together, and they change much less frequently, but I believe they are as important as translated docs (for projects that use them) to highlight to volunteers.

Again, thanks!

Brian

On January 13, 2020 2:57:48 PM GMT+08:00, Yang Cheng <great_cy_ang@...> wrote:
Dear Hyperledger community,

We are a small group of volunteers that have been translating Fabric docs to Chinese since 2018. We’d like to share our current status and rationale behinds some decisions for your reference.

Tool selection

We initially started off using github, since it’s familiar to most of developers, and other projects like k8s have been doing the same. The workflow roughly looks like this:

Admins:
1.create branches in `hyperledger-labs/fabric-docs-cn` following Fabric release tags, for example `1.4.2_zh-CN`
2.populated Github issues with untranslated docs
3.assign issues to translators upon request
4.review pull request
5.readthedocs is updated automatically upon successful merge
6.periodically pull in updates from Fabric docs in the form of new issues

Translators:
1.browse Github issues looking for unassigned issues
2.assign issue by commenting on it
3.translate and submit pull request

This workflow had served us well for a small group of contributors. Later on, translation tools, in particular Zanata and Transifex, were proposed by community members, and we decided to give them a try. However, several major drawbacks of Transifex were observed after months of trial:

1.slow access in this region, resulting in bad user experience
2.intermediate files (.po) loses annotations during conversion, resulting in bad formats
3.no commit sign-off when eventually pushed to github

Therefore, we went back to Github. However, this does not mean we rule out the option of using professional tool, which obviously has its own advantages. Our current focus is to get things done and keep handful of contributors happy. When the time comes that Github becomes bottleneck (either due to increase of volunteers, or number of languages being translated to), we are definitely open for reassessment of tooling.

Location of translated docs

It was proposed to separate docs from Fabric code repo, which can co-exist with translations, similar to k8s [1]. Although the proposal was turned down for solid reasons, and we are happily informed that readthedocs actually supports multiple Github repo setup [2]. This is so far the least invasive option to incorporate non-English docs into main site.

We do not think putting translated docs into Fabric core repo is a good idea, even with fine-grained maintainer-ship in place. The PR page would be overwhelmed by foreign characters and we are no longer able to track tasks with Github issues. Besides, it doesn’t really buy us anything beyond one less repo.

To avoid creating new repo for each language that people are interested in translation, we could also setup a repo `Fabric-i18n` containing them as separate directories, e.g. `zh`, `es`, `de`, etc.

This is how things get done today and we definitely welcome any suggestion and feedback. As the number of volunteers and languages grow, we believe a standardized process will emerge.

Thank you,
Cheng Yang

[2] here’s a demonstration website to show how to incorporate multiple github repo into one readthedocs site https://stone-fabric.readthedocs.io/zh/release-1.4_zh-cn/

--
程阳
Yang Cheng

--
Sent from my Android device with K-9 Mail. Please excuse my brevity.

3841 - 3860 of 11409