Hyperledger Foundation Meetups, 8 Aug 2022

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Ask Me Anything - Hyperledger Fabric

Description

This is a recording of a virtual meetup hosted by the Hyperledger India Chapter on July 22, 2022 featuring Pavan Adhav taking questions about Hyperledger Fabric from the meetup attendees.

Learn more about Hyperledger Fabric at: https://wiki.hyperledger.org/display/fabric/

Learn more about the Hyperledger India Chapter at: https://wiki.hyperledger.org/display/HIRC/

----
Hyperledger – Open Source Blockchain Technologies

Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use

Discord: https://discord.gg/rVCn4j7UFd
Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory

Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter

Follow-us on Twitter @Hyperledger

Learn about Hyperledger Membership: https://www.hyperledger.org/about/join

#Hyperledger #Blockchain
License
Creative Commons Attribution license (reuse allowed)

A

All right, I am recording it now. Yes,.

B

The recording started.

A

All right, when you want to just quickly um do a kind of uh introduction, your own self, if you like to.

C

Yeah sure all right yeah. Thank you guys. Thank you so much for joining this session. So yeah my name is pawan and uh I'm currently working as a senior blockchain developer in the paramount. So especially I'm responsible for uh designing the applications and writing the smart contract related to everything blockchain, even for the deployment as well. So yeah, I'm excited to answer your questions. I I will do my best to answer your most of the questions. So, let's see.

A

All right yeah so so, for I mean um I had you know uh been talking to a lot of people um out here in hyperledger uh uh fabric community, where people are you know building. So what I see is whenever people um you know enter this, the most confusing uh state they have is in how the transactions are. You know going through. Maybe if you can um show us some diagrams or you know, maybe do a whiteboarding and show us how actually a transaction happens on uh you know fabric just to start up.

A

It's just a warm-up exercise, guys pressure pressure is on you. You have to. You know make this really. You know ask questions whatever questions you have. No question is a bad question. Just in that spirit, I would like all of you to think about whatever questions you have and let's make this a very uh useful session for everyone so uh pawan over to you um for the all.

C

Right, thank you. I'm sharing my screen. I hope you guys can see me right.

A

Sure yeah yeah, we can see.

C

Okay, all right yeah. So currently, uh let's consider we have two different organizations and they want to have a communication between them in a distributed way. Okay, so, as per the diagram here, we can see two peer organizations. Are there and one order organization? Is there so this two-peer organization? Again, they are holding some kind of peers. Two peers are there and one of them is the industry in other organization. Also, the situation is exactly the same. They have their own certificate authority even for the ordering organization as well.

C

There is a dedicated certificate authority which I haven't mentioned here, but yeah, so we will have a separate certificate authority for each and every organization. So talking about the transaction flow, the first thing is to interact with our network. We require certificates, because this is a permission plus private network right with only authorized entities are allowed to interact with the network so, as per the business use case.

C

These two sorry, two organizations are in a bond with to have a communication actually and what how transaction will takes place, but before before going to the transaction, let's discuss like what exactly the process to how to create a com, create this network first step is creating all the certificate for all the participants in the network. So using those certified authority we will create all the certificate for all the participant in the organizations who are the participant peer is one of the organization cl all the clients we talk about, and admin are the users.

C

Those are the participants of the organizations for all of them. We will create the certificate using certificate authority for other organizations. Also, we will do the same thing. So what is the next step to have a communication between these two different organizations? We require to have channel artifacts. What are the channel artifacts? Those are nothing but the initial.

A

Blocks actually.

C

Genesis blocks generally create and channel the transactions file, which is responsible for creating the channel later on, so those channel artifacts will create once we have this channel. Artifacts next step is to create a medium between this organization to have a communication so for that using the channel, the transaction file will create a channel so using that channel now they have a medium to communicate, but what kind of communication they are going to do so for that purpose they will.

C

They will write some kind of business logic. That is nothing but the smart contact. Sometimes we call a jam code so that chain code will get deployed between these organizations and finally using the smart contract. They can communicate with each other, okay, so for transaction flow. Let's go to the first step, so this client is responsible for doing the transaction and, let's consider this client.

B

Is belonging to the organization one it.

C

Could be any one client like organization, one organization, so order will not have any kind of uh I mean client uh user. They will have admin uh who is responsible for doing some administrative operations, but yeah there will be always client for the organization.

A

C

And organization, too again, it depends on our use case. First thing: this client need to have the parties uh publicly private career certificate, okay, so this client, when they have the public key private, can the certificate, because using username and pas we don't have anything username and password these. These are the transactions which get invoked using the signature first and the certificate as well. So this client will have.

C

We will need to create the transaction proposal signed proposal general. We call it so the signed proposal first, they will create, and this client will send this transaction proposal to the all the industry in the network. Again, it depends on like how many endorsing payers are there during the deployment of the chain code, we define the chain code policy.

C

If we are not defining explicit explicitly implicit is the majority. So in our network we have two organization. First, is this peer organization? Second, one is this one, and uh what is the majority of the two? It's two, so it means we have to have two endorsements for successful execution of the transaction, so this client will send this transaction proposal to the industry.

C

Zero of the organization too, so in this phase, okay, so this is the most important phase that uh generally we have. uh We have one system chain code that is the escc endorsement system chain code at this phase transaction get validated how exactly validated. So this pr will check if the client is really authorized to the transactions.

C

If he is, then he will check if it has a proper in the signature. How generally signature validated during the during sending this transaction proposal to the industry peer client create the signed proposal. So what is the process for creating the signed up proposal? Let me show you okay here, so these are the proposal. Bytes, okay, left side. You can see. Let's consider this is the proposal like having some information like function, name which function you want to invoke. What are the different arguments?

C

What is on which encode you want to interact and the channel id okay? So how? Generally this digest get created? I will show you here this proposal uh at the client side, only they create a hash, the other we call it ig, sometimes or has doesn't matter so. This has gates signed with private key of the client and it happens at the client side. Only so and finally, this signature get created and you can.

B

C

Is the signed proposal? It's a signature plus proposal by original proposal and this gets sent to the.

A

Whether it's a valid or not, once you get.

B

C

Signature plus the proposal by it so in this signature, uh so this industry just decrypt with the public key of the client in this process. In the signed proposal, this client will send us a certificate as well. Okay, in the certificate at the bottom, we will have uh public.

B

Key of the certificate using this.

C

Public key, we can decrypt the signature and get one digest right and using the same way how client created the signature. We have the proposal by here using this proposal. Byte again, anyone can create a hash and we will get in these hash digest. If those are those digests are exactly same. It means this. So if the certificate is valid, then we can consider this is the valid transaction send by authorized transaction only so this is the process generally happens. Actually.

A

Okay, okay: this is quite valuable, yeah yeah, I have a small query over here. uh Normally right there will be a consensus, happens right either it's a proof of work. So what exactly the consensus mechanism here we are following, and you know uh how how you know do we do it in terms of you know the the code or in the how that you know proof of consciences work so that you know the transaction is completely valid and a very good presentation. Yeah. Thank you.

C

Yes, thank you.

A

Yeah we'll go we'll.

C

Come to the consensus after this transaction flow completion. uh Yeah, that's also a very good question, so we'll go through it once we're done with this transaction. Okay, so.

B

C

B

C

So, generally, this validation happens and, after that, this endorsing peer uh enters the transaction. What is the investment process? Endorsement? Is nothing but just again signing that proposal and create the industry is signature only that is nothing but the signature. So this owns this industry create the uh signed the proposal again at this phase. Only while by executing our smart contract, it create, read, state and the right set.

C

When we talk about the decoding the block and checking the content of the block, I will show you how exactly this looks like like read, set right set. How endorsement looks like how many endorsements are there in the transaction, we'll say.

A

C

I have decoded some of the blocks later once we're done with this. I will show you so finally, this endorsement again sent back to the client okay here this client again validate if this transaction has sufficient endorsement to execute it. So, as per the endorsement policy, uh we require two investment, so client will wait until dedicated to endorsement and he get it. He create a signed. Sorry transaction response.

D

And send back to the order.

C

The responsibility.

A

Of the order, it is.

C

Nothing but just chronologically order, these transactions create a block. After specific configuration, we define it depends on the batch time or batch size. Finally, once the block is created, one of the order will create uh sign.

B

That block again at.

C

The bottom of the each block, we will have the signature of order as well, so once this created the block and signed a block that block gets sent to the leader player of the organization, so how we will decide whether the peer 1 or ps0 is the leader pair. It totally depends on the configuration how we define so there are two configuration first, one is the static or second one is the dynamic.

C

So, let's consider at the point of time we have to.

A

This pn1 is the leader.

C

Pair and we are following the dynamic selection of the leader here- that is again, one election process happens in in between the peers and one leader get chosen at the run time. So now this peer got the block right, so this ph responsibility to just disseminate that block broadcast that block to other query in the organization.

C

Finally, I'm sorry. Finally, each peer will just decode the transaction from the block execute sorry validate the transaction. Now what kind of validation happening here so each pair will validate if he has a sufficient endorsement, each inducement is valid or not, and there is one mvcc multi version, concurrency control, so validation happens and all the validation goes successful. Then the transaction get committed into the ledger again.

B

C

Have two part: first, one is the current state database. Second, one is the chain of the block and both of the places the transaction will get added. So this is the overall transaction flow in the hypoallergenic fabric. So anyone have any questions on this part.

A

Yeah, I think I have already got a couple of questions related to that, uh so uh one is um from lakshmi, so she's asking that or he's asking is the client sending the transactions to the order? I guess this is this is solved. But if you like to answer this and further is being said, is I thought that client sent the transaction to peers.

C

Yeah yeah first in the first page, it will send to the pi once you get a sufficient endorsement. Finally, this client again saying to the order as well.

A

C

A

C

Nothing but the user. So let's consider you are part of the network and you want to remove the transactions right.

C

Okay, so uh we have some client library fabric network. Previously we used every client. So these uh these.

B

Libraries are right now doing.

C

All things for you, you generally just create a transaction proposal and send to the fabric network right, so fabric network is doing everything for us. We don't need to worry on this. I'm just trying to explain how internally it happens. Okay,.

A

Okay, okay, so that means you know the organization, one and organization two are not they know, or the sender and receiver or how we can take it up.

C

A

So I didn't get your question.

C

Okay, all right, so this client belonging to the organization one right, so it's not necessary. I mean the client could be part of anyone. Any one of the.

B

C

So they just create the transaction proposal and send to the industry.

B

C

This client will connect will get connected to all of the peers in the uh so there.

B

Is a discovery service.

C

Which is responsible for getting uh for that particular channel and the smart contract? How many endorsements are required, so this client is responsible for getting this okay at the right time. They will get this information and send this transaction to each other.

A

Okay, I got it so in general, the client used to connect to both the endorsements.

C

A

And receiver of that and then you know they try to coordinate and then the set of the batteries will start.

C

A

C

Any other question.

A

um There are a couple of questions from shane, so shane, if you can just unmute and ask your questions because they are very specific to diagrams and he's pointing to specific areas of the diagram. So that would be great if you can unmute and ask yeah. Thank you. um So in the I have a little bit confusion over the peers right because in uh documentation say there are three peers uh endorsing committing and also the um order of peers right. So in the diagram here. What is the common.

C

Thing first, I did mean okay, okay, all.

A

C

Cool yeah, so there are four rules of the pair first one is the committing. Second, one is the end of thing. Third, one is the leader here and fourth, one is the anchor pier okay. So these are nothing but roll. They are not differentiating in the pier, so one peer can play all of the four role at particular time. So previously someone have asked what about the consensus, how exactly it happens.

C

So we will uh okay go through that, but before that roles of the peer, when, when clients send this transaction to the peer that peer act as a industry peer okay, so how generally we define? We install the smart contract on the pier, and that will be uh the industry for that. Okay, so, in the initial phase, this sphere is acting as the indus npn, okay.

C

So next, when the client get the endorsements sent to the order, order, create a block and finally send that block to the leader here, okay, so I told this process is just dynamic at any point of time, this election can happen and they can so if peer one goes down so peer, zero would be the leader for that organization right. So this leader peer will again. Pl0 is again at this point leader period.

C

Next, one is the committing period when we are sending this block to the leader pair leader. We are again sending this block to all final uh all the period.

B

C

Right in the final phase, the same peer is acting as a committing peer okay, so this committing peer is nothing but just do the validation uh vstc generally, I mean it execute the vsa, that is, the system encode, which is responsible for validating the system, validating the transactions coming from that.

A

C

At that point it plays as the committing pair and what is the anchor pier okay, so this client sending this transaction to the industry period. How generally client get the information which peer is up and running? Okay, so in each organization we can define, one of the pair is anchored here and right now the network is the state forward. We have only two peers in each organization, but let's consider we have 10 pair in each organization.

C

And how we will know which pair is up and running and which pair is in the same pair, so for that purpose, this anchor pair is responsible for that.

C

Just collecting the data like which pair is up and up and running in the organization, and so that client can connect to that anchor pair only and get all the information within that organization, and once you get the sufficient endorsement like how many industry investments we require and which one of them are up and running at this particular time and that client will send the transaction to those investment pair. Only so in this way we have to total four roles of the peer industry period in the endorsement phase. Committing peer had to commit last phase.

C

That is the last phase of the transaction leader pair at the sending of the outer block from the order to one of the pair in the organization and anchor pair, which is responsible for collecting the data collecting the information about which fear is up and running in the organization we got it right. So is that answer your question.

A

Yes, thank you very much. um The second question I have is on the order itself. um So.

C

In case of these two organizations, so who should be running the orders? Okay, it depends again yeah, so, um okay, so here we have three orders right. So one of the organization can uh hold this.

B

Order organization as well, but then.

C

Why other powder, I mean organization, two: will trust on this, like order is responsible for doing that? If they want to do something, they can do it right, so it depends like uh you.

A

Can hold some of the order as well.

C

Like one order and two order- and you can decide like two or three orders- you want to uh final uh what we can say like signature okay, so this order can be hosted by each organization as well. So, let's talk about it. We have three different organization in the network and one each organization can hold one order in pair so that in the final transaction they will have always uh minimum two. I mean out of three order.

C

While we are using this for high availability, even though one up goes down it will it will, our network will be up and running. It totally depends on our business use case like who will hold this or will order our organization, so they can distribute these ordering nodes or they can one of the engine can hold this.

C

Thank you guys very clear.

C

A

About the consensus, nope just quick question, um so I'm specifically looking at um different clients transferring assets between each other and the goal is to get the approval time as short as possible. So some second, you know your transfer has been accepted by the system via consensus. So I'm curious what kind of tweaks we can do to the system uh that improve the response time to the client.

C

Okay: okay, yeah; uh okay. So when we're creating the genesis block right uh so generally, we define uh there are two parameters for the uh batch.

A

Time and the batch size, okay,.

C

Batch time is nothing, but the time required to create the block and batch size is nothing but the how many transactions you want to add in a single block, so both of them we configure we generally while creating the genesis block. Okay, what happens when client is sending the transactions right, so order will wait until they get single transactions? If, within two turn within two seconds, he is getting one.

B

Transactions, then also.

C

Order create a block and send finally to the committing period I mean leader, pair and finally, to the committing period. This is the first approach, second one if within one second, the client sent 10 transactions to the order, in that case as well order will create a block immediately, because one of the criteria is matching either block size or block time. So if one of the criteria is matching order, create a block and finally send to the committing period.

C

Finally- and I'm talking about the finally limited, so you can pick this parameter and you can make it like as soon possible when you are, you are getting one transactions or maybe within a 100 milliseconds. If you are getting only any kind of transactions, whether it's a 10 20 depends, uh you can create a block and you can take that parameter. Actually,.

A

Thank you. I have one basic query: uh where does you know the database as a concept is used when they know when the transaction happens as such, or you know how you know blockchain as such, you know, stores it as a ledger.

A

C

Yeah, okay, so finally I mean uh finally, when peer commit the transaction side, what exactly it happens so in the ledger.

A

C

A

What exactly the laser major is.

C

A

B

C

First is the chain of the block that is the block chain, and second, is the current state database that is the latest state of that particular asset uh get stored. So let's talk about one asset. Car is the asset.

C

uh Car one is the key for asset and we are adding some data like color model or whatever the data. You want to add right. So this client has done first transactions car one uh one of the field owner is the owner, one, okay, so first transaction executed and sent back to the same to the ledger. Now in the current state database, we will have the value for the carbon asset with the owner one and in the chain of the block as well.

C

We will have that information, so let's consider another one client came and he changed or maybe same client has changed the second transaction with the owner owner too. It's the same card. Okay. So at that time current state database will have the latest value of data set. That is the car one. We will have owner equal to two okay and in the chain of the block. We will have one more transactions there. It means we will process the history transaction one with the owner, one transaction two with the owner two.

C

It means at any particular point of time. We can check the history of the transition who has done for this car one, how many transactions happen in the blockchain, so first transaction was the owner one with the owner. One and second transaction was the owner too, but in the current state database we will have the latest value, for that is that answer your question.

A

Okay requesting everybody to be on mute and I am unmuting um indrajeet right now, indrajeet. You can ask your question.

E

Yeah, thank you. uh Thank you for one. Actually, I have um design question which you already touched a bit.

E

um So um whenever I bring up a blockchain network like a fabric, high pressure, hypothermic network, I always have a doubt like, should I have other orders in organization itself or to have a separate uh single organization for orders. So I I want to know. But what does it I mean?

E

What is what are the advantages and this disadvantages of having a separate organization for orders or having the others part of the organization, because I asked um these questions to a hypothetical fabric chat um long ago and then they prefer to have a when the thing is like. If you have single order organization, then it defies the uh meaning of a truly distributed network, because then it becomes a single organization which manages us, but uh I think few people who are like uh very.

A

A

Sorry about that guys, you can. You can continue now.

E

Okay, yeah, so is it like um organization can have uh as a network, uh they can have two organizations where uh one can host the peers and then other audition hosts the orders. So how do like from your experience? What's your preference to have um like orders? How do you have orders? Do you have a yeah.

C

Yes, all right, so, let's consider I mean there could be different kind of scenario. So, let's consider.

E

C

uh Putting the agreement on the blockchain network- and we have two organization: first one is the ibm. Second, one is the lenovo right: they they generally have a lot of communication in terms of the hardware like ibm, uh get some laptops from the linux and they want to have all the agreement on the blockchain right. So in that case, let's consider ibm is fully solely responsible for that to maintain english, so in that case, ibm can hold the order organization and other order. So, of course, every transaction will get executed after the business rules.

C

We define in the smart contract, but yeah in that case, uh ibm can hold the ordering load. But there are some scenarios like the.

A

Other organization never address.

C

On the other uh I mean other organization in the network, so so in that case they can hold their own ordering node. uh But again the answering the query answering to the question like.

B

It totally depends on your business.

C

Use case what they want exactly to achieve, and depending on that, we can have this ordering node to the specific uh or appear organization as well, or we can have a common ordering node as well. So some organization will, of course, maintain it.

E

A

All right so um pankaj, you can ask your question. I'm asking you to answer saying: yep.

B

Yeah, so my question is uh regarding to the client one so in this uh diagram we have shown that this one client, which is communicating with organization one and ovation two, but in a real scenario, I believe there are different clients for different organizations to communicate with their uh peer note. Right, yes, is it? Is it possible that a client which is of organization to uh can query or can do transaction on organization one directly? If somehow, let's say it has their path or everything, so is it possible still.

C

See I mean see: no no client is sending directly any data to the organization 2 or organization 1.. They are going through this consensus process. Okay, we'll talk about the consensus process like how exactly it actually, but whatever the transaction flow I mentioned here right the same way. Any of the client will send the data, even though it's a part of organization, 2 or organization 1, because creating the signed proposal is totally depends on the client, so client, creating the signed proposal by using his private key right and the library were using fabric network.

C

So this organization can also have a separate api server. This organization also can have a separate api server, but the finally, when they are talking with the industry, they want to add some data. Then, of course they will go to this process only first, this client will get the sufficient information like how many endorsing peers are there for that particular channel and the particular chain code. Once they get the data, they will send the transaction to the industry.

B

Okay, the reason I asked this question because uh I have this scenario, including by one of poc, where I have one client where I do. The network discovery everything and I just tried to query something from organization one, although my client was not part of organization one, but still it is able to do the query uh like it will able to execute the smart content and get the data from organization one itself so uh like.

B

I need to put some kind of check in my smart contract, but is it like uh a fundamental function of the hyperledger or is it some kind of, or maybe I'm doing something wrong? Okay, yeah! So query.

C

Part never goes to this transition process. They never change the state of the check blockchain they generally while doing the query right. So I'm part of this organization one and I'm doing I'm doing a query. So in that case, one of the peer will give the data directly. They will not go through this transaction process or something fetching. Data directly from the pair is possible straight forward from the.

B

Same organization.

C

B

Okay, but even the private data, like public data, I understand, but I tried to get something from implicit uh collection and.

E

B

Get the data so should not be there some kind of authentication some mechanism so that no other client organizations can connect the other information get the private data.

C

Okay, so see if we are storing the private data on the other pair, then definitely they will be able to access it right. So you have to define your use case. While uh I mean initial phase, only you have to decide like where you want to store. If you are storing only on the organization, one then definitely only the client from the organization. One only will be able to get.

B

C

The other organization to user will won't be able to get the collection. Actually I mean private data collection they won't be able to. They will get an issue access issue.

B

Actually, that is, that is the problem like I just simply create a smart contract with query function and no no checks and everything and the client from organization 2 will still able to get the private data of organization organization one. So maybe I'm doing something wrong. Maybe there's something with the certification.

C

So, depending on that, you will be able to do it, even though uh both of the organizations are part of the one channel. But there is a private data as well, then uh you can. You can constrain to the uh constraining to the smart contract itself, but as for my understanding, they won't be able to get the data private data actually until and unless they are part of that particular organization.

B

Okay, so in case of implicit organization, do we still need to provide some kind of policy or no no.

C

No implicit all the users from the same organization, they will will be able to get that data for the implicit. We don't need to worry about it by default. It is there in the fabric in 2.1.

B

Okay, okay, maybe I'm doing something wrong because I'm getting data from implicit collection and that's why I didn't provide any policy. So I was in a thought that uh every client which has its own certificate for that organization, should be able to communicate with that and not to any other organization. But in this case, client from another organization can still able to query private data from another organization.

B

C

Yeah, maybe we can maybe, after this session we can simply try to get exactly what is.

A

Loving the conversation, so um I am unmuting lakshmi um asking to unmute lakshmi you there, because yeah.

B

Yeah yeah, thank you.

A

Yeah, thank you, nice presentation, yeah, my doubt is uh like related, you were mentioning that uh the leader here can be set in the configuration file. Is it yes yeah? So I didn't see anywhere like I have seen only like uh setting the anchor pair. So is it? Is it like? Similarly, like setting the angle here, we we are setting the leader or the.

C

Network automatically.

A

Selects it our.

C

Leader pair the configuration generally defines in the docker compose of the pair. Let me show you.

C

If you want to make it static or dynamic by default, is the dynamic? So probably that's why it's not there, but in the peer configuration file you can see.

A

Maybe let me show you here.

C

You can see here with the quality literally by default. Configuration is there use leader election is the true default con. This is the default configuration you will get for the pr code or angle file or leader, false or true. It depends on, like which pair you want to make it.

A

If it is a static, okay, okay, thank you and one more doubter like regarding to the orders. So there are, there will be multiple orders. Is there so so normally do we need to go with three or more than three orders is good for a use case.

C

Again, I mean that would be depend on the business use case, so generally people prefer, like odd number of, of course, first one we have to have our number three five or seven and uh depending on your uh high availability, you may uh you can choose like any number like more than three five seven. So.

A

For a normal application is more than three standard like for good performance and or.

D

A

Is three that I know for the wrath consensus, but uh for if we are designing a use case like.

C

See I mean if you feel you're one of the order, one or two orders are getting frequently disconnected or maybe going down. So in that case your network will be, will not be up and running right I mean uh so. You will face an issue while submitting your transactions, if your only one order is there. So in that case uh you can have five, even though two of them goes down, then three will be there and uh with three also we can achieve the consensus.

A

So the ideal count do you suggest, is five or.

C

uh It again, it depends like you feel like uh if your two orders are going down frequently or one order is going frequently frequently of sometimes also I mean it depends like which service order you are using. Okay,.

A

Yeah so here here in the figure like the three order comes under one uh order organization, so can we split uh it like one or order organization can consist of two order and another order organizations of two orders like that? Do we need to decide anything? Yes,.

C

Yes, we can decide yeah. We had a discussion just before sometime right on that one yeah like it totally.

E

C

On this organization, they want to have a separate ordering organization for them or a dedicated one.

A

Okay, thank you.

A

Okay, um there was one more question um from harsh stating that how is golang useful in hyperledger- and I am unmuting you harsh if you have other questions. Thank.

E

A

Hello, everyone I'm a computer winner in hyperledger, and this is my first meetup.

C

So currently I'm learning golang.

A

C

Development- and I was.

A

Also, learning how to build microservices so.

A

How to learn hyperledger.

C

And how to start working in hyperlogic fabric and also golang is in any way useful in this journey.

C

Okay, okay! So, first of all, if you're familiar with the javascript or go language, you can start with it, because hyperledger have reproved the flexibility over the client, as well as the smart contact language. So you can write a smart contract language right now in the javascript go language or java, or there are multiple other languages. Even our client also can be in the language or javascript or java. There are multiple options. It totally depends on you like if you're familiar with already.

C

I would suggest you to start with that language on you not not necessarily to start directly with the go yeah. There are some advantages because uh fabric itself is written in the book whole fabric so and compared to the other languages. uh So this go. I had a lot of additional things like gossip dissemination and all the peer-to-peer communication. They are using grpc protocols right and go is one of the most efficient language, as per my understanding. So that's why they prefer to. I.

A

Mean initially, everything.

C

Was in the go so later because of the community community suggested some of the new clients and all uh so a lot of people are working to have this flexibility uh on the fabric like independent on the language. If you're familiar with the one language, that's fine, you can write your language, your client and smart contact, as per as per the knowledge you have in that particular language. So it's not necessary to write in the goal itself at the first page, but you are familiar with to go really.

C

You can start with the client go and do smart contract also in the.

A

uh What is the of you know? Debugging you use like you know. Is there any? You know debug tool specifically no to once you know, return your blockchain and then you know if you want to identify how that transactions are going. You know how, where the transaction you know getting stuck in some node.

C

Okay, see uh so we have total uh control over this client right. So in this client we have, we can debug directly straight forward. We don't need to worry on that, but other part internally.

C

uh So for david for that dedicated debugging, we don't have any kind of explicitly tool, but definitely we have to check the logs of each and every services. You can barely check like uh logging of the each pair like what kind of logs we are getting. Of course they have handled all of the errors and everything in the client side. We will get it so we have to.

A

Understand like.

C

Why we are getting that? That is the only thing probably itself you will get all the errors.

B

And all so, you can just save.

C

That but internally we don't have any explicitly tools for debugging, but you.

A

C

Meaningful errors.

A

So still community is working to make it more.

C

Even meaningful in case, if some of them are not there.

A

Okay, okay, just one question: sorry, uh you know, if suppose you know, if there is any issue on order side, also that also now we will be able to know to know what exactly happened from the client side itself. That.

C

Of course, if there is any issue, definitely we will get at the client side in.

A

Case you are not getting just check the logs for that.

C

Particular service, if you are feeling this order, is facing some issue. uh So probably we can just go inside the container check the log or maybe radically check the container. Oh.

A

uh Hi pawan: this is, uh we know, uh so uh we are going to implement the type of topic into the next. This is a live project on the logic project uh we are going to for the data around the dynamics level. So initially we started the.

C

A

A

So which one is the shootable, because this is a live project, it's a big project. So any suggestion from your friend.

C

Okay, all right.

A

C

Again, that's my personal suggestion. So probably you see I'm not expert in the day, especially on the kubernetes. I know the basic things, but that's a different expertise right, for I mean maintaining the some services in kubernetes. So probably I'm not the person. I mean right guy to uh answer on that question, but yeah hyperledger behavior is there there.

B

Is I feel I mean I.

C

Also, try to use it, but I feel uh there.

B

Is a lot of tight.

C

Coupling between the services, if one up goes down, otherwise it will not uh work. Something like that. So that's why we also have this server and there's all the services in the kubernetes as well, but we're not using beverly or something but yeah. Maybe you guys can explore that part very well, or maybe it hyperlapse operator is also there. So people and maintainers are trying to improve their services continuously but yeah.

C

If it's uh saving a lot of times, then I would, uh I would recommend to go through it and just explore it so even accomplish here. I will recommend for the hlf operator.

A

Because there are kind of uh open source by ibm, blockchain platform and could be used very efficiently and is a production grade available. But it depends on the choice but operator.

A

Any repositories.

C

Yeah, so meanwhile, uh there was another one question right on the consensus: okay, so how consensus is achieved in the hyperledger fabric, so in the bitcoin on the ethereum we have some different uh options in the bitcoin. We have proof of work again and in the new family, it's a ethereum 2.0. We are going to have the proof of stake as well right, but in the fabric it's different. Actually, this process is little different. How exactly this transaction get added successfully in the blockchain network?

C

uh Consensus is nothing but just the process achieving the final uh state of the particular asset in the blockchain okay. So this is the three step process. First, one is the endorsing phase. Second, one is the ordering phase. Third, one is the committing phase and in the transaction flow we have already gone through. Where is the indexing phase when client send the transaction to the inversion pair? That is the endorsement phase. Okay, so again, client validation happens.

C

There created the reset and then writes it by executing our own smart contract and send it back to the client. So this is the uh inducing phase. What is the second phase? It's a ordering phase where order just chronologically order, the transactions and, depending on the passage and batch configuration, create a block and sign that block. So this process is nothing but the ordering phase, and this is the second phase. Okay and finally, this order sent block to the final peer leader pair through the committing period and committing fear again execute the transaction.

C

You just check number of endorsements are correct and each endorsement is valid or not. So that is the committing phase, so combined wringly. These three phases are nothing but the consensus. So generally, we call this uh the three phases uh we can achieve consensus using these three phases, endorsement phase ordering phase and finally, committing phase.

B

uh I want to understand how similar the chain code is with smart contracts, that we are writing in solidity. Programming.

C

A

C

It's a different actually and if the fabric is totally different because see we are, I mean we can achieve whatever we want in uh in hyperledger fabric like you, can totally have a customized use case and you can add any kind of data or something, but in.

A

C

Smart contract, there are limitations to add the data right I mean you can add data, but yeah. It will be expensive for you. Transition.

B

And I am very new to hyperledger: can you suggest any resources or courses that are available for learning? Please.

C

Yeah, okay, so, first of all I would uh I would.

A

Recommend you to go through the doctor.

C

For I mean you are, you should be familiar with the doctor in a basic course you can, you can go with once you are done with the docker. If you're familiar with the javascript, you can start with the javascript uh api and the smart contract as well. If you're familiar with the go or java, you can do it, I mean accordingly.

C

uh First thing is official documentation. I would recommend awesome document. This is the best resource right now and uh I created some of the simple use cases, not very solid proof, but I try to cover most of the hyper relative fabric, things on the youtube channel, so.

C

Then you can go with that udemy courses. Those are little bit advanced like if you are familiar with the fabric, then how to add a new organization in the existing channel or new order in the existing.

B

C

So that is first course. Second, one is the multi-use development like how we can do this deployment on the different kind of virtual machines, so that will be kind of uh advanced. I think, and.

A

C

You should be familiar with this all the uh basic flow first, if you are familiar with this, then definitely I would recommend to go through this process so that it will be easy for you to just do this.

E

Is there a possibility to um change the logging level on the fly like once? We.

B

E

The network is up and running and then, if we want to change the logging level of players on orders from info to debug or debug to info as we require, is there a possibility to change the configuration.

C

Yeah, my understanding in the docker compose file itself right.

E

Yeah yeah once.

C

We started yourself, you can sorry here yeah, you can see the logger logging it over here right. You can just restart the server, that's it. uh It will get reflected immediately when you're restarting this docker restart and the container name just change this, and that should work. Actually I tried and uh it worked for me.

E

So we thought that the resetting of the net um each nodes- we cannot change them.

E

But you can pursue.

C

You can purchase the data for that and you can just restart this container.

C

uh What is the issue.

E

C

You're resetting it.

E

No, I I just wanted to know whether do we need to restart each and every node to change the logging levels.

C

Now, if you want to arrest, I mean you want to check the logging of the pr0 of the organization. Then you don't need to reset everything I mean now all the other containers you can do it only for the pr itself right. You want to check the log of the players here in case. If you want to do for everyone, definitely you can do it.

E

Yeah, okay, then, um I also have another question: like: is there any best practices for organizing organizing the chain code, because, like writing, every every function in a single chain code or having uh or separating functions between different chain codes? Is there any best way or it's just purely like your knowledge and business use cases.

C

Yeah, it totally depends on like say, uh chain codes are nothing but just a business logic, and so you know other.

A

C

Or maybe any other services? How, generally we maintain the what you can exactly do the same way even for the other traditional technologies. We maintain a code in a property right, the same way you can follow here itself. uh There is another one thing uh I want to mention here uh in the fabric 2.0 and ago we can install multiple smart contract within the same chain code. So what is the difference between smart contact and chain code?

C

It's nothing exactly same, but we have a privilege to have a multiple smart contract within the same chain code. So there would be only one collection for both of the smart contract but yeah within the same chain code. Also, we can manage it and we can make it so proper folder structures and everything it totally uh for the developers right. We can manage accordingly, like whatever you feel suitable, so would that.

E

Like other languages based practices, you can follow the same thing. So would in that case, that would be like uh uh I'm writing in go. So in a single go package we will have uh like uh um two or more than one chain. Smart contracts is that right.

C

Sorry, sorry, uh can you come again.

E

So let's say we are writing a chain code and then we we have a single package and then, within the package we have like more than one smart contract.

C

Yeah, that's fine, that's fine! It depends on you right if you want to make it in that way, you can make it like within. I have dedicated a video on in the youtube like uh maintaining the multiple smart contact within a single changing code as well.

E

Okay and also the endorsing policy is for the chain codes, not for the channel right.

A

C

Endorsement policy is for the uh champion code.

A

B

C

B

An administrative.

C

Policy, even for the channel and for the genesis block as well.

E

A

I I request everybody to just raise hand so that.

D

Others can also get a chance. We will allow everybody to speak, uh we don't have. uh We are running out of time, so please just raise hand will give chances to everybody and unmute others, and thank you.

A

Right right, thank.

D

A

Thank you, I think that's for the interest of and the fairity which we want to keep um all right. um So I have.

A

I have divided chandran, so I'm um okay. My question is that I want to set a local fabric network. Okay. For that uh how I can do that.

A

I know I don't want to connect to kubernetes or or like that, just in a lab lab setup or something for storing the data in that particular organization, I want to set a local fabric network. Is it possible or.

C

Sorry, actually I miss your initial uh voice. So can you repeat your question.

A

Okay, my question is that I want to set a local fabric network hyperledger fabric network. Local supernote is for an organization. I don't want to be connected with the kubernetes or like that, is it possible or how I can do that.

C

A

It's possible yeah, it is possible and in.

C

The official documentation also uh they have given the script for creating the local setup. Only right there is a test network, you can just uh you, can just go through the documentation and you can just.

A

C

A

C

B

A

C

Without creating without having kubernetes as well.

A

Okay, that means I'm not in a single system. I want to have four to five pieces.

C

Okay, five pieces.

A

Yeah, that is also possible, so.

C

uh You have to I mean you want to do this posting on the multiple uh virtual machines yeah. That is also straightforward, so you have to maintain the proper communication between them. uh Probably you can use this on document. We have to create some network and within that network you can. You can allow this container to communicate. Otherwise you have to make some privilege to discommunicate these different containers with each other.

C

So there is a dedicated, uh dedicated a course actually on this part on the udp, so you can just go through it. If you feel it's useful.

A

C

Did this same deployment on the three four different machines like one order machine one for organization, one third for organization, two and fourth for organization.

A

Okay, okay, just uh I can just search on udemy and find the course.

C

Yeah, I will share the link.

A

Okay, that will be helpful in official hyperledger documentation. It is available. Is it available.

C

No, no in the official documentation it is available for the single machine.

A

Single machine.

A

Okay, thank you. I will check out.

A

All right, so I had one question at the start of the session itself asking about: what's the difference between the admin and users by vinod.

C

Okay, admin, users and the normal users right, client,.

A

C

Okay, all right so uh see in the hyperlapse fabric, we have defined some kind of organization unit. What is the purpose of that? uh The purpose of organization unit is just segregate the function that is aggregated responsibility, so there are four no organization unit. First, one is the admin. Second, one is the client. Third, one is the order, and fourth, one is the peer okay. So in their certificate we have this information. This certificate belongs to which participant like it's a client or it's an admin. It's a peer or it's about error.

C

So, depending on this generally, we have some responsibility.

C

Endorsing peer only can endorse the transactions where we have defined by creating the certificate, so it means only peer will be allowed to endorse the transaction, not admin or the client, because they are not responsible for doing that. Only peer is responsible for endorsing the transactions. Another one is order, is responsible for signing the block. No other entity is allowed to sign the block if they don't have organization unit as the order of industries.

C

So in the state I mean in this way they have segregated the responsibility. Only client and admin can uh know the transactions peer, cannot induce it and sorry send the transactions. Our order cannot send a transaction right in the certificate. We have that all the information, so that way, we have defined this different organization unit as a separate kind of user. Only we can consider and admin have some administrative privileges like let's consider in the running network. We want to add a new organization.

C

Then client will not allow to add the new organization right. Only admin of that organization, those use those participants who have no organization unit as admin will be allowed to those administrative operations. So that's why they segregated this different kind of.

A

Participant in the network, uh a problem- I have one basic query- point see this: the order is there right. So you know at the end of the day,.

A

Actually comes into the picture.

C

Okay, see when climb so as per the diagram here we have so there are two organizations right, so both of the organizations are endorsing the transactions right in the first phase industry phase here and our chain code policy says each organization should endorse the transaction. If one of the organization is not in the sync, so in that case the transaction will be marked as a invalid. They won't be able to change the state of the blockchain, okay and then finally, achieving the state of the particular asset is maintaining this.

C

I mean going through these three process. First, one is endorsing. Second, one is the ordering and third one is the committing phase. So this is nothing but just because it's going through this distributed system only right with proper.

A

A

Okay, so that means you know who who actually know our order here, like you know it could be a you know, any of the third party vendors anybody.

C

No, I mean what uh see it depends again. uh We, I think we already have a discussion on this, but yeah see uh this order can hold by one of the organization. If the other.

A

C

uh Trust on this organization or oh, they can hold some of the ordering node between these uh clusters. Okay, okay, okay, sorry, I missed.

A

C

That's fine, that's fine! Yeah! I mean yeah, that's a good question. It's one of the confusing questions for everyone like for the new and intermediate people, or even for sometimes the expert as well.

A

Yeah, okay: I am um asking arjuna to ask a question.

A

That's now you there sorry, I was talking on mute.

A

So like is there a way where we can get the history for the pdc implementation.

C

History for the pdc yeah: we don't have that in uh right. Now, it's not yet implemented. I mean getting history for the private data collection.

A

C

Yeah, it's not it. I mean implemented in the fabric. You cannot get the history for the private data collection.

A

Now- and my second question is like uh we have three coach pb, uh the thing is like: we have manipulated the data in the three quantitative ways. Now uh I like from the organization one, I tried invoking the transaction and I could able to update the data on the manipulated data.

C

Okay, so what is your chain code policy? It's a one, two or three.

A

C

Or, or of any one of them.

A

Yes, yes, okay,.

C

Yeah this is a very good question, see here uh right now, we'll talk about these two organizations we're having two peer right. So, let's consider we have added one transactions car, one with the older one. Okay and now, when we are talking about doing some other transactions uh with, I want to change the owner three, but before that only uh one of the users from this organization. They have access to the cause deal of this peer right, so they have changed the owner with uh owner fio.

C

In this peer and this you uh one of the users.

B

From this organization also changed with the owner, six.

C

Okay and this another one, one user- they want to change the owner he is considering owner is the one and he wants to change with the owner too. Here, okay, so what this client does? He create the signed proposal and send to one of the industrial payer, because he requires only one eight endorsement.

C

Yes, okay and here so you were changing yeah. So in the industry phase, what this peer does he just create the reset and the writes it depending on the whatever the data they have right. So right.

C

A

I won't just unmute yourself, I have muted all so you have to unmute okay,.

C

All right, sorry, all right yeah so here so this pl will just simulate the transactions and create the read sheet and the right set doesn't matter like what data he has. He will just update the data, because this peer is considering what are the data? They have it's a valid one, so in that case he will write the owner six with the owner too, and this client will just simulate. Sorry client will just check if you got one indoor saying endorsement.

A

Endorsement and just.

C

Send to the order order, create a block and final say committed into the laser. So that's why our chain code, we should have multiple endorsement if these two endorsements are not matching in case as per the investment policy, we have only one. So that's why it's getting executed successfully. It's not checking anything, even though you can check the history in the transaction, but it will get the transaction get added successfully.

C

So, while uh deploying our smart contract, we have to have the chain code policy majority either or each organization should indulge the transaction. So in this phase, let's consider we have. We have a chaincode policy that each organization should induce a transaction.

C

So in that case here itself the client will check if each endorsement is correct or not.

C

If if each right side is correct or not here, except the transition get discarded. So that's why we it's important to have multiple uh organization endorsement.

A

C

Here is that answer equation.

A

uh Means one more clarity: it is like a couch db urls. We have right yeah.

C

A

To be logged in and there it simply changed the data. Yeah, that's great yeah, so yeah means the data would be like it would be sinking right across all the creators. Even.

A

C

Okay, syncing of the data see uh in the cause db. We have one uh one thing: it's a height of the laser okay. If the height of the laser is same exactly on the pier, they will not sync up. If some peer misses some transactions, let's consider peer one, how the ledger height legend laser height is nothing but the block number. Let's consider we have created total 10 blocks till now, and this peer message: some transactions, like we have done again, 10 transactions later and two blocks get added.

C

This peer will get synced with this other peer. If the this.

B

Later height is mismatching.

C

If it is mismatching, then only the data gets synced. Otherwise, if the legend height is exactly same, it means they they are considering. They have the latest data on it.

A

Okay, okay, thank you.

C

Any other question.

A

All right, I am I'm asking for newt pankaj singh.

B

Yeah, so my question is uh related with the private integration. Only so let's say I have put some details about my car in my private data collection, which is the car model name and its price and everything I want to sell it to some other person on the same network. The other person has a process where he can. Let's get the data off the chain, he can generate a hash of that and match. If the hash is certified, hash is available on the network or not. Yes, he can identify. That is the correct data.

B

Now in the previous process, when I put my private data collection user, one has put the details with honor user. One.

A

B

Verified its fine now he has sell the car, and now owner has been changed to uh user two. Now, if this user to wants to resell the car to user three, if user three will uh like create the same hash with user one name so that hash, because that has been already on the chain, so will did this also work. uh Will this work as well, because.

A

Now, at this point,.

C

B

The corner has been changed.

C

Yeah, okay, see this your top when you're talking about the private data, so organization, one is selling this data, this car to the organization to user right.

B

C

So, in that case see when we're adding any kind of private data, any private data for the car one with the owner one so that has get added onto the on-chain okay and other data is available in the private data collection itself. Now you are saying you want to sell this car to the user of the organization too.

C

Okay, so this organization, two should say: save same information with the owner to in a private data collection and send the transit transaction on the on chain actually and when organization two is uh selling again car to some other organization user. So in that case, also that same verification will happen. So you have to uh you have to add some privilege there like, against which data you want to verify this card.

C

If it's uh organization, two and what was the owner of the organization doing the private data collection just give him information, ask him, create a hash and just check on the on-chain.

C

So in that way I mean we can manage that.

B

In the same process where the organization one has been initially updated,.

B

One- and there is a hash already generated for that now- I have updated the same record with user two. Now there is.

C

In the same organization, in the same.

B

Organization, the same private database, I just had updated the key with user12. There is another transaction which has been uh hashed and distributed across the system right, so there are two hashes now one hash which says honor is.

A

Used too another how we.

B

Can uh like remove this process, because at this moment the actual data is with user two updated data. But if.

A

B

Organization user will create a hash with user one. It will also get matched with the hash which has been generated previously, but it's not the actual state of that particular uh car. So is there any way we can avoid this process? We can identify okay. uh This is the invalid hash now, because the current data has been updated with user1 to user2.

A

C

So see, first of all, why we are uh storing the data into the private data collection. I didn't get see if you are saving in the private data collection. It means confidential data right and but you don't want to share with the other organization. If, even though you are changing the owner or something how someone can verify it, you can. Of course you can share directly data to him and just ask him to verify the hash. If the owner's ownership is getting changed, then that should be on the direct public chain.

C

I feel and the confidential data, just like a price or something which is not required to know everyone. I mean if a transaction happening within organization one and two the price was p1 and the same car is changing ownership from owner.

C

Two to honestly, the price is something different: that data we can maintain into the private data collection, but in case, if you want to change, if you want to change uh the ownership of the car, one two two then organization to should just keep the data of that particular car in his own collection and ask anyone to just verify it in case. If you want.

B

Okay, maybe maybe I get get the wrong example with honor: let's take it, this is the price one. Initially. I have created uh this record in my private data collection, with price equals to one lakh. Let's say for the car and.

A

B

I updated that price to 80 000.. I have asked two different users uh to verify the data, one with the 80 000 and one with the one lakh, because the hash of this two transition has already been distributed. If they try to create a hash and verify they will they, both reports will get verified right or it will not.

A

C

Okay, what all right yeah see- I mean private. In the private data collection, we are showing the sensitive information just like a buys or something you can add any number of transactions and all the transition will get validated against the data and the hash will be valid, but it totally depends like other organization or who do? Is there right? They want to verify the data against which data they will verify. It totally depends on their.

A

Communication like they.

C

Want to they want to compare with the price one price. Two prime is three it. It depends on their uh I mean requirement. I feel it totally uh depend on the business use case.

B

D

Question is about the implementation.

D

A

Implementation.

D

Of the deployment, so when we want to deploy the hyperledger fabric for two organizations- okay, the organization- one- is a manufacturer. Another one is a seller for an example, so manufacturer initiates at a deployment and uh they do the ca organization one and the order and everything right and the same thing. The same infra has to be set up in the different data center or it has to be done along with the same data center. In all the servers now how these two organizations set up in two different locations, how this peer joins?

D

The the first organization channels.

C

Okay: okay, there are two things: first, one is the having communication or maybe making on this different centers different location, uh okay and related to the fabric. They are already they are distributed and they are having a communication between them.

C

So in case, if you want to make this in a separate virtual machines, first thing or if you want to make it multi-region or something so probably for that, you have to maintain the network accordingly, so why we are having this multiple pair here, because if some of the ph goes down, other peer will be up and running, even in case of the order as well right.

C

So we have to make sure these peers are not in the same virtual machine if the virtual machines itself goes down and both of the peers we made for the high availability, then that would make that would not make any sense right, because if the virtual machine itself goes down so our network.

A

C

Be down so in that case we have to maintain these peers on the different kind of uh region or maybe different kind of virtual machines. It totally depends on us and we can follow the same approaches. How, generally we do for the traditional technology? Maybe other services not only for the fabric but general purpose.

D

Okay, yeah, that is, for the high availability of the appears correct.

A

D

Go for multiple availability zone and then we distribute it. So that is clear to me now the organization too, I'm talking about it. So I have let's say I have an aws account: I'm setting a multiple parent, different availability zones to meet the high valuability. Now, when talking about organization 2, let's say they are in on premises. They are not intervened into the cloud.

A

D

And they do they do installation of their own high pressure fabric and they create orders right. Every organization have their own order or.

A

D

Or we have one order set of orders for the entire fabric.

D

How does it work uh pawan, it's my.

C

D

So now, in this diagram we have order one two: three: okay organization, one and organization, two correct this organization, one let's say that they are in a different data center. Let's say they are in uh you know, atlanta.

C

D

And this organization, two uh they are in ohio, two different organizations: two geographic locations; okay, now this organization, one they're gonna, have their own set of orders in their location.

C

Okay, that's fine.

D

And this organization too, they are also going to build their own orders in their data center. Yes,.

C

D

Now, let's say assume that we got vpn in between them all the.

C

D

Connectivity has been established, okay, now how this organization 2 in a different ohio region, how they're going to join uh the organization one fabric which is based in atlanta, I'm.

C

D

About only the fabric level setups, I'm you're getting it right. What kind of commands.

C

D

To establish to uh okay.

C

D

Go communicate.

C

Without your question, okay, all right first thing is: they need to have this. This is a distributed and decentralized system right. These different kind of services are communicate with communicating with each other. If they are on the same machine, then they will be able to directly communicate with that particular addresses, even though they are on a different machine. We have to allow them to communicate with each other directly right. So.

A

For that purpose,.

C

There could be multiple options uh in the devs how people generally allow different services to communicate.

B

With each other.

C

But swam is one of them some network. You can create between these different kind of.

B

Virtual machines.

C

Or maybe different region, of course region will have that virtual machines and inside that virtual machines you can create.

A

C

Swarm network between these different virtual machines and allow them directly to communicate because this order are communicating with each other, and this order are also communicating with the peer.

A

C

So it's totally depending on the configuration like how you are making or how you are allowing them to communicate so docker swarm is one of the options even on the so otherwise you have to export some those ips and all.

D

But if we go for docker swarm right, it is one organization installing docker swamp and creating multiple organizations inside it right. But now we are talking about real-time example. The organization 2 is a different company altogether.

C

Yeah, that is fine right.

D

But see if, if I do a docker swarm, let's say I am the docker admin or kubernetes admin right when I deploy it, I'm installing my containers for organization one. I have everything I have to install organization two, which I myself is doing. It's not the second organization which is doing it right.

C

Yeah, of course that's fine see uh when we are creating the song network right. We should throw some sort of key actually to join in the same network from the other organization. So what I will do, okay, you have created some network on your machines and you have created all the participant information or certification all and you have created uh these uh all the peers and everything your own order. Now what you're saying?

C

Okay, so we, you created song network- and you are asking me to just join this form network and allow your network's container to communicate with mine. So for that, so we will create a channel artifacts right in the channel. Artifacts will mention that information, okay,.

D

So in that case, if we do this, then again, this everything is becoming coming under one swarm cluster right, but we want to be geographically distributed in a different uh part of the world or different part of the country. Correct see.

B

If we take everything under docker.

D

Swarm then it becomes like one swag docker swarm network and then all our part of the organization. We are not uh achieving the real. uh You know purity, pure and geographically, distributed or set up right. So my only thing is that you know company in australia and company in america. They want to have a fabric chain, okay, but nowhere that they must have one common infrastructure. They must have different uh infrastructure in their own data centers. How they're going to form a channel? I'm not finding.

C

D

Documentation of the uh hyperledger, I got a question yes uh see this.

C

Is not related to the something uh I mean strictly to the fabric. This is just a general communication. How people do their communication between different kind of services right? They have to expose some kind, of course, or they have to expose some kind of domain so that they can communicate directly uh in the channel configuration sorry in the blog genesis block at the bottom. We define all the orders endpoints you can see here and they are communicating with each other. Using this uh order addresses you can see my screen right.

D

C

So here you, then, you have to maintain like so which are exposed so other um so order. One sorry, one of the order cluster have three order, and other organization also have three order. They have to mention these all the orders here and they should be able to communicate. You have to make some uh adjustment so that they can communicate without any issue these on unleashed domain.

C

You got my uh question right, so we have to uh make some kind of privilege they can communicate without an issue directly. That is the only thing we have to.

A

See maintain, I can add to it. There are two different different systems sitting just next to each other on different different continents, but the two system have their different different ip addresses. Those ip addresses can communicate.

D

With either this that's more related to the communication and basics of docker.

A

Swarm and kubernetes can also do the same. One part is running in one different system with different ip address. One pod is running a different system with another ipad.

A

There are two ip addresses which actually used to communicate with each other and in docker form you can make all the ip addresses as uh master and slave either all of them will be mastered or all of them will escape. So it's the equal like rights distributed to all the system.

A

Then we have to drink more into docker, some rather than uh fabric, and that's.

C

A

C

That's correct and one thing like see order.com, so, let's consider we are in a different region. You have your own virtual machines. I have my own observations. So let's talk about without the swarm as well, so how order will know how they will resolve this order, dot, com and 705 vote? They will try to search.

D

In the local right.

C

First in their local, or they will just try to search with this domain as well, so how they will identify. If this domain have a proper service running behind that particular domain, then we have to.

C

Let them know like this is the domain which is running under this vm, or we have to make some kind of uh this is just a purely communication part actually how this different container communicate with each other, yeah and character water mentioned. I think.

A

Addresses used to communicate to uh system wherever it is generally two ip address and fourth number used to complete it. That's absolutely so in interest of time I think we are running out of time. We will be able to take only one or two question marks so just identify when we can just I have all the repeat, uh uh participants now uh so and uh so I'll I'll just pick one of them. uh First, you can go ahead.

A

C

A

No not necessary.

B

A

B

C

B

Then, later on, devops.

D

C

Totally different right kubernetes it comes under the daves, so you don't need to necessarily know you need it immediately, but yeah. Definitely it's good to know. You can start with these basic things. First and later you can uh expert and get expertise on that part. Kubernetes.

A

Also, what are the perspectives of career perspectives in high quality fabric like demand and average compensation.

B

C

Maybe uh you can, I mean market would be the good uh things to comment on that, but, like market will, let you know but yeah there are a lot of jobs actually on the private blockchain. There are multiple frameworks as well, but fabric is one of the most famous framework right now in the industry, so yeah. Definitely there would be lot of opportunity.

A

All right, I think, with that we can close this uh uh session. Thank you very much. Pawan. Looking at the kind of you know, the number of hands which are still risen at this point looks like we have to do. You know uh another session.

A

But but absolutely wonderful to have the questions and um great responses from you, pawan and everybody who just joined in and put the answers across, and I see a lot of people who are also writing on the chat, which means that we have such a great public to you know, help each other out. uh Let's keep building um and if any of you want to get added to the whatsapp group or want to have more details.

A

Please subscribe to this meetup link which, where we, you have actually registered this so um hyperledger india and hyperledger heights,.

C

Okay, one more thing like uh maybe just let me create some coupon codes for you guys like if you guys are interested in that. So I try to create a please don't share with anyone.

A

This is wonderful, see. Thank you.

A

A lot of people are thanking you already.

C

Just give me a moment: I'm just creating your coupons.

A

C

And we have limited number of involvement on the under.

A

C

Don't uh share with anyone please, otherwise you won't be able to enter.

A

Also, if you can take this coupon code to our whatsapp group, that will be also helpful. Okay,.

C

Let us find it, I will share.

A

All right with that, um okay, just um I'm just messaging everybody, my number, so you guys can just drop me a whatsapp message and ask for a request to the whatsapp group and also just give a brief introduction of who you are, what you do and why you want to be a part of that group and that will help us do a bit of a screening and add you there right.

A

Thank you so much.

C

Yeah, thank you guys. Thank you. So much for joining.

A

Thank you have a great weekend guys thanks pawan thanks. Thank you so much for having this thing very beautiful. Thank you. Thank you.

C

B