I got to the project that is based on BYFN sample, but limited to one organization only.
In the app when I call following code:
private Contract getContract(Gateway gw) {
return gw.getNetwork("mychannel").getContract("realchain");
}
following error is produced:
2019-10-30 09:34:51.433 INFO 23108 --- [nio-8080-exec-3] org.hyperledger.fabric.gateway.Gateway : Unable to load channel configuration from connection profile:
org.hyperledger.fabric.sdk.exception.NetworkConfigurationException: Channel configuration has no channels defined.
at org.hyperledger.fabric.sdk.NetworkConfig.loadChannel(NetworkConfig.java:519) ~[fabric-sdk-java-1.4.5-20190620.151745-1.jar:na]
at org.hyperledger.fabric.sdk.HFClient.loadChannelFromConfig(HFClient.java:161) ~[fabric-sdk-java-1.4.5-20190620.151745-1.jar:na]
at org.hyperledger.fabric.gateway.impl.GatewayImpl.getNetwork(GatewayImpl.java:258) ~[fabric-gateway-java-1.4.0-20191002.055106-31.jar:na]
...
...the app recovers somehow itself, but I suspect this error to slow down the whole interaction with the ledger.
Also I'd love to keep my logs clean from any exceptions if possible.
Anyone encountered and resolved the same error?
That is normal behaviour. If the connection profile does not contain a channel definition then the client uses peer definitions (currently for your client identity's organisation only) and assumes the channel exists for those peers. It should not impact performance at all.
You have the option of adding a channel definition to the connection profile, which will avoid this log message occurring, but we wanted to avoid this being necessary.
I agree that the exception appearing in the log gives the false impression that a real error has occurred and I have removed that in the current development code.
Related
I've got Caliper configured as a subtree of my custom Fabric blockchain. I have it configured correctly in order to connect the network and am running a small set of tests at the moment. The arguments I'm providing are absolutely correct in terms of the custom chaincode being run, but whenever I try to create a transaction via Caliper I get the error:
Failed to perform submit transaction [CreateAsset] using arguments [<ARGS>], with error: TypeError: Cannot read property 'status' of undefined
Anybody encountered this before with Caliper V0.4.2? It's interfacing with a Fabric 2.3.3 instance on local machine via Kubernetes, with the latest SDK bound.
The error occurs on line 237 of transactions.js inside the node_module fabric-network, where there is a strict property comparison on a commit object.
Also it's worth noting that the querying part of my test rounds is yielding expected results... none of those assets get created due to the error thrown, and when querying for a given asset an error like so occurs:
Failed to perform query transaction [ReadAsset] using arguments [<ID>], with error: Error: error in simulation: transaction returned with failure: Error: The asset <ID> does not exist
The closest thing I've been able to find is this ~2 year old issue which isn't exactly what I'm facing but it's the same sort of error at least... any help or pointers would be greatly appreciated because I'm struggling to glean useful information. (Link: https://github.com/hyperledger/caliper/issues/727)
The issue was due to the host names being converted to localhost when you want to work anything except a network running locally via docker. You need to launch caliper with the following extra option
--caliper-fabric-gateway-localhost false
I am using Hyperledger Fabric 1.4.1, with Kafka ordering service, Couch DB as stateDB, Java chaincode and Java SDK.
I experience timeout expired when I try to send large data to the channel and then query it back.
The data size of the transaction I sent is 512KB, and I can query it from peers.
I tried sending it 7 times with the same id, and I can get all 7 histories with shim function getHistoryForKey. But when I sent it 8 times with the same id, getHistoryForKey failed.
The error message is shown below (from peers):
Error: endorsement failure during query. response: status:500 message:"failed to execute transaction fdfbc5b72efd688136c777f5b114a3cedc62339e42f6c39a27e86ca11b8e3d14: error sending: timeout expired while executing transaction"
I have tried to expand the timeout with following options and their combinations in peer and client:
CORE_CHAINCODE_EXECUTETIMEOUT="120s"
CORE_LEDGER_STATE_COUCHDBCONFIG_REQUESTTIMEOUT="120s"
# (seems this setting is not for Hyperledger Fabric)
CORE_PEER_GRPCOPTIONS_REQUESET_TIMEOUT="120s"
But they didn't help in my case.
I also noticed that the 7 copies of testing data are about 3.5MB, and I tried to query 4MB data (which is about 8 copies) from a peer but failed.
When we check the chaincode runtime docker, we see the following error message:
SEVERE: An error occured on the chaincode stream. Shutting down the chaincode stream.
io.grpc.StatusRuntimeException: RESOURCE_EXHAUSTED:
io.grpc.netty.NettyClientTransport$3: Frame size 6294852 exceeds maximum: 4194304.
It seems like there is a grpc limitation on running the java chaincode docker.
We tested with go chaincode, we do not experience the same issue.
We managed to fix that by overriding the ChaincodeBase class
ManagedChannelBuilder<?> newChannelBuilder() throws IOException {
final NettyChannelBuilder builder = NettyChannelBuilder.forAddress(host, port);
logger.info("Configuring channel connection to peer.");
builder.maxInboundMessageSize(104857600);
logger.info("maxInboundMessageSize is 104857600.");
if (tlsEnabled) {
builder.negotiationType(NegotiationType.TLS);
builder.sslContext(createSSLContext());
} else {
builder.usePlaintext(true);
}
return builder;
}
I have submitted a ticket on fabric Jira to support changing the variable without overriding.
I use Hyperledger Fabric 1.2 to build a blockchain cluster, which contains 3 peers and 3 orderers. I can successfully deploy and invoke the chaincode via both CLI and Java SDK. Everything works fine. However, when I notice the service discovery function and try to use it, I met two problems. First, after I build the discover tool and try to use it to get some discovered information, I can't access the peer and get the message as follow
"failed connecting to discovery service: failed to create new
connection: context deadline exceeded"
The config command is
discover --configFile conf.yaml --userKey ./crypto-config/peerOrganizations/org1.forchain.com/peers/peer0.org1.forchain.com/msp/keystore/7458b29b1fb6a89768585430dbf0e522a40ff4aefe600fc1e4fafe62c3c972e4_sk --userCert ./crypto-config/peerOrganizations/org1.forchain.com/peers/peer0.org1.forchain.com/msp/signcerts/peer0.org1.forchain.com-cert.pem --MSP Org1MSP saveConfig
The query command is
discover --configFile conf.yaml peers --channel lajiao --server localhost:6051
I guess it may be caused by the TLS config so I canceled the TLS and tried again. This time I successfully access the peer and get some messages, but I met another problem. When I use 'discover peers xxx ' command, I always get null result, in fact there are two peers in that channel. When I use 'discover endorsers xxx' command, I always get the following error message
'failed constructing descriptor for chaincodes:'
In the meantime, the peer log outputs the following message:
'Principal set computation failed: chaincode isn't installed on
sufficient organizations required by the endorsement policy 2018-08-01
10:21:50.860 UTC [discovery] chaincodeQuery -> ERRO 1441 Failed
constructing descriptor for chaincode chaincodes:
,: chaincode isn't installed on sufficient organizations required by
the endorsement policy'
I can assure that the chaincode is successfully installed in all peers. And I didn't use the endorsement policy when I instantiated the chaincode. I think it is not the policy problem because I still can invoke the chaincode and propose a transaction.
I also tried to use the Java SDK and found that I can get the orderer nodes info but I can't get the other peer nodes or chaincode info. The log always output: "Discover of chaincode names was empty.". But the chaincode is definitely instantiated and can be invoke via SDK. I refered to the test code in "org.hyperledger.fabric.sdkintegration.ServiceDiscoveryIT" and some key Java code is as follow:
channel.addPeer(peer, createPeerOptions().setPeerRoles(EnumSet.of(Peer.PeerRole.SERVICE_DISCOVERY,Peer.PeerRole.LEDGER_QUERY, Peer.PeerRole.EVENT_SOURCE,Peer.PeerRole.CHAINCODE_QUERY)));
channel.initialize();
System.out.println("================ orderer ===============");
for (Orderer orderer : channel.getOrderers()) {
System.out.println(orderer.getName());
}
System.out.println("================ peer ===============");
for (Peer p: channel.getPeers()) {
System.out.println(p.getName());
}
System.out.println("================ chaincode ===============");
for (String s: channel.getDiscoveredChaincodeNames()) {
System.out.println(s);
}
So, how can I use the 'discover' command under TLS configuration and how can I get the discovered information?
For the config command - you need to pass a TLS root CA, via --peerTLSCA. Please look at the examples in the documentation and act accordingly.
Now - for the second problem, I think that the peers might not know each other in the channel.
Make sure you have anchor peers defined in the channel and that both peers have external endpoints configured.
Feel free to bug me (yacovm) on chat.hyperledger.org if you're struggling for too long and can't solve the problem.
You must add an anchor peer from each organization in the channel, this solved the problem for me. Anchor peers are required for the service discovery since the service discovery uses gossip protocol- thanks #yacovm
I stumbled a similar error (regarding to service discovery) as below.
Go Fabric Client logs:
Failed to get endorsing peers: error getting channel response for channel [myc]:
Discovery status Code: (11) UNKNOWN. Description: error received from Discovery Server:
failed constructing descriptor for chaincodes:<name:"mycc">
Peer logs:
Failed constructing descriptor for chaincode chaincodes:<name:"mycc" > ,:
cannot satisfy any principal combination
It's fixed when I provide CORE_PEER_GOSSIP_EXTERNALENDPOINT environment attribute with a correct value on peer's configuration (in docker yaml file in my case).
As I understood since this attribute is missing, discovery services running on peers failed to communicate with each other to have a conclusion of what current network looks like.
Currently chaincode events in hyperledger will only raise duplicate events n number of times, where n is the number of chaincode events in a block and the event raised is the first event in the block.
const profileRegId = this.event_hub.registerChaincodeEvent(request.chaincodeId, "Profile Added", event => {
this.event_hub.unregisterChaincodeEvent(profileRegId);
em.emit(event.payload);
});
Above is how we are calling the registerChaincodeEvent function in our node application.
createEvent(APIstub, "Profile Added", profile)
Above is how we implement in the chaincode.
Is there a way to raise this as a bug with hyperledger myself?
The procedure of opening a new bugs or submitting issues into Hyperledger Fabric is fairly simple, you need to register your linux foundation id (read here the details) and login into https://jira.hyperledger.org/, once done you can open an issue.
While from your description it's not really clear/obvious there is an issue, if you have n valid transactions each has created an event not sure whenever it was expected to have only single notification. Also please note, since release of v1.1.0 of Fabric there is a new event delivery service: FAB-7069 and here some docs about it.
However, if you still think there is a bug or possible improvement, please submit JIRA.
JIRA issue: https://jira.hyperledger.org/browse/FAB-7420
I'm trying to set up a network of four organisations, each with one peer (set as admin), and an orderer with kafka ordering service along with a client, which makes 7 VMs.
Setting up the cryptographic material, starting every channel (two of them), joining them with the appropriate peers and installing/instantiating chaincode happens with no trouble.
When trying to invoke chaincode, the chaincode can be successfully invoked (status:200) however the results do not seem to be recorded on the ledger. When running chaincode which returns the value for a/multiple key(s), the resulting payload is simply null. This seems to be due to a fault in the ordering service: I get keeping the error message "Error reading from stream: rpc error: code = Canceled desc = context canceled".
Here are logs for the orderer and endorsing peers of the transaction (which returns status:200)
Orderer: https://gist.github.com/alexanderlhicks/91f32fb5ed2b6d5d232dcdad0572ffee
Peer1 (initiating transaction): https://gist.github.com/alexanderlhicks/a2c7a37d3e692336dbc9a8a9f7fd66ba
Peer2 (has the chaincode installed): https://gist.github.com/alexanderlhicks/75c76b11a26963d7a6962f51295ea816
Peer3 (on channel but does not have that chaincode installed): https://gist.github.com/alexanderlhicks/951757e880d8c2f658d2d188e25b9274
What could be causing the "Error reading from stream" message which seems to be the only error happening? I'm tried playing around with peer and orderer configurations but to no avail. Searching for this error also returns issues in channel creation, or joining channels, which I assume has gone fine since I can invoke chaincode on the channels with my peers.
The txid in the logs is 61f54319421bae9c47fab505e502b932ff6c6b58d7d3795afe1400466e663a5d.
Cheers.