Include specific gRPC TLS implementation in dependencies instead of relying on an appropriate one to be present as a transitive dependency.
Signed-off-by: Mark S. Lewis <Mark.S.Lewis@outlook.com>
- Set correct CERTIFICATE file name in application ChaincodeEvents exercise.
- Add required environment variables to application AssetTransfer exercise, rather than rely on the RunApplication exercise being run in the same terminal first.
Signed-off-by: Mark S. Lewis <Mark.S.Lewis@outlook.com>
A typo in the 10-appdev-e2e.sh script caused the shell to exit with an error status, and the killall command appears to match some key processes in the runner so removing.
Also:
- Update sample application and smart contract to Node 18, using the latest fabric-contact-api release.
- Use release versions of Fabric v2.5 and update tool versions in workflow set up.
- Use GitHub Action to install just instead of curl of install script.
Signed-off-by: Mark S. Lewis <mark_lewis@uk.ibm.com>
* Fix npm problems on asset-transfer-basic/chaincode-typescript
This patch fixes npm problems on
asset-transfer-basic/chaincode-typescript.
- Sync npm-shrinkwrap.json with package.json to fix the npm ci error
- Update the version of TypeScript to avoid the type error on logform
Signed-off-by: Tatsuya Sato <tatsuya.sato.so@hitachi.com>
* Update npm-shrinkwrap.json
Fix dependencies
Signed-off-by: Arnaud J Le Hors <lehors@us.ibm.com>
---------
Signed-off-by: Tatsuya Sato <tatsuya.sato.so@hitachi.com>
Signed-off-by: Arnaud J Le Hors <lehors@us.ibm.com>
Co-authored-by: Arnaud J Le Hors <lehors@us.ibm.com>
For the asset transfer example (https://hyperledger-fabric.readthedocs.io/en/release-2.5/private_data_tutorial.html#pd-use-case)
the are three private data collections, one per org and one shared between the orgs.
The shared collection didn't have an endorsement policy so inheritted the chaincodes; this was specifically set away from the default
to be OR(Org1MSP,Org2MSP). The documentation says this is to ensure that either organization can create and asset.
However this isn't really necassary, the endorsement policy should be lowest level; so this PR moves the endorsement policy to
the collection. The documentation does I believe lead to a false understanding
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
Ensure that all sample applications return exit code 1 upon failure
so that github actions can report the failure.
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
Flag --csr.hosts is only relevant for tls enrollments, it adds the SAN to the TLS cert.
For msp enrollments, the cert is only used for identity, therefore there is no need for SAN and no need to pass --csr.hosts.
Having --csr.hosts for msp enrollments is misleading, this commit removes it.
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
Also:
- Tidy up README documentation.
- Add MacOS/Homebrew (Apple silicon) default SoftHSM library location to HSM scripts and sample applications to avoid the need to explicitly override the library location when running the sample.
Signed-off-by: Mark S. Lewis <mark_lewis@uk.ibm.com>
* Update go dependencies
Update go dependencies in sample chaincodes and applications
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
* Update to Go 1.19.6
Update to Go 1.19.6
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
---------
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
* Return exit(1) if application-java fails
If application-java fails, return an exit(1) code,
so that callers such as Github Actions CI can detect the failure.
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
* asset-transfer-basic CI improvements
- Application failure should result in CI failure
- Automatically remove wallet from prior runs
- Fix chaincode name issues, allow chaincode name to be passed from CI
- Fix appUser collisions (duplicate registration failures)
- Fix key create collisions across apps (in cases where same chaincode is used for multiple apps)
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
---------
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
This patch fixes the gradle version used in asset-transfer-basic/
chaincode-java build to 7.
The reason is because the gradle version in the gradle:jdk11-alpine
image is updated from 7 to 8, which seems to cause the build to fail.
Signed-off-by: Tatsuya Sato <tatsuya.sato.so@hitachi.com>
The samples repository has been refactored to focus on an asset transfer series of samples.
The goal of the samples repository is to align with the fabric documentation tutorials
with minimal other distractions.
The heritage samples were left in place for a period of time until the asset transfer series stabilized.
It is now time to purge the heritage samples.
Signed-off-by: David Enyeart <enyeart@us.ibm.com>
Rely on the dependencies of @hyperledger/fabric-gateway making these packages transitively available. Having explicit dependency versions in the application can cause type conflicts due to private member changes in the gRPC Client class across versions.
Also fix some missing or incorrect Node types and engines version contraints.
Signed-off-by: Mark S. Lewis <mark_lewis@uk.ibm.com>
The Commercial Paper example code hasn't kept pace with the state-of-the-art of writing applications and chaincode
Whilst the scenario is still a valid usecase, the example itself is not helping - in fact causing confusion.
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
* Add in examples and notes on using PurgePrivateData
- Update the configtx.yaml to enable the 2.5 capabillity
- Added purge into the chancode
- Added pruge into the application client code.
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
* Temporarily remove java cc test
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
* run test suites with 2.5 binaries and images
Signed-off-by: Josh Kneubuhl <jkneubuh@us.ibm.com>
* Swap to use the Helper Function to create policies (#907)
The old way used the direct protobuf functions, there is a helper function
in the chaincode does exactly that.
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
Signed-off-by: Josh Kneubuhl <jkneubuh@us.ibm.com>
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
Co-authored-by: Matthew B White <whitemat@uk.ibm.com>
The old way used the direct protobuf functions, there is a helper function
in the chaincode does exactly that.
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
Late 2022 it was discovered that when used via the CA, the certificate had a fixed named covention. cert.pem
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
Signed-off-by: Matthew B White <whitemat@uk.ibm.com>
