Although every attempt has been made to make this information as accurate as possible, please know there may be things that are omitted, not fully developed yet, or updates since this publication that were not included in the information below. Only the most pressing or significant items have been listed. For the entire list of tickets and or specific information about any given item, please visit the list at Hyperleder Indy's Jira. Once logged in, simply navigate to Projects > Indy.
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.12.4 |
| indy-node | 1.12.4 |
| sovrin | 1.1.89 |
| sovtoken | 1.0.9 |
| sovtokefees | 1.0.9 |
- NYM dynamic validation check transaction w/out verkey or role
| Description | Additional Information | Ticket Number |
|---|---|---|
| NYM dynamic validation check transaction w/out verkey or role |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.12.3 |
| indy-node | 1.12.3 |
| sovrin | 1.1.81 |
| sovtoken | 1.0.8 |
| sovtokefees | 1.0.8 |
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| TAA signature's validation fix (milliseconds sending broke the primary) |
Stop indy-node service on demoted nodes to avoid a minor issue with client's requests processing (see Known Issues for details).
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.12.2 |
| indy-node | 1.12.2 |
| sovrin | 1.1.71 |
| sovtoken | 1.0.7 |
| sovtokefees | 1.0.7 |
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| WARNING messages incorrectly logged if tokens are not used | INDY-2221 | |
REV_REG_DEF tag field is not validated |
INDY-2314 | |
| A node may re-send messages in a loop in case of connection issues | INDY-2318 | |
| Up to F Nodes may not be able to finish View Change if there are uncommitted NODE txns | INDY-2319 | |
| A node lagging behind may not be able to finish view change if nodes have been added/demoted | INDY-2308 | |
| A lagging node may use wrong N and F quorum values and never finish view change if there are NODE txns being processed | INDY-2320 | |
| A lagging node may be the only one who started view change in case of F Nodes added/promoted in 1 batch | INDY-2322 | |
| Debug View Change when nodes added/demoted/promoted | INDY-2326 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Demoted Node should not process client's requests | INDY-2334 |
Use forced simultaneous pool upgrade.
Please be careful with demoting/promoting/adding nodes (see Known Issues for details).
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.12.1 |
| indy-node | 1.12.1 |
| sovrin | 1.1.67 |
| sovtoken | 1.0.6 |
| sovtokefees | 1.0.6 |
- Multiple active TAAs implementation
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| Sovrin upgrade fails due to incorrect package dependencies | INDY-2303 | |
| GET_CRED_DEF for a Schema with a lot of attributes may fail with Timeout | INDY-2306 | |
| Only Trustee or Node owner can be the author of NODE demotion txn regardless of endorsement or auth constraint rules set | INDY-2024 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Allow multiple active TAAs | INDY-2302 | |
| Allow multiple active TAAs: Debug | INDY-2316 | |
| Improve TAA acceptance date validation | INDY-2313 | |
| Get TAA should return the hash | INDY-2297 | |
| Auth_Rules documentation should explain how endorsers work | INDY-2304 | |
| Document PBFT view change protocol | INDY-2138 | |
| Backups should start ordering in new view only after master instance ordered till prepared cert from NewView | INDY-2299 | |
| Get rid of transport batches | INDY-2294 | |
| Enable zeroMQ auto-reconnection | INDY-2289 | |
| Improve simulation tests to include NODE txns | INDY-2286 | |
| Improve BLS signature performance | INDY-2280 | |
| Improve simulation tests to include processing of InstanceChanges | INDY-2263 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| A node lagging behind may not be able to finish view change if nodes have been added/demoted | INDY-2308 | |
| Up to F Nodes may not be able to finish View Change if there are uncommitted NODE txns | INDY-2319 | |
| A node may start re-sending messages in a loop in case of connection issues | INDY-2318 | |
| A lagging node may use wrong N and F quorum values and never finish view change if there are NODE txns being processed | INDY-2320 |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.12.0 |
| indy-node | 1.12.0 |
| sovrin | 1.1.63 |
| sovtoken | 1.0.5 |
| sovtokefees | 1.0.5 |
- Improve primary selection algorithm
- Take into account transaction history when recovering state for new nodes
- Fix new nodes adding when there are old AUTH_RULE or plugin transactions
| Description | Additional Information | Ticket Number |
|---|---|---|
| The problem with a config state | INDY-2283 | |
| Node loses consensus and unreachable by clients | INDY-2287 | |
| A new added node failed to reach consensus | INDY-2254 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| All nodes need to select the same primary during view change | INDY-2262 | |
| Take into account txn history when recovering state from the ledger for new nodes | INDY-2292 | |
| Do not restore Primaries from the audit ledger | INDY-2298 | |
| Start View change on receiving a quorum of ViewChange messages | INDY-2236 | |
| PBFT View change: cleanup and debug Part 3 | INDY-2267 | |
| A Node missing a View Change may not be able to finish it if NODE txns have been sent | INDY-2275 | |
| PrePrepare's Digest need to take into account all PrePrepare's fields | INDY-2285 | |
| Investigate reasons of hundreds VCs during 15 txns per sec production load | INDY-2295 | |
| Support historical req handlers for non-config ledgers | INDY-2307 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| A node lagging behind may not be able to finish view change if nodes have been added/demoted | INDY-2308 | |
| GET_CRED_DEF for a Schema with a lot of attributes may fail with Timeout | INDY-2306 | |
| Only Trustee or Node owner can be the author of NODE demotion txn regardless of endorsement or auth constraint rules set | INDY-2024 |
Please be careful with demoting/promoting/adding nodes (see Known Issues for details).
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.11.0 |
| indy-node | 1.11.0 |
| sovrin | 1.1.60 |
| sovtoken | 1.0.4 |
| sovtokefees | 1.0.4 |
- Switch to PBFT View Change protocol
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| One node doesn't catch up after promotion | INDY-2222 | |
| A Replica may process messages from other Replicas | INDY-2248 | |
| Up to F nodes are out of consensus after >3 hours of load | INDY-2268 | |
| A Node may not connect to another Node after restart | INDY-2274 | |
| Two View Changes happen during master or backup primary demotion | INDY-2247 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Debug: Integrate PBFT viewchanger service into current codebase | INDY-2140 | |
| Request missing ViewChange messages when receiving NewView | INDY-2178 | |
| Basic integration tests with a new View Change protocol need to pass | INDY-2223 | |
| Recover from a situation when View Change is finished on >= N-F of other nodes | INDY-2224 | |
| A Primary lagging behind a stable chedkpoints should not send NewView | INDY-2230 | |
| Do not stabilize checkpoint after the view change if a Replica doesn't have this checkpoint | INDY-2231 | |
| Save PrePrepare's BatchID in audit ledger and restore list of preprepares and prepares on node startup | INDY-2235 | |
| PBFT View Change Debug: Part 2 | INDY-2244 | |
| Optimize Propagate logic | INDY-2257 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| All nodes need to select the same primary during view change | INDY-2262 | |
| A Node missing a View Change may not be able to finish it if NODE txns have been sent | INDY-2275 | |
| A new node joining the pool during the view change may not be able to start ordering immediately | INDY-2276 | |
| Summary: If there are NODE txns for adding/removing nodes interleaved with View Changes (not any view changes, but a specific subset), then either up to F or all Nodes may not be able to finish view change. Please see the details and conditions when it may happen in INDY-2262. |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended. PBFT View Change was implemented but not enabled so old View Change is active now.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.10.0 |
| indy-node | 1.10.0 |
| sovrin | 1.1.58 |
| sovtoken | 1.0.3 |
| sovtokefees | 1.0.3 |
- PBFT View Change implementation (not enabled yet) and corresponding code improvements
- BLS multi-signature fixes and improvements
- The latest version of ZMQ library support
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| GET_TXN doesn't work with old libindy | INDY-2233 | |
| Need to improve error message with invalid signature | INDY-2103 | |
| A node may not be able to connect to another node if another node was able to connect | INDY-2183 | |
| ZMQError: Address already in use when restarting client stack | INDY-2212 | |
| State proofs and audit proofs are not working for token ledger after transaction with FEEs is written | ST-623 | |
| Unexpected change to file access permissions | ST-621 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| All ledgers in a batch need to be BLS multi-signed | INDY-2228 | |
| Drop ppSeqNo on Backups after View Change | INDY-2226 | |
| Move 3PC Message Request logic into a separate service | INDY-2220 | |
| Bump pyzmq to the latest version | INDY-2213 | |
| Integration of Services: Cleanup | INDY-2208 | |
| Integrate Checkpointer Service into Replica | INDY-2179 | |
| Use audit ledger in Checkpoints | INDY-2177 | |
| Integrate OrderingService into Replica | INDY-2169 | |
| Integrate PrimarySelector into View Change Service | INDY-2167 | |
| Integrate view change property-based tests into CI | INDY-2150 | |
| Integrate and run PBFT View Changer simulation tests with a real implementation | INDY-2149 | |
| Implement PBFT viewchanger service with most basic functionality | INDY-2147 | |
| Extract and integrate ConsensusDataProvider from Replica | INDY-2139 | |
| Extract Checkpointer service from Replica | INDY-2137 | |
| Extract Orderer service from Replica | INDY-2136 | |
| Simulation tests for View Changer (no integration) | INDY-2135 | |
| Implementation: Make PBFT view change working | INDY-1340 | |
| Implement network, executor, orderer and checkpointer as adaptors for existing codebase | INDY-1339 | |
| Define Interfaces needed for View Change Service | INDY-1338 | |
| Modify WriteReqManager to meet Executor interface needs | INDY-1337 | |
| Stop resetting ppSeqNo (and relying on this) in new view | INDY-1336 | |
| Enable full ordering of batches from last view that have been already ordered, make execution on replicas that executed them no-op | INDY-1335 | |
| Balance checker script | ST-620 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| One node doesn't catch up after promotion | INDY-2222 |
Migration script will be applied for buildernet only and will return error message in python shell for any other pools (if manual migration will be performed). There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.9.2 |
| indy-node | 1.9.2 |
| sovrin | 1.1.56 |
| sovtoken | 1.0.2 |
| sovtokefees | 1.0.2 |
- Stability fixes
- Endorser support fixes and improvements
- Improving GET_TXN to be able to query just one node the same way as for other GET requests
| Description | Additional Information | Ticket Number |
|---|---|---|
| New nodes added after last upgrade (1.9.1) are not in consensus | INDY-2211 | |
| indy-node broken by indy-plenum and python-dateutil | INDY-2176 | |
| Issue with non utf-8 decoding | INDY-2218 | |
| Endorsers must be specified within the transaction | INDY-2199 | |
| One node doesn't catch up | INDY-2215 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| As a user, I need to be able to know what was the last update time of the ledger when querying a txn via GET_TXN request | INDY-1954 | |
| Endorser field can contain a DID with a known role only | INDY-2198 | |
| TokenAuthNr needs to extend LedgerBasedAuthNr, not CoreAuthNr | ST-618 |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.9.1 |
| indy-node | 1.9.1 |
| sovrin | 1.1.52 |
| sovtoken | 1.0.1 |
| sovtokefees | 1.0.1 |
- New DIDs can be created without endorsers
- Transaction authors don't need to be endorsers
- TAA acceptance should use date, not time
- Bug fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| Incorrect request validation | INDY-2164 | |
| Need to make "reask_ledger_status" repeatable | INDY-2112 | |
| When view change takes too long instance change should be sent periodically | INDY-2143 | |
| Hotfix: Request GET_UTXO needs to take into account max limit for a message | ST-600 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| New DIDs can be created without endorsers | INDY-2171 | |
| Transaction authors don't need to be endorsers | INDY-2173 | |
| Grab pool data for failed system tests | INDY-2141 | |
| Memory profiling needs to be removed from GET_VALIDATOR_INFO output | INDY-2182 | |
| Implement PBFT viewchanger service with most basic functionality | INDY-2147 | |
| Extract Orderer service from Replica | INDY-2136 | |
| Extract and integrate ConsensusDataProvider from Replica | INDY-2139 | |
| TAA acceptance should use date, not time | INDY-2157 | |
| Clean-up Pluggable Request Handlers | INDY-2154 | |
| GET_UTXO needs to use pagination to be able to return all unspent UTXOs regardless of max message size | ST-602 | |
| Extended State Proof verification for GET_UTXO | ST-604 |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
Some nodes can fail to send a REJECT or REPLY to client under specific network conditions. See Know Issues for more details.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.9.0 |
| indy-node | 1.9.0 |
| sovrin | 1.1.50 |
| sovtoken | 1.0.0 |
| sovtokefees | 1.0.0 |
- Pluggable Request Handlers have been implemented
| Description | Additional Information | Ticket Number |
|---|---|---|
| Propagates with invalid requests can lead to node crashes | INDY-2144 | |
| There is no validation of the ISSUANCE_TYPE field for the transaction REVOC_REG_DEF | INDY-2142 | |
| Reduce CONS_PROOF timeout to speed up catchup under the load | INDY-2083 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| As a Trustee(s), I need to have a way to set multiple AUTH_RULES by one command | INDY-2087 | |
| Make more system tests to be ready for Indy Node CD pipeline | INDY-2127 | |
| Integrate new handlers into the codebase | INDY-1861 | |
| Define Interfaces needed for View Change Service | INDY-1338 | |
| Rename TRUST_ANCHOR to ENDORSER | INDY-1950 | |
| Update PBFT view change plan of attack | INDY-2134 | |
| Apply a new Docker-in-docker approach for system tests | INDY-2131 | |
| More tests for pluggable request handlers | INDY-2108 | |
| Remove ANYONE_CAN_WRITE | INDY-1956 | |
| [Design] ViewChange protocol must be as defined in PBFT | INDY-1290 | |
| Batch containing some already executed requests should be applied correctly | INDY-1405 | |
| Update Pluggable Req Handlers | INDY-2097 | |
| As a Network Admin, I need to be able to forbid an action in AUTH_RULE, so that no changes in code are needed | INDY-2077 | |
| Create Builders for handlers | INDY-1860 | |
| Apply new request handlers approach to Token Plugins | ST-523 | |
| Implementation: Apply new request handlers approach to Token Req Handlers | ST-510 | |
| Implementation: Apply new request handlers approach to sovtoken FEEs Handlers | ST-578 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Incorrect request validation | INDY-2164 | |
| PoA: Request GET_UTXO needs to take into account max limit for a message | ST-575 |
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.8.1 |
| indy-node | 1.8.1 |
| sovrin | 1.1.46 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| All BuilderNet nodes are restarting every 30-50 seconds | INDY-2128 | |
| Primaries are not updated in audit ledger if one of the primaries is demoted | INDY-2129 | |
| A client may receive NACK for a payment transfer request, but the transaction will be eventually ordered (payment transferred) | INDY-2122 |
Payment transaction can return NACK from the pool but in fact it will be eventually ordered (see more details below).
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
Pool upgrade to sovrin 1.1.32 and above should be performed simultaneously for all nodes due to txn format changes.
Pool upgrade to indy-node 1.8.0 should be performed simultaneously for all nodes due to audit ledger.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.8.0 |
| indy-node | 1.8.0 |
| sovrin | 1.1.45 |
- Add Transaction Author Agreement Acceptance Mechanisms and Transaction Author Agreement support
- Configurable Auth rules improvements
- Stability fixes
| Description | Additional Information | Ticket Number |
|---|---|---|
| Issues with catch up and ordering under the load | INDY-2064 | |
| Editing of CLAIM_DEF uses auth rules for Adding a Claim Def | INDY-2078 | |
| Faulty primary can order and write already ordered and written request | INDY-1709 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| As a Network Admin, I would like to use GET_AUTH_RULE output as an input for AUTH_RULE | INDY-2102 | |
| Get Transaction Author Agreement Acceptance Mechanisms from the Config Ledger | INDY-2071 | |
| Support Transaction Author Agreement in Write Requests | INDY-2072 | |
| Validate transaction author agreement as part of consensus | INDY-2073 | |
| Write Transaction Author Agreement to Config Ledger | INDY-2066 | |
| Get Transaction Author Agreement from the config ledger | INDY-2067 | |
| Write Transaction Author Agreement Acceptance Mechanisms to the Config Ledger | INDY-2068 | |
| Catch-up should take into account state of other nodes when sending requests | INDY-2053 | |
| Added FeesAuthorizer | ST-535 ST-529 | |
| Update SET_FEES and GET_FEES logic for Auth Rules | ST-531 | |
| Update GET_FEES to use aliases | GET_FEE request was added | ST-559 |
| Support TAA for XFER txn | ST-557 | |
| Register auth rule for MINT, XFER_PUBLIC and SET_FEES requests in auth_map | ST-536 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| A client may receive NACK for a payment transfer request, but the transaction will be eventually ordered (payment transferred) | INDY-2122 | |
| Incorrect auth constraint for node demotion | INDY-2024 | |
| Request GET_UTXO may fail if payment address contains too many UTXOs | ST-566 |
(!) Don't use GET_AUTH_RULE txn - it will cause the service to crash.
(!) There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
(!) Pool upgrade to sovrin 1.1.32 should be performed simultaneously for all nodes due to txn format changes.
(!) Pool upgrade to indy-node 1.7.1 should be performed simultaneously for all nodes due to audit ledger. There should be no fees set up.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.7.1 |
| indy-node | 1.7.1 |
| sovrin | 1.1.41 |
- Audit Ledger
- helps keep all other ledgers in sync
- helps recovery of pool state by new or restarted nodes
- can be used for external audit
- Correct support of multi-signatures
- Configurable Auth Rules in config state
- Stability fixes
| Description | Ticket Number |
|---|---|
| Validator-info doesn't show view change information and sometimes shows node info as unknown | INDY-2008 |
| Schema can't be written with error "'Version' object has no attribute 'dev'" | INDY-2020 |
| Node fails to start after the load | INDY-2018 |
| POA: Sovrin TestNet lost consensus | INDY-2022 |
| Nodes can fail on first start after upgrading from version without audit ledger to version with audit ledger | INDY-2047 |
| Pool is getting out of consensus after a forced view change and writes to all the ledgers | INDY-2035 |
| View Change processing - replica ends up with incorrect primaries | INDY-1720 |
| Validator node shows False for consensus | INDY-2031 |
| Watermarks may not be updated correctly after view change by a lagging node | INDY-2060 |
| ATTRIB doesn't have auth rules in auth map | INDY-2061 |
| Some nodes are stalled and throw an error under load | INDY-2050 |
| Some nodes failed to join consensus after upgrade | INDY-2055 |
| Description | Ticket Number |
|---|---|
| Implementation: Restore current 3PC state from audit ledger | INDY-1946 |
| Implementation (not active): As a user/steward I want to have better understanding of release version and changelog | INDY-1992 |
| Implement auth rule maps in config ledger | INDY-2001 |
| Add audit ledger | INDY-1944 |
| INSTANCE_CHANGE messages should be persisted between restarts | INDY-1984 |
| Add updateState method for ConfigReqHandler | INDY-2006 |
| Use auth constraints from config ledger for validation | INDY-2002 |
| Implementation: Improve catch-up to use audit ledger for consistency | INDY-1945 |
| Implement a command to set auth constraints | INDY-2003 |
| Debug and validation: Move the auth_map structure to the config ledger | INDY-1995 |
| Need to enhance write permissions for Revocation transactions | INDY-1554 |
| Implement a command to get auth constraint | INDY-2010 |
| Integrate testinfra-based system tests to Indy CD | INDY-2016 |
| Debug and Validation: As a user/steward I want to have better understanding of release version and changelog | INDY-2019 |
| As a QA I want system tests to be run in parallel in CD pipeline | INDY-2028 |
| Debug and Validation: Audit Ledger and improving catch-up to use audit ledger for consistency | INDY-1993 |
| Need to track same transactions with different multi-signatures | INDY-1757 |
| Debug and Validation: Restore current 3PC state from audit ledger - Phase 1 | INDY-2025 |
| A Node need to be able to order stashed requests after long catch-ups | INDY-1983 |
| Need to account fields from PLUGIN_CLIENT_REQUEST_FIELDS when calculating digest | INDY-1674 |
| Debug and validation: Multi-signature support | INDY-2046 |
| Debug and Validation: Restore current 3PC state from audit ledger - Phase 2 | INDY-2051 |
| Description | Ticket Number |
|---|---|
| Get auth rule txn breaks the indy-node service if plugins are on | ST-566 |
(!) There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
(!) Pool upgrade to sovrin 1.1.32 should be performed simultaneously for all nodes due to txn format changes.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.58 |
| indy-node | 1.6.83 |
| sovrin | 1.1.35 |
| Description | Ticket Number |
|---|---|
| validator-info to client times out if there are many upgrade attempts by node | INDY-1922 |
| Node on Sovrin TestNet did not upgrade automatically | INDY-1919 |
| Node that does not upgrade spams the config ledger | INDY-1918 |
| Incorrect pool upgrade txn validation | INDY-1953 |
| Upgrade appears to have broken "validator-info --nagios" | INDY-1920 |
| Node can't order after view change and catch up | INDY-1955 |
| Unclear error messages when Trustee send a NYM with the same verkey | INDY-1963 |
| A role that has been removed can't be added back | INDY-1971 |
| Description | Ticket Number |
|---|---|
| Limit the number of attributes in schema | INDY-1914 |
| Enable Clear Request Queue strategy | INDY-1836 |
| A Node needs to be able to order requests received during catch-up | INDY-1876 |
| Network maintenance role | INDY-1916 |
| There should always be fresh enough signature of a state | INDY-933 |
| Node stops working without any services failure | INDY-1949 |
| As a user of Valdiator Info script, I need to know whether the pool has write consensus and when the state was updated the last time | INDY-1928 |
| Trust anchor permission not needed for ledger writes | INDY-1528 |
| Description | Ticket Number |
|---|---|
| Some of the nodes lagging behind during the load test | INDY-1965 |
(!) There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
(!) Pool upgrade to sovrin 1.1.32 should be performed simultaneously for all nodes due to txn format changes.
(!) indy-node 1.6.82 isn't using the latest versions of libindy-crypto (must be 0.9.5) and python3-indy-crypto(must be 0.9.5). Set these packages versions in case of new node manual installation.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.57 |
| indy-node | 1.6.82 |
| release version number | 1.1.34 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Added old instance change messages discarding. | INDY-1909 | |
| Increased ToleratePrimaryDisconnection and bind re-try time. | INDY-1836 | |
| Added check for None of replica's primary name during logging. | INDY-1926 | |
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
Pool upgrade to sovrin 1.1.32 should be performed simultaneously for all nodes due to txn format changes.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.55 |
| indy-node | 1.6.80 |
| release version number | 1.1.33 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed an issue with the pool doing a restart if there was no consensus. | INDY-1896 | |
| Fixed an issue with intermittent test failure: test_primary_selection_increase_f | INDY-1872 | |
| Fixed the throughput class creation bug. | INDY-1888 | |
Warning: Embedded command-line tool indy is no longer available. For further pool interaction use the indy-cli package https://github.com/hyperledger/indy-sdk/tree/master/cli
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
Validator-info output has been changed. If you use validator-info as a data source make sure that you have done necessary changes for compatibility.
The INDY-1818 (Init Indy Node should output Base58-encrypted verkey already) affects nodes adding.
Pool upgrade to sovrin 1.1.32 should be performed simultaneously for all nodes due to txn format changes.
There are still some issues with token functionality.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.54 |
| indy-node | 1.6.79 |
| release version number | 1.1.32 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed an issue with intermittent test failure: test_primary_selection_increase_f. | INDY-1872 | |
| Fixed an issue with the pool stopped writing after production load with fees. | INDY-1867 | |
| Node wasn't validating CLAIM_DEF's filed ref. | INDY-1862 | |
| Fixed an issue where the Sovrin package couldn't be upgraded. | INDY-1850 | |
| There was an error with Expected object or value error during batch handling. | INDY-1849 | |
| Fixed an issue with a node that wasn't on the network was showing as 'unreachable'. | INDY-1842 | |
| There wasn't enough information about upgrade in journalctl. | INDY-1834 | |
| Fixed 'aws_manage' playbook that was failing when inventory directory wasn't specified. | INDY-1827 | |
| A Validator on Sovrin MainNet was failing to upgrade, then was failing to revert. | INDY-1824 | |
| The pool stopped writing during load testing against domain and pool ledgers together. | INDY-1823 | |
| The node service was stopping during node key validation. | INDY-1820 | |
| Made it so that Init Indy Node outputs the Base58-encrypted verkey already. | INDY-1818 | |
| Investigated slowness on TestNet due to demotion. | INDY-1816 | |
| Fixed and issue where the new instance was removed after creating it. | INDY-1815 | |
| Fixed issues relating to upgrading from 1.6.645+ version resulting in errors about packages versions in journalctl. | INDY-1781 | |
| Fixed an issue with RequestQueue in Replica wasn't clearing after View Change. | INDY-1765 | |
| Description | Workaround | Ticket |
|---|---|---|
| Created a Diagram for Components. | INDY-1870 | |
| Created a Catch-up Sequence Diagram. | INDY-1869 | |
| Made a Write and Read request flow. | INDY-1868 | |
| Created a Plenum Consensus Protocol Diagram. | INDY-1851 | |
| Made it so that change pool state root hash for BLS-signature is in Commit messages. | INDY-1846 | |
| Made a 3rd party open source manifest. | INDY-1839 | |
| Enabled PreViewChange Strategy | INDY-1835 | |
| Added Names to AWS ec2 instances and security groups. | INDY-1828 | |
| Securely automated SSH authenticity checking. | INDY-1826 | |
| Limited RocksDB memory consumption. | INDY-1822 | |
| Ran a very long load test on a small local pool. | INDY-1821 | |
| Created AWS tags for pool automation AWS resources. | INDY-1813 | |
| Adjusted last_ordered_3pc and performed GC when detecting lag in checkpoints on backup. | INDY-1795 | |
| Improved usability of current pool automation PoC. | INDY-1792 | |
| Created the ability to refer different groups in the same namespace using one inventory. | INDY-1788 | |
| Removed security groups at tear-down phase for both tests and playbooks. | INDY-1784 | |
| Cleared the Requests queue periodically. | INDY-1780 | |
| Tested ZMQ Memory Consumption with restarting of listener on every X connections. | INDY-1776 | |
| Got information about how many client connections is usually in progress. | INDY-1775 | |
| Did a long test with a load pool can handle. | INDY-1774 | |
| Found out why validation of PrePrepares with Fees takes so long. | INDY-1773 | |
| Checked why backup instances stop ordering so often. | INDY-1772 | |
| Made it so that you can perform tests on docker. | INDY-1771 | |
| Tested ZMQ Memory Consumption with restricted number of client connections. | INDY-1770 | |
| Ran load tests with file storage. | INDY-1769 | |
| Changed dependency building for upgrade procedure. | INDY-1762 | |
| Used persisted last_pp_seq_no for recovery of backup primaries. | INDY-1759 | |
| Extended the Load Script with GET_TXN. | INDY-1756 | |
| Avoided redundant static validation during signature verification. | INDY-1753 | |
| Found out why max node prod time increases during long load test. | INDY-1747 | |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Upgrade to latest Sovrin doesn't work if it depends on a newer IndyNode. | INDY-1764 | |
| Verify Upgrade of the MainNet to the latest IndyNode. | INDY-1763 | |
| Failed to upgrade from Sovrin 1.1.24 to Sovrin 1.1.26. | INDY-1748 | |
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.53 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.6.78 |
| release version number | 1.1.30 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed and issue where the ledger statuses and maximal consistency proofs is not canceled. | INDY-1740 | |
| Fixed an issue where Bug in calling notifier methods in . | INDY-1741 | |
| Fixed an issue where 35 view changes were happened during 10 minutes after nodes failure because of an invalid request. | INDY-1696 | |
| Fixed an issue where the requests queue is not cleared in case of reject-nym transactions. | INDY-1700 | |
| Fixed an issue where throughput critically decreases without causing view_change | INDY-1672 | |
| Fixed an issue where Node can't catch up large ledger. | INDY-1595 | |
| Fixed an issue where we were unable to demote node in STN. | INDY-1621 | |
| Fixed an issue where view changes happen when all responses should be rejected during load testing scenario. | INDY-1653 | |
| Fixed an issue where node doesn't write txns after disconnection from the rest nodes. | INDY-1580 | |
| Fixed an issue where throughput is degrading if backup primary is stopped. | INDY-1618 | |
| Description | Workaround | Ticket |
|---|---|---|
| Switch off a replica that stopped because disconnected from a backup primary. | INDY-1681 | |
| Extended load scripts emulating non-smooth load according to the changes in the core script. | INDY-1667 | |
| Proof of stability under load. | INDY-1607 | |
| Investigate Out of memory issues with the current load testing. | INDY-1688 | |
| Do not re-verify signature for Propagates with already verified requests. | INDY-1649 | |
| POA: Require multiple signatures for important transactions. | INDY-1704 | |
| Support all FEEs txns in the load script. | INDY-1665 | |
| Test domain transactions with FEEs. | INDY-1661 | |
| 3PC Batch should preserve the order of requests when applying PrePrepare on non-primary. | INDY-1642 | |
| Ability to switch off (remove) replicas with no changes of F value. | INDY-1680 | |
| A node should be able to participate in BLS multi-signature only if it has a valid proof of possession. | INDY-1589 | |
| Make validator info as a historical data. | INDY-1637 | |
| Known Issue: There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended. | ||
| Known Issue: INDY-1447 - Upgrade failed on pool from 1.3.62 to 1.4.66 | Note that INDY-1447 was fixed in indy-node 1.5.68, but it still presents in indy-node 1.3.62 and 1.4.66 code. So, some of the nodes may not to be upgraded during simultaneous pool-upgrade. If this problem will appear, stewards should perform manual upgrade of indy-node in accordance with this instruction. | |
Pool upgrade from indy-node 1.3.62 to indy-node 1.6.78 should be performed simultaneously for all nodes due to txn format changes.
To reduce the risk of reproducing INDY-1447, it is recommended to use the old CLI for pool upgrade from indy-node 1.3.62.
All indy-cli pools should be recreated with actual genesis files. For more details about txn format changes see [INDY-1421]((https://jira.hyperledger.org/browse/INDY-1421).
Important: Several iterations were done very rapidly between the last release and this one. All of the changes, upgrades, etc... are included in this new release. Simply upgrading will include them all from 1.1.13 until 1.1.17. To see further, specific numerous changes, please reference the appropriate tickets in the Hyperledger Jira ticketing system.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.51 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.6.73 |
| release version number | 1.1.17 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed and issue where the pool stopped writing after F change. | INDY-1583 | |
| Fixed an issue where read_ledger was passing incorrectly formatted stdout and breaks convention | INDY-1645 | |
| Fixed an issue where the node couldn't catch up a large ledger. | INDY-1595 | |
| Fixed an issue where the Validator Info may hang for a couple of minutes. | INDY-1603 | |
| Fixed an issue where the read_ledger tool is not able to read the sovrin plugin ledger. | INDY-1548 | |
| Description | Workaround | Ticket |
|---|---|---|
| Made it so that the 3PC Batch should preserve the order of requests when applying PrePrepare on non-primary. | INDY-1642 | |
| Made it so that Monitor takes into account requests not passing the dynamic validation when triggering view change. | INDY-1643 | |
| Improved throughput calculation to reduce a chance of false positive View Changes. | INDY-1565 | |
| Made it so that the performance of monitor is improved. | INDY-1660 | |
| Made it so that Stewards, can have a script that can generates Proof of possession for their BLS key. That value can now be used in a NODE txn. | INDY-1588 | |
| Added Support Proof of Possession for BLS keys. | INDY-1389 | |
| Made it so that the average is not used when calculating total throughput/latency of backups. | INDY-1582 | |
| Made it so that any client requests are discarded during view change. | INDY-1564 | |
| Created a simple tool to show graphical representation of some common metrics. | INDY-1568 | |
| Changed default configs for better performance and stability. | INDY-1549 | |
| Known Issue: Upgrade failed on pool from 1.3.62 to 1.4.66. | Note that INDY-1447 was fixed in indy-node 1.5.68, but it still presents in indy-node 1.3.62 and 1.4.66 code. So, some of the nodes may not to be upgraded during simultaneous pool-upgrade. If this problem will appear, stewards should perform manual upgrade of indy-node in accordance with this instruction (!) To reduce the risk of reproducing INDY-1447, it is recommended to use old CLI for pool upgrade. | INDY-1447 |
Pool upgrade from indy-node 1.3.62 to indy-node 1.6.73 should be performed simultaneously for all nodes due to txn format changes.
There must be sovrin package upgrade to 1.1.17 version after indy-node package upgrade. You need to specify package=sovrin in pool-upgrade command to do this.
All indy-cli pools should be recreated with actual genesis files.
For more details about txn format changes see INDY-1421 .
There are possible OOM issues during 3+ hours of target load or large catch-ups at 8 GB RAM nodes pool so 32 GB is recommended.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.6.49 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.6.70 |
| release version number | 1.1.13 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed and issue where several nodes (less than f) were getting ahead the rest ones under load. | INDY-1473 | |
| Fixed an issue where the pool has stopped to write txns. | INDY-1539 | |
| Fixed an issue where re-send messages to disconnected remotes. | INDY-1497 | |
| Fixed an issue where the pool stopped writing under 20txns/sec load. | INDY-1478 | |
| Fixed an issue where 1.3.62 -> 1.5.67 forced upgrade without one node in schedule failed. | INDY-1519 | |
| Fixed an issue where tmp.log must have unique name. | INDY-1502 | |
| Fixed an issue where a node needed to hook up to a lower viewChange. | INDY-1199 | |
| Fixed an issue where the one of the nodes lagged behind others after forced view changes. | INDY-1470 | |
| Made it so that View Change should not be triggered by re-sending Primary disconnected if Primary is not disconnected anymore. | INDY-1544 | |
| Description | Workaround | Ticket |
|---|---|---|
| Made it so that as a Trustee running POOL_UPGRADE txn, you can specify any package depending on indy-node, so that the package with the dependencies get upgraded. | INDY-1491 | |
| Made it so that Monitor is reset after the view change. | INDY-1555 | |
| Made it so that GC by Checkpoints are not triggered during View Change. | INDY-1545 | |
| Made it so that the validator info must show committed and uncommitted roots for all states. | INDY-1542 | |
| Explored timing and execution time. | INDY-1475 | |
| Memory leaks profiling. | INDY-1493 | |
| Bound connection socket to NODE_IP | INDY-1531 | |
| Enable TRACK_CONNECTED_CLIENTS_NUM option | INDY-1496 | |
| Updated revocation registry delta value during REG_ENTRY_REVOC writing. | INDY-1378 | |
| Support latest SDK in Indy Plenum and Node. | INDY-1480 | |
| Latency measurements in monitor are windowed. | INDY-1468 | |
| Trust anchor permissions are not needed for ledger writes. | INDY-1528 | |
| Known Issue: Docker pool can't be built because of new python3-indy-crypto in sdk repo. The problem described in INDY-1517 will be fixed in the next release of indy-node. | Workaround for this problem is to add python3-indy-crypto=0.4.1 to the list of packages to be installed. | INDY-1517 |
| Known Issue: Upgrade failed on pool from 1.3.62 to 1.4.66. Note that INDY-1447 was fixed in indy-node 1.5.68, but it still presents in indy-node 1.3.62 and 1.4.66 code. | So, some of the nodes may not to be upgraded during simultaneous pool-upgrade. If this problem will appear, stewards should perform manual upgrade of indy-node in accordance with this instruction:(!) To reduce the risk of reproducing INDY-1447, it is recommended to use old CLI for pool upgrade. | INDY-1447 |
Pool upgrade from indy-node 1.3.62 to indy-node 1.6.70 should be performed simultaneously for all nodes due to txn format changes.
There must be sovrin package upgrade to 1.1.13 version after indy-node package upgrade. You need to specify package=sovrin in pool-upgrade command to do this.
All indy-cli pools should be recreated with actual genesis files.
For more details about txn format changes see INDY-1421 .
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.5.48 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.5.68 |
| release version number | 1.1.12 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed and issue where logs were appearing in the old CLI. | INDY-1471 | |
| Fixed an issue where there were numerous blacklists under high loads. | INDY-1461 | |
| Fixed an issue where the pool stopped writing after 1114k txns (different view_no). | INDY-1460 | |
| Fixed an issue where the "AttributeError: NoneType object has no attribute 'request' during load"; was appearing. | INDY-1464 | |
| Fixed an issue where the validator-info was reading an empty file. | INDY-1406 | |
| Fixed an issue where validator-info -v --json wasn't producing valid JSON. | INDY-1443 | |
Fixed an issue where the first Pre-Prepare message had incorrect state trie root right after view_change (on master replica). |
INDY-1459 | |
| Fixed an issue where the pool could not order transactions because Node set incorrect watermarks after its restart. | INDY-1455 | |
| Fixed an issue where the pool stopped working due to several incomplete view changes. | INDY-1454 | |
| Fixed an issue where the node crashes on _remove_stashed_checkpoints. | INDY-1427 | |
| Fixed an issue where memory was running out during non-completed viewChange process (under load). | INDY-1360 | |
Fixed an issue where part of nodes continued ordering txns after incorrect state trie under load. |
INDY-1422 | |
| Fixed an issue where the upgrade failed on pool from 1.3.62 to 1.4.66. | INDY-1447 | |
| Fixed an issue where a forced upgrade from 1.3.62 -> 1.5.67 without one node in schedule failed. | INDY-1519 | |
| Description | Workaround | Ticket |
|---|---|---|
| Implemented periodic restart of client stack to allow new clients to connect. | INDY-1431 | |
| Got rid of peersWithoutRemotes. | INDY-1467 | |
| High Watermark on backup may be reset to 300. | INDY-1462 | |
| We now allow optional field in node-to-node and client-to-node. | INDY-1494 | |
| Catchup during view change may last forever under the load. | INDY-1463 | |
| Propagate Primary mode should not be set for already started view change. | INDY-1458 | |
| Catchup needs to be finished during high load. | INDY-1450 | |
| Included reviewed logging strings in Indy. | INDY-1416 | |
| Added benchmark performance impact of recorder tool. | INDY-1483 | |
| Decreased the amount of logging with INFO level. | INDY-1311 | |
| Made it so that throughput measurements in monitor should are windowed. | INDY-1435 | |
| Limited the number of requested PROPAGATES in MessageRequests. | INDY-1386 | |
| Made it so that any client requests during view change are not processed. | INDY-1453 | |
| Made it so that a node must send LEDGER_STATUS with correct last ordered 3PC after catch-up. | INDY-1452 | |
| Fixed calculation of prepared certificates during View Change. | INDY-1385 | |
| Made it so that catchup should not be interrupted by external events. | INDY-1404 | |
| Known Issue: Upgrade failed on pool from 1.3.62 to 1.4.66. Note that INDY-1447 was fixed in indy-node 1.5.68, but it still presents in indy-node 1.3.62 and 1.4.66 code. So, some of the nodes may not to be upgraded during simultaneous pool-upgrade. If this problem will appear, stewards should perform manual upgrade of indy-node in accordance with this instruction: (!) To reduce the risk of reproducing INDY-1447, it is recommended to use old CLI for pool upgrade. | INDY-1447 | |
Pool upgrade from indy-node 1.3.62 should be performed simultaneously for all nodes due to txn format changes.
All indy-cli pools should be recreated with actual genesis files. For more details about txn format changes see INDY-1421.
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.4.45 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.4.66 |
| release version number | 1.1.11 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed and issues where one of the nodes stopped writing after 44287 txns with errors in status. | INDY-1410 | |
| Fixed an issue where the pool stopped accepting transactions on 5731 txns (1 sec delays, no logging). | INDY-1365 | |
| Fixed an issue where the pool stopped writing after ~300,000 txns from 5 clients | INDY-1315 | |
| Fixed an issue where STN was not accepting transactions with only one node down. | INDY-1351 | |
| Fixed an issue where the pool stops taking txns at ~178k txns written in ledger. | INDY-1260 | |
Fixed an issue where ReqIdrToTxn does not store information about the ledger. |
INDY-1327 | |
| Made simple Timeout fixes of the current View Change protocol. | INDY-1341 | |
| Fixed an issue where the migration fails in case of upgrade to version with new transactions format. | INDY-1379 | |
Fixed an issue where --network parameter of read_ledger doesn't work. |
INDY-1318 | |
| Fixed an issue where the /var/log/indy/validator-info.log was inappropriately owned by root. | INDY-1310 | |
| Created a fix around the issues found in the current logic of catch-up. | INDY-1298 | |
| Fixed GetValidatorInfo so it has correct validation for signature and permissions. | INDY-1363 | |
| Fixed an issue where there was an unhandled exception during node working. | INDY-1316 | |
Fixed an issue where validator-info and read_ledger were giving inconsistent responses in node on provisional. |
INDY-1219 | |
| Fixed an issue where the pool stops taking txns at 3000 writing connections. | INDY-1259 | |
| Description | Workaround | Ticket |
|---|---|---|
Reviewed and replaced assert with exceptions in indy-plenum where needed. |
INDY-810 | |
| Tuned RocksDB options for the best performance. | INDY-1245 | |
| Created a migration guide from Indy-node 1.3 to 1.4. | INDY-1392 | |
| Сhanged a key in the requests map and field reqIdr in Pre Prepare and Ordered. | INDY-1370 | |
| Investigated issues found during load testing of 25-nodes pool with increased timeouts for catchups and viewchange. | INDY-1400 | |
| We now support binding on separate NICs for Client-to-Node and Node-to-Node communication. | INDY-1332 | |
| Added short checkpoints stabilization without matching digests. | INDY-1329 | |
| Added indy-crypto package to the hold list. | INDY-1323 | |
| Removed ledger status based catch-up trigger together with the wrong catch-up workflow. | INDY-1297 | |
| Read-ledger without storage copy in case of RocksDB (RocksDB read-only mode support). | INDY-1243 | |
| Applied state machine to Catchup code. | INDY-971 | |
| Refactored the common Request structure. | INDY-1124 | |
| Refactored the common transactions structure. | INDY-1123 | |
| We now support the new libindy with changed txn format. | INDY-1319 | |
| Explored config parameters to find the best performance/stability settings. | INDY-1334 | |
| Extended the Validator Info tool to provide more information about the current state of the pool. | INDY-1175 | |
| A Steward needs to be able to get validator-info from all nodes. | INDY-1184 | |
| Modified existing load scripts for a better load testing. | INDY-1279 | |
| Performed a migration from LevelDB to RocksDB | INDY-1244 | |
| A Trustee needs to be able to restart the pool in critical situations. | INDY-1173 | |
| Move the log compression into separate process. | INDY-1275 | |
| Known Issue: There's an incorrect read_ledger info with seq_no parameter. | INDY-1415 | |
| Known Issue: Pool upgrade should be performed simultaneously for all nodes due to txn format changes. All indy-cli pools should be recreated with actual genesis files. | ||
| List of breaking changes for migration from indy-node 1.3 to 1.4: | 1.3-1.4 Migration Guide |
Pool upgrade should be performed simultaneously for all nodes due to txn format changes.
All indy-cli pools should be recreated with actual genesis files.
Old CLI (indy):
upgrade from 1.3 to 1.4 version
delete ~.ind-cli/networks/<network_name>/data folder
replace both old genesis files by new ones (from 1.4 node or from sovrin repo)
New CLI (indy-cli):
upgrade from 1.4 to 1.5 version recreate indy-cli pool using 1.4 pool genesis file (from 1.4 node or from sovrin repo)
List of breaking changes for migration from indy-node 1.3 to 1.4:
https://github.com/hyperledger/indy-node/blob/master/docs/1.3\_to\_1.4\_migration\_guide.md
IndyNode 1.4 and LibIndy 1.5 compatibility:
General
By default LibIndy 1.5 will be compatible with IndyNode 1.3 (current stable), and not 1.4 (the new one).
LibIndy 1.5 can become compatible with IndyNode 1.4 if indy_set_protocol_version(2) is called during app initialization.
Guideline for teams and apps
Applications can freely update to LibIndy 1.5 and still use stable Node 1.3
If an app wants to work with the latest master or Stable Node 1.4, then they need to support breaking changes (there are not so many, mostly a new reply for write txns as txn format is changed, see 1.3_to_1.4_migration_guide.md)
call indy_set_protocol_version(2) during app initialization
Use https://github.com/hyperledger/indy-sdk/blob/b4a2bb82087e2eafe5e55bddb20a3069e5fb7d0b/cli/README.md#old-python-based-cli-migration to export dids from your old CLI wallet to the new one (new indy-cli).
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.2.42 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.3.62 |
| release version number | 1.1.10 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Fixed an issue where the STN was losing consensus. | INDY-1256 | |
| Fixed an issue where we were unable to use the read_ledger tool with the parameter "to". | INDY-1284 | |
| Fixed the upgrade from 1.2.223 (1.3.55 stable analogue) to 1.3.410 (rocksdb) wasn't working. | INDY-1330 | |
| Description | Workaround | Ticket |
|---|---|---|
| Support was added for supervisord. | https://github.com/hyperledger/indy-node/pull/588 | |
| Indy-node dependencies are fixed. | ||
None for this release.
None at this time.
| Components | Version Numbers |
|---|---|
| indy-node | 1.3.57 |
| release version number | 1.1.9 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| The Node was restarting because of an "Out of memory" error. | INDY-1238 | |
| The pool was not working after not simultaneous manual pool upgrades. | INDY-1197 | |
| When adding a new schema, field 'attr_names' of schema json can be an empty list. | INDY-1169 | |
| This prevents an Identity Owner from creating a schema or claimDef. | INDY-1111 | |
| There was the same primary for both instances 0 and 1. | INDY-1112 | |
| The node logs were being duplicated in syslog. | INDY-1102 | |
| It was possible to create several nodes with the same alias. | INDY-1148 | |
| There was ambiguous behavior after node demotion. | INDY-1179 | |
| One of the nodes were not responding to libindy after several running load tests. | INDY-1180 | |
| When returning N-F nodes to the pool, "View change" was not occurring if the Primary node was stopped. | INDY-1151 | |
| There was a failed restart after getting the "unhandled exception (KeyError)". | INDY-1152 | |
| Fixed a bug where you were unable to install indy-node if sdk repo is in sources.list | INDY-1269 | |
| Description | Workaround | Ticket |
|---|---|---|
| Made it so that a developer can distinguish logs of each replica. | INDY-1186 | |
| Made it so a developer, can track the path of each request. | INDY-1187 | |
| Made it so that you can use RocksDB as a key-value storage. | INDY-1205 | |
| Refactored the common Request structure. | INDY-1124 | |
| Made it so that it supports anoncreds revocation in Indy. | INDY-680 | |
| Made it so that it supports REVOC_REG_DEF transaction. | INDY-1134 | |
| Made it so that it supports GET_REVOC_REG_DEF request. | INDY-1135 | |
| Made it so that it supports REVOC_REG_ENTRY transaction. | INDY-1136 | |
| Made it so that it supports GET_REVOC_REG request. | INDY-1137 | |
| Made it so that it supports getting state root by timestamp. | INDY-1138 | |
| Got rid of the RAET code. | INDY-1057 | |
| Incubation: Move CI part of pipelines to Hyperledger infrastructure. | INDY-837 | |
| Made it so that a user can revoke a connection by rotating the new key to nothing. | INDY-582 | |
| Known Issue: Define the policy how to restore node from the state when it's stashing all the reqs and there is a risk of running out of memory. | INDY-1250 | |
| Known Issue: Re-promoted node cannot hook up to a lower viewChange. | INDY-1199 | |
| Known Issue: One of the nodes does not respond to libindy after several running load test. | INDY-1180 | |
| Known Issue: One node fails behind others during the load_test with a high load. | INDY-1188 | |
| Known Issue: Pool can be broken by primary node reboot in case of network issues between nodes. Note: RocksDB was added as dependency (INDY-1205). It is used for revocation, but the rest part of node functionality is still using LevelDB. | INDY-1256 | |
None for this release.
None at this time.
Important: Upgrade to this version should be performed simultaneously for all nodes (with force=True).
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.2.34 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.3.55 |
| release version number | 1.1.8 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Transactions were missing from the config ledger after the upgrade. | INDY-799 | |
| The node was broken after a load_test.py run. | INDY-960 | |
| The pool stopped taking transactions after sending 1,000 simultaneous transactions. | INDY-911 | |
| The pool stopped working: Node services stop with 1,000 simultaneous clients doing GET_NYM reads | INDY-986 | |
| The node is broken after adding it to the pool. | INDY-948 | |
| The generate_indy_pool_transactions command can be run only by an indy user. | INDY-1048 | |
| Made it so that updates to existing Schemas are not allowed. | INDY-1035 | |
| The pool was unable to write txns after two nodes adding. | INDY-1018 | |
| Fixed a bug where it was possible to override CLAIM_DEF for existing schema-did pair. | INDY-1083 | |
| Fixed a bug where here was a huge amount of calls and a lot of execution time in kv_store.py. | INDY-1077 | |
| One of added nodes wasn't catching up. | INDY-1029 | |
| The pool stopped working and lost consensus while new node was performing a catch-up. | INDY-1025 | |
| Performing a View Change on large pools of 19 or more nodes can cause pool to stop functioning. | INDY-1054 | |
| Performing a View Change issue stopped the pool from accepting new transactions. | INDY-1034 | |
| We were unable to send transactions in STN. | INDY-1076 INDY-1079 | |
| Replica.lastPrePrepareSeqNo may not be reset on view change. | INDY-1061 | |
| We were unable to send an upgrade transaction without including demoted nodes. | INDY-897 | |
| The Nym request to STN was resulting in inconsistent responses. | INDY-1069 | |
| The validator node was being re-promoted during view change. | INDY-959 | |
| There was a false cancel message during an upgrade. | INDY-1078 | |
| Transactions were being added to nodes in STN during system reboot.. | INDY-1045 | |
| There were problems with nodes demotion during load test. | INDY-1033 | |
| The node monitoring tool (email plugin) wasn't working. | INDY-995 | |
| ATTRIB transaction with ENC and HASH wasn't working. | INDY-1074 | |
| When returning N-F nodes to the pool, View Change does not occur if Primary node is stopped. | INDY-1151 | |
| We were unable to recover write consensus at n-f after f+1 descent. | INDY-1166 | |
| Newly upgraded STN fails to accept transactions (pool has been broken after upgrade because of one not upgraded node). | INDY-1183 | |
| We were unable to submit upgrade transactions to STN. | INDY-1190 | |
| Description | Workaround | Ticket |
|---|---|---|
| Added indy-sdk test dependency to plenum and use indy-sdk for plenum tests. | INDY-900 INDY-901 | |
| Published docker images to dockerhub. | INDY-962 | |
| Simplified the view change code. | INDY-480 | |
| Refactored config.py to reflect file folder re-factoring for Incubation. | INDY-878 | |
| Added Abstract Observers Support. | INDY-628 | |
| Moved scripts from sovrin-environment to one of Indy repos. | INDY-1055 | |
| Got rid of Sovrin dependency in the environment scripts. | INDY-1064 | |
| Updated information in "Getting Started with Indy". | INDY-1062 | |
| Updated information in "Setting Up a Test Indy Network in VMs". | INDY-1062 | |
| Add iptables rules to limit the number of clients connections. | INDY-1087 | |
| Knowledge transfer on Indy build processes. | INDY-1088 | |
| Incubation: Move CI part of pipelines to Hyperledger infrastructure. | INDY-837 | |
| Made it so that a user can revoke a connection by rotating the new key to nothing. | INDY-582 | |
| Client needs to be able to make sure that we have the latest State Proof. | INDY-928 | |
| Created it so that anyone could have access to an up-to-date Technical overview of plenum and indy. | INDY-1022 | |
| Known Issue: Pool has lost consensus after primary demotion (with 4 nodes setup only). | INDY-1163 | |
| Known Issue: Ambiguous behavior after node demotion. | INDY-1179 | |
| Known Issue: One of the nodes does not respond to libindy after several running load test. | INDY-1180 | |
| Known Issue: Pool does not work after not simultaneous manual pool upgrades. | INDY-1197 | |
| Known Issue: Pool stops working if the primary node was not included to schedule in the upgrade transaction. | INDY-1198 | |
Node promoting is not recommended for 1.3.52 version according to known issues because backup protocol instances may work incorrectly until next view change.
As mentioned above, upgrade to this version should be performed simultaneously for all nodes (with force=True).
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.2.29 |
| indy-anoncreds | 1.0.11 |
| indy-node | 1.2.50 |
| release version number | 1.1.7 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| A node was maintaining a pace with the network exactly 12 transactions behind. | INDY-759 | |
| New nodes added to an existing pool were unable to sync ledgers with the pool. | INDY-895 | |
| Scheduled upgrades were happening at the current time on some of the nodes. | INDY-231 | |
| Some nodes were not restarting after a canceled pool upgrade. | INDY-157 | |
| Sovrin logs were insufficient for failed upgrade. | INDY-801 | |
A node was getting the wrong upgrade_log entries after restarting and was running the wrong upgrade. |
INDY-917 | |
An earlier pool_upgrade was not happening when there was an upgrade to schedule to happen in the future. |
INDY-701 | |
| A validator was running instance change continually on the live pool. | INDY-932 | |
| New nodes added to an existing pool were unable to participate in consensus after the upgrade. | INDY-909 | |
| The node logs were repeating the message, "NodeRequestSuspiciousSpike suspicious spike has been noticed." | INDY-541 | |
| Unable to catch up the agent if a validator was down. | INDY-941 | |
| The pool was unable to write nyms after BLS keys enabling. | INDY-958 | |
The last pool node is failed to upgrade; during a pool upgrade. |
INDY-953 | |
| State Proof creating is fixed. | INDY-954 | |
| State Proof verifying is fixed. | INDY-949 | |
| Description | Workaround | Ticket |
|---|---|---|
| Signed State implementation | INDY-670 | |
| State Proofs implementation | INDY-790 | |
| Removed all non-Indy branding from the indy-plenum repo. | INDY-829 | |
| Removed all non-Indy branding from the indy-anoncreds repo. | INDY-855 | |
| Removed all non-Indy branding from the indy-node repo. | INDY-830 | |
| Backward compatibility of nodes with state proofs support with old clients. | INDY-877 | |
| Supported rebranding in sovrin package. | INDY-880 | |
| Supported rebranding in Docker and Vagrant environments of sovrin-environments. | INDY-891 | |
| Support of multiple pool networks by Indy Node. | INDY-831 | |
| Support of multiple pool networks by Indy Client (CLI). | INDY-832 | |
| Created proper file folder paths for system service. | INDY-833 | |
| Client needs to be able to send read requests to one Node only. | INDY-927 | |
| Client needs to be able to make sure that we have the latest State Proof. | INDY-928 | |
Known Issue: Node is broken after load_test.py run |
INDY-960 | |
Mapping of all file/folder changes are located here.
-
Send Pool Upgrade command so all nodes upgrade.
-
Sometime later each Steward will need to do the following steps to add their BLS Keys:
From the Validator Node:
- Generate a new 32-byte seed for the bls key (we recommend pwgen):
$ sudo apt install pwgen
$ pwgen -s -y -B 32 1
If the output has a single-quote symbol ('), rerun until it doesn't.
NOTE: This is not your Steward or Node seed.
-
Record the seed somewhere secure.
-
Switch to the indy user.
$ sudo su - indy
- Configure the BLS key.
$ init_bls_keys --name <NODE_ALIAS> --seed '<SEED>'
The --seed is the seed you generated above, and will be used to create the BLS key.
Example with Seed:
$ init_bls_keys --name Node1 --seed 'xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx'
Capture the stdout at the end of the output, which looks like the following, and record it.
BLS Public key is 3AfkzUZVn2WT9mxW2zQXMgX39FXSY5qzohnMVpdvNS5KSath1YG5Ux4u9ubTFTaP6W55XX9Yx7xPWeYos489oyY53WzwNBG7X4o32ESnZ9xacLmNsQLBjqc6oqpWGTbEXv4edFTrZ88n93sEh4fjFhQMumaXxDfWJgd9aj7KCSpf38F
- Exit the indy user.
$ exit
From the CLI Node:
- Manually upgrade the CLI.
$ sudo apt update
$ sudo apt upgrade
- Launch the CLI.
$ indy
The first time running the upgraded CLI you will be prompted to migrate your previous settings. Answer "Yes."
- Connect to the pool.
indy> connect live
- Set your Steward as the signer in the CLI.
indy@live> use DID <Steward DID>
Example:
indy@live> use DID Th7MpTaRZVRYnPiabds81Y
Note: If your DID is not found in the wallet, you will need to use your steward seed:
indy@live> new key with seed <steward_seed>
- Now you will send a node transaction like what you did when you added the node to the pool. You will add the BLS key as a new parameter to the transaction to update the pool ledger with this additional public key. For 'dest', use the same base58 value for this that was used when you initially onboarded your VM onto the provisional pool.
indy@live> send NODE dest=<node_dest> data={'alias':'<node name>','blskey': '<key_generated_by_init_bls_keys>'}
Example:
indy@live> send NODE dest=Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv data={'alias':'Node1','blskey': '3AfkzUZVn2WT9mxW2zQXMgX39FXSY5qzohnMVpdvNS5KSath1YG5Ux4u9ubTFTaP6W55XX9Yx7xPWeYos489oyY53WzwNBG7X4o32ESnZ9xacLmNsQLBjqc6oqpWGTbEXv4edFTrZ88n93sEh4fjFhQMumaXxDfWJgd9aj7KCSpf38F'}
Note: The 'node_dest' value can be found on the node with sudo read_ledger --type pool.
What does BLS stand for?
Boneh-Lynn-Shacham - The BLS signature scheme is used to verify that a signer is authentic.
How does the CLI use State Proof for confirmation?
When the CLI requests information about a transaction it checks the BLS signatures to verify the transaction was written by nodes that are part of the validator pool. The CLI sends a request to one node (arbitrary one). If the Reply doesn't have a State Proof, or the reply is incorrect/invalid, then CLI falls back to sending requests to all Nodes and waiting for f+1 equal Replies.
What if not all nodes in the pool have BLS signing keys for a transaction?
Transactions only get signed if all nodes reaching consensus can sign it (>= n-f Nodes with correct BLS signatures).
Can the bls_seed be any 32 character seed like the Steward seed?
Yes.
When adding a new node to an existing pool where do I find my BLS key?
When initializing your node using init_indy_node the output will display the keys for the node including the BLS key. It can be found in /var/lib/indy/<network_name>/keys/<node_name>/bls_keys/bls_pk file (e.g.: /var/lib/indy/sandbox/keys/Node1/bls_keys/bls_pk)
When you send the transaction to add the new node to the pool it will also contain the BLS key in the transaction shown in this example.
Example of send node command with BLS for 5th node in test pool:
send NODE dest=4Tn3wZMNCvhSTXPcLinQDnHyj56DTLQtL61ki4jo2Loc data= {'client_port': 9702, 'client_ip': '10.0.0.105', 'alias': 'Node5', 'node_ip': '10.0.0.105', 'node_port': 9701, 'services': ['VALIDATOR'], 'blskey':'2RdajPq6rCidK5gQbMzSJo1NfBMYiS3e44GxjTqZUk3RhBdtF28qEABHRo4MgHS2hwekoLWRTza9XiGEMRCompeujWpX85MPt87WdbTMysXZfb7J1ZXUEMrtE5aZahfx6p2YdhZdrArFvTmFWdojaD2V5SuvuaQL4G92anZ1yteay3R'}
Can I use a seed when generating my BLS keys?
For a new node when using init_indy_node if you specify a seed for this script that same seed is used to generate your BLS keys.
For existing nodes being upgraded to 1.2.50, which includes state proofs, you would use the script init_bls_keys where you can specify a 32-character seed on the command line.
init_bls_keys --name <NODE_ALIAS> --seed '<SEED>'
After running init_bls_keys, Stewards of existing nodes will be required use their CLI node to update their validator's information on the ledger to include the bls keys:
send NODE dest=<node_dest> data={'alias':'<node name>', 'blskey': '<key_generated_by_init_bls_keys>'}
Where do I find the configuration file settings?
With file and folder changes the new location for indy_config.py is in the directory location /etc/indy/. The configuration file has a new setting called "NETWORK_NAME" which is used to identify which network and associated genesis transaction files to use, such as sandbox or live. If adding a new node to a live pool, change this setting before initializing the node.
The genesis files are now located in their own directory based off the network name "/var/lib/indy/NETWORK_NAME". The defaults are live, local, and sandbox. Setting the "NETWORK_NAME" in the indy_config.py file will determine which network is used. The default setting in the indy_config.py file is ""NETWORK_NAME=sandbox".
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.1.27 |
| indy-anoncreds | 1.0.10 |
| indy-node | 1.1.43 |
| release version number | 1.1.6 |
| Description | Additional Information | Ticket Number |
|---|---|---|
Added a migration script which eliminates redundant fields with null values from legacy transactions in the domain ledger. |
INDY-895 INDY-869 | |
Added a constraint on version field of POOL_UPGRADE transaction that denies values lower than the current installed version. |
INDY-895 INDY-869 | |
Added prevention of upgrade to a lower version to Upgrader class. |
INDY-895 INDY-869 | |
Fixed a bug in Upgrader class in search for a POOL_UPGRADE cancel transaction for the last POOL_UPGRADE start transaction. |
INDY-895 INDY-869 | |
| Added a test verifying prevention of upgrade to a lower version. | INDY-895 INDY-869 | |
| Corrected existing tests according to introduced prevention of upgrade to a lower version. | INDY-895 INDY-869 | |
| Components | Version Numbers |
|---|---|
| indy-plenum | 1.1.27 |
| indy-anoncreds | 1.0.10 |
| indy-node | 1.1.37 |
| release version number | 1.1.6 |
| Description | Additional Information | Ticket Number |
|---|---|---|
| Stewards can now demote and promote their own nodes. | INDY-410 | |
| Fixed problem with timezones for timestamp in a transaction. | INDY-466 | |
| Limited incoming message size from 128k to 128MB (Temporary solution). | INDY-25 | |
Fixed send CLAIM_DEF command. |
INDY-378 | |
| Masked private information in the CLI logs/output. | INDY-725 | |
| Fixes crashes on ubuntu 17.04. | INDY-8 | |
| Python interpreter is executed in optimized mode. | INDY-211 | |
| Memory leak fixes. | INDY-223 | |
| Some minor stability fixes. | ||
| Fixed a problem with migration during manual upgrades. | INDY-808 | |
| Fixed a problem with the message length limitation. This was a permanent solution of INDY-25. | INDY-765 | |
| Fixed a problem when the pool was writing transactions when more than F nodes were stopped. | INDY -786 | |
| Fixed a problem when the pool was broken after processing lots of transactions at once. | INDY-760 | |
| Fixed a problem when the pool doesn't come back to consensus in cases when less than n-f nodes are alive. | INDY-804 | |
| Partially fixed a problem when the pool responded with outdated data. | INDY-761 | |
| Description | Workaround | Ticket |
|---|---|---|
| New ledger serialization is supported and Leveldb is used as a storage for all ledgers : msgpack is used for the ledger serialization (both transaction log and merkle tree). | ||
The new serialization change created changes to the directory structure for the nodes. The directory name changes are located on a node under .sovrin/data/nodes/<node name>/<directories>. The change removes the ledger files as plain text files and creates them as binary files. A new tool was created to view the ledger entries called read_ledger. This tool also provides you with a count of the transactions. To learn more about this tool and to see a list of available commands, run this as the sovrin user: read_ledger --h |
||
| Genesis transaction files are renamed adding a _genesis to the end of each file name. | ||
| Added the commands to the POOL_UPGRADE to support downgrade and re-installation. However both have issues and should not be used at this time. | INDY-735 INDY-755 | |
| Fixes to upgrade procedure, in particular an issue which caused an infinite loop. | INDY-316 | |
A new CLI command was added to ease the process of rotating a verification key (verkey). The command is change current key or change current key with seed xxxxx. |
||
| Improvements to log messages. | ||
| Publishing only to repo.sovrin.org | ||
| In your sources.list you only need the entry "deb https://repo.sovrin.org/deb xenial stable". | ||
| Implemented a command line tool to provide validator status. | INDY-715 | |
| "Debug" mode for tests was moved to parameter. | INDY-716 | |
| Log levels were changed on some debug level messages to an info level. | INDY-800 | |
| If the pool loses enough nodes and cannot reach consensus when enough nodes become available, the pool will still not reach consensus. | If you restart all the nodes in the pool, it will start reaching consensus again. | INDY-849 |
| Description | Additional Information |
|---|---|
| Identities can be created, and their exercise and evolution can be controlled by ed25519 key pairs (where the public verkey is stored and updated on the ledger). | |
| Identities are fully self-sovereign by default. Guardianship, as described in The Sovrin Provisional Trust Framework, is also possible. | The Sovrin Provisional Trust Framework |
| Identities can have an agent, and the agent endpoint can be declared on the ledger. | |
| Identities are associated with the roles defined in The Sovrin Provisional Trust Framework. Specifically see: trustee, steward, trust anchor, TGB member, and normal identity. | The Sovrin Provisional Trust Framework |
| Identities with special roles have the appropriate permissions. | http://bit.ly/2eRfeIV |
| The ledger supports a core set of transaction types suitable for early workflows | http://bit.ly/2v1OAmO |
| Byzantine consensus is robust and transparent, providing diffuse trust. | |
| Ledger state is enstructured as a Patricia Trie. This is the precondition for implementing Observer nodes and for having high-trust answers with any number of invalid high-trust answers with any number of malicious intermediaries. | |
| The network generates log files that stewards can use as an audit trail. | |
| The network can be upgraded without manual tasks for stewards, using a POOL_UPGRADE transaction. | |
| Batching is functional, allowing bulk submission and reasonable throughput. (Note, however, that very large batches of more than a few thousand transactions are discouraged because they consume network resources unfairly and may cause brownouts for other users.) | |
| Simple safeguards are in place to react to hacking and DDoS attempts. | |
| Terminology in the CLI has been updated to align with usage in the Provisional Trust Framework and the DID specification. In particular, this release is DID-centric, whereas earlier previews were cryptonym-centric. | |
| Registering simple schemas for claims, and then deriving claims and proofs from them, is possible. This supports many of the use cases envisioned for simple verifiable claims, but not the advanced privacy-preserving features offered by Camenisch-Lysyanskya-style schemes. | |
| Description | Additional Information |
|---|---|
| DDO support: Currently, the DID support in Sovrin uses ATTRIB transactions. This limits the complexity and granularity of identity control. Full DDO support will be necessary to support models such as individual signers being empowered to represent a corporation, for example. | |
| Rich schema: W3C verifiable claim schema thinking is still evolving, and we need time to experiment with the best ways to represent certain constructs. Very complex credential types may not yet be representable. | |
| Revocation: Although the cryptographic foundation for revocation is fully implemented in the anoncreds component that Sovrin depends on, the ledger currently lacks a transaction type to update accumulators. This means initial claims won't support revocation. | |
| Observers and state proofs: Although the state trie is present in the ledger, code to return a state proof with each answer from the ledger has not yet been released. This means clients need to hear f+1 consistent answers from validator nodes before they can trust any response from the ledger. The Indy SDK supports this type of response checking, and will cut over to using the more efficient state proof mechanism as soon as Sovrin does. Once we have that mechanism in place, observer nodes become trustworthy because they cannot simulate state proofs. This drastically increases the scale and performance of the system. | Indy SDK |
| Description | Workaround | Ticket |
|---|---|---|
| Stewards demoting a node: Stewards should be able to demote and promote their own node. Demoting a node removes it from the pool so it does not participate in the pool reaching consensus. | INDY-410 | |
| Ubuntu 17.04 support: Currently the CLI and Node are not supported on Ubuntu 17.04. | INDY-8 | |
| Timezones not matching: If a system timezone of the primary does not match that of the other nodes the pool cannot function. | 1. Shut down the sovrin-node service. 2. Use "sudo dpkg-reconfigure tzdata" to change your timezone to UTC. (Select "None of the Above" and then "UTC".) | INDY-466 |
| Installing client as root: Installing the client as root does not allow users to access the transaction files and they are not able to connect to the pool. | INDY-24 | |
Please be aware that the following terminology has changed:
- . keyring is now "wallet"
- . link is now "connection"
- . invitation is now "request"
- . identifier is now "DID"
For emergencies where the Network is under a DOS attack, there is a new command for Trustees only that will put the Pool into a read-only mode.
send POOL\_CONFIG writes=False
or optional
send POOL\_CONFIG writes=False force=False
The following are bool parameters: writes and force. Writes is required. Force is not required. The default for force is False.
With the terminology change from "keyring" to "wallet," the wallets that have already been created will be the client in the .sovrin/keyrings/live directory. When you upgrade, you'll have to start the CLI and connect to live just once for the wallet directory to be created. The new directory is .sovrin/wallets/live. The upgrade does not move your existing wallets from /keyrings/live to /wallets/live since the wallet directory does not exist until you connect for the first time. You will need to move the existing wallets from /keyrings/live to /wallets/live and they will be ready for use.
With the latest builds (which will be on your new stable) you will start to see a time stamp in the ledger for the transactions. It will look like the following:
V4SGRU86Z58d6TV7PBUe6f|1500090331922505|5iBLSL3iujkNsX4RiGN7RjDk5UVtE8GHsn8XHsRsmhUUN9Jz8GEmYtpPrso8e96jo3YafhYSwPkd9QXYA3wYyst4|1500090332|1|Pqs6Tyj1huDsjKYsiF8w4f|~WwZD24Fk4ka7ZY47MPopoo||||||2||
The time stamp is the entry that shows 1500090332.
The 1500090332 is the utc epoch and a converter can be found at this site https://www.epochconverter.com/ so you can see the human readable version of the time.