checks at FullNode side by KryptSai · Pull Request #431 · rubixchain/rubixgoplatform

KryptSai · 2025-12-31T11:55:35Z

In this PR, we are adding the following validations at fullnode side while syncing them from other nodes.

While adding an incoming block to fullnode's levelDB it checks whether, previous block_ID of the incoming block matches with block_ID of the existing latest block or not.
ex: let's say FullNode is having 10blocks, 11th block it is syncing from other node, then 11th block's previous block ID should be same as existing 10th block's ID.
If the incoming block is a transfer block, FullNode checks whether the sender of the incoming block matches with the owner of the latest existing block or not.
ex: let's say FullNode is having 10blocks, 11th block it is syncing from other node, then 11th block's senderDID, should be same as existing 10th block's OwnerDID.

Fullnode handles syncing token in the following 3 ways:
Lets say FullNode has 7 blocks of a particular token t1.

case:If the publisher node publishes any blocks with block numbers greater than 7th(lets say from 8th to 10th)
In this case, FullNode need to sync the incoming block. So while adding each block to Fullnode's levelDB it checks the above conditions. It will add only the above checks pass.
case: If the publisher node publishes a block with block number 7.
In this case, Fullnode checks whether is it the same block which is coming or is it the different block. If it is the different block, then it will add it into a double spend tokens table with DID1 &DID2 having the same block.
case: If the publisher node publishes a block with block number less than 7(lets say 5).
In this case, Fullnode doesn't need to sync. So it checks whether the incoming block(5th block), is it the same as the 5th block which the FullNode is having, If the Fullnode has a different 5th block than what publisher is publishing, It will add the token to double spend tokens table with DID1 & DID2 as the owners.

More possible exploits but not resolving for now(If we assume all the publisher nodes are under our control, no need to resolve this case):
case description:
let's say nodeA has 8 blocks. A has done double spend transaction, A -> B and A -> B' both with blocknumber 9. Now
B->C, C->D , D->E transactions has happend. So currently E is having 12th block. B' is having 9th block only. When Firstly, if FullNode sync from B', FullNode will get till 9th block. After that if E also publishes tokenchain with length 12. Fullnode checks, previous blockID of the incoming 10th block won't match with the already existing 9th block. so we will note down both DIDs as the same token holder for later reference while giving the tokens to them. If we do in this way any node can simply add a dummy block to their levelDB and publish with a different DID as a publisherDID then Fullnode treat it as a double spend transaction and note down both DIDs to distribute the tokens later.

Some more checks to discuss:

lets say a Fullnode has received part token's genesis block, then fullnode should get it's parent tokenID and check whether it has already burnt block of the parent token with it, but the issue would be both, parent tokens burnt block and part tokens genesis block, will be get created at almost same point of time, so there is a chance that fullnode might not received in the same order in which they both get published, so this check might not be a good fit for the fullnodes.

…lidity check at fullnode side.

…rom Did to PublisherDid for the SendTokenDetailsInfo struct; add error message column to Double spenttokenInfo table; handling double spend tokens in all 3 cases where fullnode has already more length token chain, or same length tokenchain or less length tokenchain.

…in fullnodeRBTTable; if synced token is a part token, then checks how many part tokens it's parent token is haviing if it is having more than 2 will add in multiple parenttokensTable.

…r occurs in case of syncing;

…different blockIDs get published in case of regular transfer because of some double spending transaction.

…ock is migrated type in case of token transfer; adding some debug statements in validateSigner function.

…gner;

…ming block number; added all checks for incoming txnBlock; added a seperate function by putting all 3 checks together; added functions to Validate the token content;

…articular block;

…omes a generic function to validate the token content;

…e changes in GetTokenContentAndValidate function;

…d to the publisher in ProcessRegular transfer; calling ValidateIncomingTokenBlock function in SyncFullTokenChainForFullNode function; modifying GetTokenBlockByNumber function to GetFullNodeTokenBlockByNumber function;

…smart contract and NFTs.

…ypes; handling the token owner check in case of smart contract token;

…esqlDB;

… to avoid code duplication;

…etest tokens so that local tokens will be having token_level and token_number similar to faucet tokens or mainnet tokens;

… generatetest tokens so that local tokens will be having token_level and token_number similar to faucet tokens or mainnet tokens;" This reverts commit ee02743.

…as a background process;

…tion

arnabghose997

Ensure the errors are returned properly

arnabghose997 · 2026-02-23T07:52:32Z

-	if len(elems) < 2 {
-		cmd.log.Error(fmt.Sprintf("Invalid token: %v", cmd.token))
-		return
+	if cmd.assetType == "rbt" || cmd.assetType == "RBT" {


Lets' have only "rbt" as value to denote RBT Asset type

arnabghose997 · 2026-02-23T07:56:48Z

+
+	// If no child tokens exist, return nil (as requested)
+	if len(childTokens) == 0 {
+		return nil, nil


Two scenarios:

If this condition is expected to return an error, the error should be returned instead of nil

If this condition is not expected to return an error, instead of returning nil for list, an empty list should be returned

arnabghose997 · 2026-02-23T07:58:35Z

+					}
+
+					if err := c.w.AddFailedTokensToTable(info); err != nil {
+						c.log.Error("Failed to record failed token sync in DB", "token", tokenId, "error", err)


Please return error here

arnabghose997 · 2026-02-23T07:58:54Z

-					c.log.Error("failed to sync token chain for token ", tokenId, "error", err)
-					return fmt.Errorf("failed to get latest block for token %s - may need sync", tokenId)
-
+					handled, _ := c.HandleSyncErrorAsDoubleSpent(err, tokenId, newEvent.AssetType, tokenType, newEvent.PublisherDID, previousOwner, txnBlockOwner, fmt.Sprintf("%s, %s both dids are claiming the same token or same did would have double spent the token,dual ownership issue", previousOwner, txnBlockOwner))


Return the error here

arnabghose997 · 2026-02-23T07:59:56Z

+				incomingBlockID := txnBlockID
+				publishersLatestBlockNumber := currentBlockNumber
+				//get Fullnode block whose block number is publisher side latest blockID
+				fullnodesideBlockBytes, err := c.w.GetFullNodeTokenBlockByNumber(tokenId, tokenType, publishersLatestBlockNumber)


The err is not returned here. Please return it

arnabghose997 · 2026-02-23T08:01:59Z

+			err := c.ReadTokenContentFromPSQL(tokenId, newEvent.AssetType)
+			if err != nil {
+				if err := c.AddTokenContentToPSQL(tokenId, newEvent.AssetType); err != nil {
+					c.log.Error("failed to add token's ipfs content to psql db, err: %v", err)


Please return the error

arnabghose997 · 2026-02-23T08:02:14Z

+				}
+
+				if err := c.w.AddFailedTokensToTable(info); err != nil {
+					c.log.Error("Failed to record failed token sync in DB", "token", tokenId, "error", err)


Please return the error here

arnabghose997 · 2026-02-23T08:02:37Z

+				}
+				// store double spent token info in DoubleSpentTokens table, and remove it from respective tokens table
+				err = c.StoreDoubleSpentTokenInfo(doubleSpentTokenInfo)
+				if err != nil {


error is not returned here

arnabghose997 · 2026-02-23T08:02:57Z

+	// Do all 3 checks, If any check fails, handle the failure here also just before adding the block to leveldb.
+	err = c.ValidateIncomingTokenBlock(*txnBlock, latestTokenBlock, tokenId, p, newEvent.AssetType)
+	if err != nil {
+		handled, _ := c.HandleSyncErrorAsDoubleSpent(err, tokenId, newEvent.AssetType, tokenType, newEvent.PublisherDID, previousOwner, txnBlockOwner, fmt.Sprintf("Incoming block %s is wrong,error %v", txnBlockID, err))


error is not returned here

arnabghose997 · 2026-02-23T08:03:09Z

+	err = c.ReadTokenContentFromPSQL(tokenId, newEvent.AssetType)
+	if err != nil {
+		if err := c.AddTokenContentToPSQL(tokenId, newEvent.AssetType); err != nil {
+			c.log.Error("failed to add token's ipfs content to psql db, err: %v", err)


error is not returned here

…ncErrorAsDoubleSpent function;

…oken genuinity check at quorum side; return error from the HandleSyncErrorAsDoubleSpent handled in processReceivedTokenDetails function;

KryptSai added 30 commits December 29, 2025 17:48

adding checks at fullnode side while intial syncing.

39bad3a

improving the validation condition and log messages.

1c1f1a3

adding reason column in the failed to sync tokens table; improving va…

df9d2bc

…lidity check at fullnode side.

adding fullnodemultiplechildtokenstable; adding parenttokenID column …

ffe82fa

…in fullnodeRBTTable; if synced token is a part token, then checks how many part tokens it's parent token is haviing if it is having more than 2 will add in multiple parenttokensTable.

handling some cases in processReceivedTokenDetails, if there are erro…

c37cb87

…r occurs in case of syncing;

adding signature check at fullnode side;

93673a7

handling the case where 2 blocks with the same block number but with …

832838f

…different blockIDs get published in case of regular transfer because of some double spending transaction.

fixing the issue of token value not getting updated if the genesis bl…

2ff4493

…ock is migrated type in case of token transfer; adding some debug statements in validateSigner function.

adding logs to debug the signature verification failed in validate si…

5ed4513

…gner;

Handled the case where Fullnode already has more blocks than the inco…

f059be0

…ming block number; added all checks for incoming txnBlock; added a seperate function by putting all 3 checks together; added functions to Validate the token content;

Modified the remove token chain api such a way that we can remove a p…

25f62e4

…articular block;

Merge branch 'development' into sai/checks-at-fullnode

7df38fd

modifying the ValidateNewTokenContent function such a way that it bec…

2771c01

…omes a generic function to validate the token content;

Merge branch 'development' into sai/checks-at-fullnode

461bc29

Removed code related to adding token content to psql in fullnode; som…

b7a8b4e

…e changes in GetTokenContentAndValidate function;

Merge branch 'development' into sai/checks-at-fullnode

67951df

adding MaxPossiblePartsIndexByMaxDecimalPlaces function;

d8cde7c

Added startIndex as a flag with default value -1 to generate test rbt;

f617a92

handled a small error in GenerateRBT function;

d36594a

Merge branch 'development' into sai/checks-at-fullnode

95a711f

removing unnecessary code

78a8683

Adding missing block checks and validating incoming block in case of …

686dfd2

…smart contract and NFTs.

handling invalid token error in dumptokenchain for non native asset t…

d6052c5

…ypes; handling the token owner check in case of smart contract token;

In processContractExecution adding token details to FullNode's postgr…

c4f8da2

…esqlDB;

handling errors of SyncFullTokenChainForFullNode function;

c44f8a5

adding HandleSyncErrorAsDoubleSpent to handle sync token chain errors…

127be76

… to avoid code duplication;

updated tokenMap according to the recent white paper; updated generat…

ee02743

…etest tokens so that local tokens will be having token_level and token_number similar to faucet tokens or mainnet tokens;

Revert "updated tokenMap according to the recent white paper; updated…

00ee5e6

… generatetest tokens so that local tokens will be having token_level and token_number similar to faucet tokens or mainnet tokens;" This reverts commit ee02743.

KryptSai added 4 commits February 17, 2026 11:43

Merge branch 'development' into sai/checks-at-fullnode

ad48f48

commented out RetryFailedTOSyncTokens for fullnode which was running …

609e630

…as a background process;

modified token genuinity check according to new test net token genera…

3cda0bf

…tion

Merge branch 'development' into sai/checks-at-fullnode

f322858

KryptSai marked this pull request as ready for review February 23, 2026 06:24

removing unnecessary loggers and commented out code.

439556d

KryptSai requested review from Allen-Cherian, arnabghose997 and maneeSHA-256 February 23, 2026 07:05

removed a debug statement.

f1e11e1

arnabghose997 requested changes Feb 23, 2026

View reviewed changes

KryptSai added 2 commits February 23, 2026 23:10

return error when it has not been handled in case of calling HandleSy…

992b45a

…ncErrorAsDoubleSpent function;

added block number check for genesis blocks at fullnode side; added t…

6c5319c

…oken genuinity check at quorum side; return error from the HandleSyncErrorAsDoubleSpent handled in processReceivedTokenDetails function;

Allen-Cherian closed this Jun 9, 2026

Navigation Menu

Uh oh!

checks at FullNode side - #431

checks at FullNode side#431
KryptSai wants to merge 38 commits into
developmentfrom
sai/checks-at-fullnode

KryptSai commented Dec 31, 2025 •

edited

Loading

Uh oh!

arnabghose997 left a comment

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

arnabghose997 Feb 23, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Conversation

KryptSai commented Dec 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

arnabghose997 left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

KryptSai commented Dec 31, 2025 •

edited

Loading