sodonnel commented on code in PR #6520:
URL: https://github.com/apache/ozone/pull/6520#discussion_r1565542323
##########
hadoop-hdds/client/src/main/java/org/apache/hadoop/hdds/scm/storage/ECBlockOutputStream.java:
##########
@@ -115,20 +115,43 @@ ContainerCommandResponseProto> executePutBlock(boolean
close,
}
BlockData checksumBlockData = null;
+ boolean foundStripeChecksum = false;
+ BlockID blockID = null;
//Reverse Traversal as all parity will have checksumBytes
for (int i = blockData.length - 1; i >= 0; i--) {
BlockData bd = blockData[i];
if (bd == null) {
continue;
}
+ if (blockID == null) {
+ // store the BlockID for logging
+ blockID = bd.getBlockID();
+ }
List<ChunkInfo> chunks = bd.getChunks();
- if (chunks != null && chunks.size() > 0 && chunks.get(0)
- .hasStripeChecksum()) {
- checksumBlockData = bd;
- break;
+ if (chunks != null && chunks.size() > 0) {
+ if (chunks.get(0).hasStripeChecksum()) {
+ checksumBlockData = bd;
+ foundStripeChecksum = true;
+ break;
+ } else {
+ ChunkInfo chunk = chunks.get(0);
+ LOG.info("The first chunk in block with index {} does not have
stripeChecksum. BlockID: {}, Block size: {}." +
Review Comment:
The checksum data is only stored in replica indexes 1 and the parity
indexes. I think the `BlockData[]` array could hold any indexes, so it would be
expected for some of them to not have it. Therefore this log might be a bit
noisy and cause confusion.
I think it would be OK to make this debug, or just depend on the log you
added below that warns if there was none found in any indexes.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
