From 6e243bde72ede15f4b7726986ad18420bed40966 Mon Sep 17 00:00:00 2001 From: Brandon Pitman Date: Fri, 4 Oct 2024 16:13:01 -0700 Subject: [PATCH] Remove replay check from "input share validation" section. The related implementation note is removed, too -- it is nowadays duplicative of the text in the Replay Protection section. I think this was an intended part of #582, based on the commit text of that PR. --- draft-ietf-ppm-dap.md | 12 ------------ 1 file changed, 12 deletions(-) diff --git a/draft-ietf-ppm-dap.md b/draft-ietf-ppm-dap.md index 95a8cff3..6ecfc0f2 100644 --- a/draft-ietf-ppm-dap.md +++ b/draft-ietf-ppm-dap.md @@ -1819,18 +1819,6 @@ following checks: the same. If so, the Aggregator MUST mark the input share as invalid with error `invalid_message`. -1. Check if the report has been previously aggregated. If so, the input share - MUST be marked as invalid with the error `report_replayed`. A report is - considered aggregated if its contribution would be included in a relevant - collection job. - - * Implementation note: To detect replay attacks, each Aggregator is required - to keep track of the set of reports it has processed for a given task. - Because honest Clients choose the report ID at random, it is sufficient to - store the set of IDs of processed reports. However, implementations may - find it helpful to track additional information, like the timestamp, so - that the storage used for anti-replay can be sharded efficiently. - 1. If the report pertains to a batch that was previously collected, then the input share MUST be marked as invalid with error `batch_collected`.