Wait to mark reports as aggregated until just before committing #374
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
cjpatton
commented
Jul 21, 2023
| max_time: out_share.time, | ||
| checksum: out_share.checksum, | ||
| data: Some(out_share.data), | ||
| })?; |
There was a problem hiding this comment.
We can't aggregate the output shares yet because we still may need to reject in case of replay.
cjpatton
commented
Jul 21, 2023
| // replayed. This would happen if, for example, a fresh report was marked as | ||
| // aggregated in the failed request, then marked as replayed by a retried | ||
| // request. In this case we cannot determine definitively if the report was | ||
| // replayed, so we drop it to be safe. |
There was a problem hiding this comment.
We no longer mark the reports as aggregated, so this logic for handling failed DO requests is no longer necessary. Retry should be safe.
cjpatton
requested review from
armfazh,
bhalleycf,
captain-mota,
chris-wood,
mendess and
will118
as code owners
bhalleycf
reviewed
Jul 21, 2023
bhalleycf
approved these changes
Jul 21, 2023
Loads that require a lot of CPU time can trigger the Workers runtime to reset itself, causing a 500 error. This is most likely to happen when processing the AggregationJobInitReq, as this involves the expensive, VDAF prep initialization step. Some reports may still get marked as aggregated; if the peer retries the request, those reports will get rejected. This problem is so common that we need to make this request idempotent. As a first step, wait to mark the reports as aggregated until just before we have committed to aggregating them. In particular, instead of doing this in `DapReportInitializer::initialize_reports()`, wait until `DapAggregator::put_out_shares()`.
cjpatton
force-pushed
the
cjpatton/defer-mark-aggregated
branch
from
01b1bf9
to
7c08da7
Compare
bhalleycf
approved these changes
Jul 21, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Loads that require a lot of CPU time can trigger the Workers runtime to reset itself, causing a 500 error. This is most likely to happen when processing the AggregationJobInitReq, as this involves the expensive, VDAF prep initialization step. Some reports may still get marked as aggregated; if the peer retries the request, those reports will get rejected.
This problem is so common that we need to make this request idempotent. As a first step, wait to mark the reports as aggregated until just before we have committed to aggregating them. In particular, instead of doing this in
DapReportInitializer::initialize_reports(), wait untilDapAggregator::put_out_shares().