From d0c265bdf576d2a59a8d685afa6385580f6185c6 Mon Sep 17 00:00:00 2001 From: lak hyeon Kim Date: Fri, 1 Sep 2023 00:44:02 +0900 Subject: [PATCH] Fix typo in retention design document (#633) * The figure explains that there is a problem when deleting S3, not S2 --- design/retention.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/design/retention.md b/design/retention.md index 54ce78c25..06262385f 100644 --- a/design/retention.md +++ b/design/retention.md @@ -56,7 +56,7 @@ In conclusion, when a snapshot is created, it should not simply delete the chang This can be expressed as a picture above(when SnapshotInterval=10, SnapshotThreshold=5). Assuming that there are a series of C (changes) in chronological order (ServerSeq), S (snapshots) are being created at intervals of 10 and the synchronized ServerSeq is being recorded in SyncedSeq. At this time, there may be a situation where Client A's synchronization is delayed for some reason. -In this situation, if all previous Cs are deleted when S2 is created, Client A must pull C19 and C20 for synchronization, but it is already deleted and does not exist. This is the reason why the previous changes are deleted based on the minimum synced ServerSeq in the actual implementation. +In this situation, if all previous Cs are deleted when S3 is created, Client A must pull C19 and C20 for synchronization, but it is already deleted and does not exist. This is the reason why the previous changes are deleted based on the minimum synced ServerSeq in the actual implementation. ### How it was implemented as code