You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@kudu.apache.org by ad...@apache.org on 2019/09/10 03:54:44 UTC
[kudu] 01/02: [docs] fix typo in location-awareness design doc
This is an automated email from the ASF dual-hosted git repository.
adar pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/kudu.git
commit 20dffb3659d7e577bd421d89c938d72be0b10b93
Author: Alexey Serbin <al...@apache.org>
AuthorDate: Mon Sep 9 09:19:51 2019 -0700
[docs] fix typo in location-awareness design doc
Fixed scan mode name: READ_CLOSEST --> CLOSEST_REPLICA
Change-Id: I28f7ea1137561fbb94096d6b41b69adeec19693f
Reviewed-on: http://gerrit.cloudera.org:8080/14196
Tested-by: Kudu Jenkins
Reviewed-by: Andrew Wong <aw...@cloudera.com>
---
docs/design-docs/location-awareness.md | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/docs/design-docs/location-awareness.md b/docs/design-docs/location-awareness.md
index 58bb688..a3677de 100644
--- a/docs/design-docs/location-awareness.md
+++ b/docs/design-docs/location-awareness.md
@@ -95,7 +95,7 @@ Requirements
connection between locations is not reliable or has high latency.
- Support location-aware clients so at minimum a location-aware client
- performing a READ\_CLOSEST scan will read from the same location if
+ performing a CLOSEST\_REPLICA scan will read from the same location if
possible.
- Enhance the rebalancing tool to preserve placement policies if they
hold and to move replicas to enforce placement policies if they are
@@ -250,7 +250,7 @@ When the client connects to the cluster, it will include the hostname or
IP of the machine it is running on. The master will assign it a location
and return that to the client. It will also return location information
to the client about tablet servers. The client will use this information
-to select a server to read from in READ\_CLOSEST mode. It will prefer a
+to select a server to read from in CLOSEST\_REPLICA mode. It will prefer a
local server, as it does now, and then after that prefer servers in the
same location (with ties by random choice). This procedure generalizes
to a hierarchical scheme: the client would prefer servers having the