This is an automated email from the ASF dual-hosted git repository.
alexey pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/kudu.git
The following commit(s) were added to refs/heads/master by this push:
new 8770e72 docs: add a blurb about RF=1 tables when doing a rolling
restart
8770e72 is described below
commit 8770e722fa99fb3d4c971d78bf8e95e08e2be68a
Author: Andrew Wong <[email protected]>
AuthorDate: Thu May 21 17:20:59 2020 -0700
docs: add a blurb about RF=1 tables when doing a rolling restart
The documented steps don't mention anything about how a RF=1 table will
not make progress in quiescing. We don't expect such tables to be very
common; nevertheless it'd be nice to at least mention the limitation.
While not yet easily automated since there isn't a convenient way to get
the tables that have RF=1, this adds a warning and a recommendation of
tooling that might help.
A stage version can be found here:
https://github.com/andrwng/kudu/blob/docs-1.12/docs/administration.adoc#rolling_restart
Change-Id: I2d43d0331dc6d67a9ccb9e84c38b302eed3ae6a6
Reviewed-on: http://gerrit.cloudera.org:8080/15976
Tested-by: Kudu Jenkins
Reviewed-by: Alexey Serbin <[email protected]>
---
docs/administration.adoc | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/docs/administration.adoc b/docs/administration.adoc
index 2c12ac7..0eb1cfd 100644
--- a/docs/administration.adoc
+++ b/docs/administration.adoc
@@ -1557,6 +1557,13 @@ perform such a "rolling restart", perform the following
sequence:
tablet server.
. Repeat these steps for all tablet servers in the cluster.
+WARNING: If any tables in the cluster have a replication factor of 1, some
+quiescing tablet servers will never become fully quiesced, as single-replica
+tablets will not naturally relinquish leadership. If such tables exist, use the
+`kudu cluster rebalance` tool to move replicas of these tables away from the
+quiescing tablet server by specifying the `--ignored_tservers`,
+`--move_replicas_from_ignored_tservers`, and `--tables` options.
+
NOTE: If running with <<rack_awareness,rack awareness>>, the above steps can be
performed restarting multiple tablet servers within a single rack at the same
time. Users should use `ksck` to ensure the location assignment policy is
