This is an automated email from the ASF dual-hosted git repository.
zhangyifan 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 8cb4b6385 [docs] Update tablet server decommission documentation
8cb4b6385 is described below
commit 8cb4b6385f680e65be9702e30d7a709063999d81
Author: zhangyifan27 <[email protected]>
AuthorDate: Mon Jul 3 18:07:41 2023 +0800
[docs] Update tablet server decommission documentation
This patch updates the last step in tablet server decommission section,
mentioning using `kudu tserver unregister` tool to remove a dead tablet
server from the cluster in newer Kudu versions.
Change-Id: I60043175ebdc0484ba655788aa7f5fb062e2a4ae
Reviewed-on: http://gerrit.cloudera.org:8080/20153
Tested-by: Kudu Jenkins
Reviewed-by: Yingchun Lai <[email protected]>
Reviewed-by: Marton Greber <[email protected]>
---
docs/administration.adoc | 12 +++++++++---
1 file changed, 9 insertions(+), 3 deletions(-)
diff --git a/docs/administration.adoc b/docs/administration.adoc
index ed9a88727..3ec10d0bd 100644
--- a/docs/administration.adoc
+++ b/docs/administration.adoc
@@ -1611,9 +1611,15 @@ the cluster and `ksck` is healthy before shutting down
the next.
healthy state.
. The decommissioned tablet server can be brought offline.
. To completely remove it from the cluster so `ksck` shows the cluster as
- completely healthy, restart the masters. In the case of a single master,
- this will cause cluster downtime. With multi-master, restart the masters in
- sequence to avoid cluster downtime.
+ completely healthy, run the `kudu tserver unregister` tool.
++
+NOTE: From Kudu version 1.16.0 onwards, the `kudu tserver unregister` tool has
been
+added that unregisters a dead tablet server from the cluster. In Kudu versions
that
+do not support this tool, all the Kudu masters must be restarted to completely
remove
+the dead tablet server from the cluster so `ksck` shows the cluster as
completely healthy.
+In the case of a single master, this will cause cluster downtime. With
multi-master,
+restart the masters in sequence to avoid cluster downtime.
++
In Kudu versions that do not support the above tooling, different steps must be
followed to decommission a tablet server:
