From 4b6ce0fcbf68f570831c56ba8dcb4fe8fdbf2b28 Mon Sep 17 00:00:00 2001 From: stack Date: Thu, 12 Dec 2019 13:51:35 -0800 Subject: [PATCH] HBASE-23572 In 'HBCK Report', distringush between live, dead, and unknown servers Signed-off-by: Sean Busbey --- .../resources/hbase-webapps/master/hbck.jsp | 70 ++++++++++++------- 1 file changed, 46 insertions(+), 24 deletions(-) diff --git a/hbase-server/src/main/resources/hbase-webapps/master/hbck.jsp b/hbase-server/src/main/resources/hbase-webapps/master/hbck.jsp index f89aac8bce6..e9a8658286a 100644 --- a/hbase-server/src/main/resources/hbase-webapps/master/hbck.jsp +++ b/hbase-server/src/main/resources/hbase-webapps/master/hbck.jsp @@ -31,13 +31,14 @@ <%@ page import="org.apache.hadoop.hbase.client.RegionInfo" %> <%@ page import="org.apache.hadoop.hbase.master.HbckChore" %> <%@ page import="org.apache.hadoop.hbase.master.HMaster" %> +<%@ page import="org.apache.hadoop.hbase.master.ServerManager" %> <%@ page import="org.apache.hadoop.hbase.ServerName" %> <%@ page import="org.apache.hadoop.hbase.util.Bytes" %> <%@ page import="org.apache.hadoop.hbase.util.Pair" %> <%@ page import="org.apache.hadoop.hbase.master.CatalogJanitor" %> <%@ page import="org.apache.hadoop.hbase.master.CatalogJanitor.Report" %> <% - HMaster master = (HMaster) getServletContext().getAttribute(HMaster.MASTER); + final HMaster master = (HMaster) getServletContext().getAttribute(HMaster.MASTER); pageContext.setAttribute("pageTitle", "HBase Master HBCK Report: " + master.getServerName()); HbckChore hbckChore = master.getHbckChore(); Map>> inconsistentRegions = null; @@ -60,13 +61,13 @@ String iso8601end = startTimestamp == 0? "-1": zdt.format(DateTimeFormatter.ISO_OFFSET_DATE_TIME); CatalogJanitor cj = master.getCatalogJanitor(); CatalogJanitor.Report report = cj == null? null: cj.getLastReport(); + final ServerManager serverManager = master.getServerManager(); %>
- <% if (!master.isInitialized()) { %>
@@ -78,7 +79,7 @@
-

This page displays two reports: the 'HBCK Chore Report' and the 'CatalogJanitor Consistency Issues' report. Only titles show if there are no problems to report. Note some conditions are transitory as regions migrate.

+

This page displays two reports: the HBCK Chore Report and the CatalogJanitor Consistency Issues report. Only report titles show if there are no problems to list. Note some conditions are transitory as regions migrate. See below for how to run reports. ServerNames will be links if server is live, italic if dead, and plain if unknown.

@@ -98,8 +99,6 @@
- - <% if (inconsistentRegions != null && inconsistentRegions.size() > 0) { %>
@@ -108,10 +107,11 @@

- There are three cases: 1. Master thought this region opened, but no regionserver reported it (Fix: use assigns + There are three cases: 1. Master thought this region opened, but no regionserver reported it (Fix: use assign command); 2. Master thought this region opened on Server1, but regionserver reported Server2 (Fix: - need to check the server is still exist. If not, schedule SCP for it. If exist, restart Server2 and Server1): - 3. More than one regionservers reported opened this region (Fix: restart the RegionServers). + need to check the server still exists. If not, schedule ServerCrashProcedure for it. If exists, + restart Server2 and Server1): + 3. More than one regionserver reports opened this region (Fix: restart the RegionServers). Notice: the reported online regionservers may be not right when there are regions in transition. Please check them in regionserver's web UI. @@ -123,15 +123,14 @@ Location in META Reported Online RegionServers - <% for (Map.Entry>> entry : inconsistentRegions.entrySet()) { %> + <% for (Map.Entry>> entry : inconsistentRegions.entrySet()) {%> <%= entry.getKey() %> - <%= entry.getValue().getFirst() %> - <%= entry.getValue().getSecond().stream().map(ServerName::getServerName) - .collect(Collectors.joining(", ")) %> + <%= formatServerName(master, serverManager, entry.getValue().getFirst()) %> + <%= entry.getValue().getSecond().stream().map(s -> formatServerName(master, serverManager, s)). + collect(Collectors.joining(", ")) %> <% } %> -

<%= inconsistentRegions.size() %> region(s) in set.

<% } %> @@ -142,14 +141,6 @@

Orphan Regions on RegionServer

-

- - The below are Regions we've lost account of. To be safe, run bulk load of any data found in these Region orphan directories back into the HBase cluster. - First make sure hbase:meta is in healthy state; run 'hbkc2 fixMeta' to be sure. Once this is done, per Region below, run a bulk - load -- '$ hbase completebulkload REGION_DIR_PATH TABLE_NAME' -- and then delete the desiccated directory content (HFiles are removed upon successful load; all that is left are empty directories - and occasionally a seqid marking file). - -

@@ -159,10 +150,9 @@ <% for (Map.Entry entry : orphanRegionsOnRS.entrySet()) { %> - + <% } %> -

<%= orphanRegionsOnRS.size() %> region(s) in set.

<%= entry.getKey() %><%= entry.getValue() %><%= formatServerName(master, serverManager, entry.getValue()) %>
<% } %> @@ -173,7 +163,14 @@

Orphan Regions on FileSystem

- +

+ + The below are Regions we've lost account of. To be safe, run bulk load of any data found in these Region orphan directories back into the HBase cluster. + First make sure hbase:meta is in a healthy state; run hbck2 fixMeta to be sure. Once this is done, per Region below, run a bulk + load -- $ hbase completebulkload REGION_DIR_PATH TABLE_NAME -- and then delete the desiccated directory content (HFiles are removed upon + successful load; all that is left are empty directories and occasionally a seqid marking file). + +

@@ -301,3 +298,28 @@ + +<%! +/** + * Format serverName for display. + * If a live server reference, make it a link. + * If dead, make it italic. + * If unknown, make it plain. + */ +private static String formatServerName(HMaster master, + ServerManager serverManager, ServerName serverName) { + String sn = serverName.toString(); + if (serverManager.isServerOnline(serverName)) { + int infoPort = master.getRegionServerInfoPort(serverName); + if (infoPort > 0) { + return "" + sn + ""; + } else { + return "" + sn + ""; + } + } else if (serverManager.isServerDead(serverName)) { + return "" + sn + ""; + } + return sn; +} +%>
Region Encoded Name