This is an automated email from the ASF dual-hosted git repository.
siddhant pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/ozone.git
The following commit(s) were added to refs/heads/master by this push:
new 366d07461a HDDS-10859. Improve error messages when decommission and
maintenance fail-early (#6678)
366d07461a is described below
commit 366d07461af11a62b2b19b12ea40000a4fed589b
Author: Tejaskriya <[email protected]>
AuthorDate: Thu May 16 14:46:59 2024 +0530
HDDS-10859. Improve error messages when decommission and maintenance
fail-early (#6678)
---
.../hdds/scm/node/NodeDecommissionManager.java | 21 ++++++++++-----------
1 file changed, 10 insertions(+), 11 deletions(-)
diff --git
a/hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/node/NodeDecommissionManager.java
b/hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/node/NodeDecommissionManager.java
index 96628aa6cf..df224d1d44 100644
---
a/hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/node/NodeDecommissionManager.java
+++
b/hadoop-hdds/server-scm/src/main/java/org/apache/hadoop/hdds/scm/node/NodeDecommissionManager.java
@@ -327,7 +327,6 @@ public class NodeDecommissionManager {
boolean decommissionPossible = checkIfDecommissionPossible(dns, errors);
if (!decommissionPossible) {
LOG.error("Cannot decommission nodes as sufficient node are not
available.");
- errors.add(new DatanodeAdminError("AllHosts", "Sufficient nodes are
not available."));
return errors;
}
} else {
@@ -436,10 +435,11 @@ public class NodeDecommissionManager {
}
int reqNodes = cif.getReplicationConfig().getRequiredNodes();
if ((inServiceTotal - numDecom) < reqNodes) {
- LOG.info("Cannot decommission nodes. Tried to decommission {}
nodes of which valid nodes = {}. " +
- "Cluster state: In-service nodes = {}, nodes required for
replication = {}. " +
- "Failing due to datanode : {}, container : {}",
- dns.size(), numDecom, inServiceTotal, reqNodes, dn, cid);
+ String errorMsg = "Insufficient nodes. Tried to decommission " +
dns.size() +
+ " nodes of which " + numDecom + " nodes were valid. Cluster
has " + inServiceTotal +
+ " IN-SERVICE nodes, " + reqNodes + " of which are required for
minimum replication. ";
+ LOG.info(errorMsg + "Failing due to datanode : {}, container :
{}", dn, cid);
+ errors.add(new DatanodeAdminError("AllHosts", errorMsg));
return false;
}
}
@@ -495,7 +495,6 @@ public class NodeDecommissionManager {
boolean maintenancePossible = checkIfMaintenancePossible(dns, errors);
if (!maintenancePossible) {
LOG.error("Cannot put nodes to maintenance as sufficient node are not
available.");
- errors.add(new DatanodeAdminError("AllHosts", "Sufficient nodes are
not available."));
return errors;
}
} else {
@@ -600,11 +599,11 @@ public class NodeDecommissionManager {
minInService = maintenanceReplicaMinimum;
}
if ((inServiceTotal - numMaintenance) < minInService) {
- LOG.info("Cannot enter nodes into maintenance. Tried to start
maintenance for {} nodes " +
- "of which valid nodes = {}. " +
- "Cluster state: In-service nodes = {}, nodes required for
replication = {}. " +
- "Failing due to datanode : {}, container : {}",
- dns.size(), numMaintenance, inServiceTotal, minInService, dn,
cid);
+ String errorMsg = "Insufficient nodes. Tried to start maintenance
for " + dns.size() +
+ " nodes of which " + numMaintenance + " nodes were valid.
Cluster has " + inServiceTotal +
+ " IN-SERVICE nodes, " + minInService + " of which are required
for minimum replication. ";
+ LOG.info(errorMsg + "Failing due to datanode : {}, container :
{}", dn, cid);
+ errors.add(new DatanodeAdminError("AllHosts", errorMsg));
return false;
}
}
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
