Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[BACKPORT] Do not suppress exceptions in ICMP ping task #13826

Merged
merged 1 commit into from Sep 24, 2018
Merged

[BACKPORT] Do not suppress exceptions in ICMP ping task #13826

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
64 changes: 30 additions & 34 deletions hazelcast/src/main/java/com/hazelcast/internal/cluster/impl/ClusterHeartbeatManager.java
View file
Open in desktop
Expand Up @@ -741,25 +741,21 @@ private class PingTask
}

public void run() {
try {
Address address = member.getAddress();
logger.warning(format("%s will ping %s", node.getThisAddress(), address));
for (int i = 0; i < MAX_PING_RETRY_COUNT; i++) {
if (doPing(address, Level.INFO)) {
return;
}
Address address = member.getAddress();
logger.warning(format("%s will ping %s", node.getThisAddress(), address));
for (int i = 0; i < MAX_PING_RETRY_COUNT; i++) {
if (doPing(address, Level.INFO)) {
return;
}
// host not reachable
String reason = format("%s could not ping %s", node.getThisAddress(), address);
logger.warning(reason);
clusterService.suspectMember(member, reason, true);
} catch (Throwable ignored) {
ignore(ignored);
}

// host not reachable
String reason = format("%s could not ping %s", node.getThisAddress(), address);
logger.warning(reason);
clusterService.suspectMember(member, reason, true);
}

boolean doPing(Address address, Level level)
throws IOException {
boolean doPing(Address address, Level level) {
try {
if (address.getInetAddress().isReachable(null, icmpTtl, icmpTimeoutMillis)) {
String msg = format("%s pinged %s successfully", node.getThisAddress(), address);
Expand All @@ -769,6 +765,10 @@ boolean doPing(Address address, Level level)
} catch (ConnectException ignored) {
// no route to host, means we cannot connect anymore
ignore(ignored);
} catch (IOException e) {
if (logger.isFinestEnabled()) {
logger.finest("Failed while pinging " + address, e);
}
}
return false;
}
Expand All @@ -783,29 +783,25 @@ private class PeriodicPingTask
}

public void run() {
try {
Address address = member.getAddress();
logger.fine(format("%s will ping %s", node.getThisAddress(), address));
if (doPing(address, Level.FINE)) {
boolean pingRestored = (icmpFailureDetector.heartbeat(member) > 0);
if (pingRestored) {
quorumService.onPingRestored(member);
}
return;
Address address = member.getAddress();
logger.fine(format("%s will ping %s", node.getThisAddress(), address));
if (doPing(address, Level.FINE)) {
boolean pingRestored = (icmpFailureDetector.heartbeat(member) > 0);
if (pingRestored) {
quorumService.onPingRestored(member);
}
return;
}

icmpFailureDetector.logAttempt(member);
quorumService.onPingLost(member);
icmpFailureDetector.logAttempt(member);
quorumService.onPingLost(member);

// host not reachable
String reason = format("%s could not ping %s", node.getThisAddress(), address);
logger.warning(reason);
// host not reachable
String reason = format("%s could not ping %s", node.getThisAddress(), address);
logger.warning(reason);

if (!icmpFailureDetector.isAlive(member)) {
clusterService.suspectMember(member, reason, true);
}
} catch (Throwable ignored) {
ignore(ignored);
if (!icmpFailureDetector.isAlive(member)) {
clusterService.suspectMember(member, reason, true);
}
}
}
Expand Down