I'm trying to use Chainsaw v2 from http://people.apache.org/~sdeboy
I don't want to use zero configuration. Just a simple socketAppender/SocketReceiver combo.
I'm using log4j2 with the following configuration
<?xml version="1.0" encoding="UTF-8"?>
<Configuration status="WARN" >
<Appenders>
<Console name="CONSOLE" target="SYSTEM_OUT">
<PatternLayout pattern="%d{HH:mm:ss.SSS} [%t] %-5level %logger{36} - %msg%n" />
</Console>
<Socket name="SharathZeroConf" host="localhost" port="4445">
</Socket>
</Appenders>
<Loggers>
<Root level="debug">
<AppenderRef ref="SharathZeroConf" />
<AppenderRef ref="CONSOLE" />
</Root>
</Loggers>
</Configuration>
On ChainSaw, I'm selecting the option "Receive events from network" with port 4445.
However chainsaw doesnt log anything.
I've verified that the appender configuration is correct on log4j side by using the builtin socketserver
java -cp ~/.m2/reposiry/org/apache/logging/log4j/log4j-api/2.0.2/log4j-api-2.0.2.jar org.apache.logging.log4j.core.net.server.TcpSocketServer 4445
So the bug must be on chainsaw side. Any pointers #Scott ?
You're right, I got the same issue. I just tried with LogMX instead, and it works like a charm:
I just had to copy Log4j JARs in LogMX lib/ directory (i.e. log4j-api-2.xx.jar and log4j-core-2.xx.jar)
Related
I have a docker image for a Spring Boot app with the log file location as --logging.config=/conf/logs/logback.xml and the log file is as follows.
I am able to get the logs as
kubectl log POD_NAME
But, unable to find the log file when I log in to the pod. Is there any default location where the log file is placed as I haven't mentioned the logging location in the logback.xml file.
Logback file:
<?xml version="1.0" ?>
<configuration>
<property name="server.encoder.pattern"
value="%d{yyyy-MM-dd'T'HH:mm:ss.SSSZ} %-5level : loggerName="%logger{36}" threadName="%thread" txnId="%X{txnId}" %msg%n" />
<property name="metrics.encoder.pattern"
value="%d{yyyy-MM-dd'T'HH:mm:ss.SSSZ} %-5level : %msg%n" />
<!-- Enable LevelChangePropagator for jul-to-slf4j optimization -->
<contextListener class="ch.qos.logback.classic.jul.LevelChangePropagator" />
<appender name="METRICS" class="ch.qos.logback.core.ConsoleAppender">
<encoder>
<pattern>${metrics.encoder.pattern}</pattern>
</encoder>
</appender>
<logger name="appengAluminumMetricsLogger" additivity="false">
<appender-ref ref="METRICS" />
</logger>
<appender name="SERVER" class="ch.qos.logback.core.ConsoleAppender">
<encoder>
<pattern>${server.encoder.pattern}</pattern>
</encoder>
</appender>
<root level="INFO">
<appender-ref ref="SERVER" />
</root>
</configuration>
What you see from kubectl logs is console log from your service. Only console log can be seen like that and this is via docker logs support.
I am running a Log4J2 TCPSocketServer on an edge node in a cluster. All the data nodes send log events to the TCPSocketServer on the edge node and also log locally in the data node using the log4j2.xml configuration file as shown below. The Application Name is stored as a System property and is accessible in the data node or client's log4j2.xml configuration using ${sys:ABC.appname}. How can I send the same appname to the edge node where TCPSocketServer is running using the log4j2.xml. I would be using the same Application Name in the log4j2-server.xml to log events into separate log files just like I am doing locally on data node.
Sample snippet from data node or Client - log4j2.xml :
<?xml version="1.0" encoding="UTF-8"?>
<Configuration status="warn" name="MyApp" >
<Appenders>
<Socket name="socket" host="localhost" port="12345" >
<SerializedLayout />
</Socket>
<File name="MyFile" fileName="/var/log/${sys:ABC.appname}.log" >
<PatternLayout>
<Pattern>%d{ISO8601} %p %c{1.} [%t] %m%n</Pattern>
</PatternLayout>
</File>
</Appenders>
<Loggers>
<Root level="debug">
<AppenderRef ref="socket"/>
<AppenderRef ref="MyFile"/>
</Root>
</Loggers>
</Configuration>
Sample snippet from edge node or Server - log4j2-server.xml
<?xml version="1.0" encoding="UTF-8"?>
<Configuration status="WARN">
<Appenders>
<File name="MyFile" fileName="/var/log/data/${hostName}-<**This is where I would like to see the appname from data node**>.log" >
<PatternLayout>
<Pattern>%d{ISO8601} %p %c{1.} [%t] %m%n</Pattern>
</PatternLayout>
</File>
<Async name="AsyncFile">
<AppenderRef ref="MyFile" />
</Async>
</Appenders>
<Loggers>
<Root level="WARN">
<AppenderRef ref="AsyncFile"/>
</Root>
</Loggers>
</Configuration>
I used ThreadContext to resolve this issue. Its pretty easy to add Thread Context into the codebase and later use Routing to segregate the log events based on the ThreadContext. I followed the example on this link to do the same https://logging.apache.org/log4j/2.x/faq.html#separate_log_files
I am using Log4j2 (RollingFile with routes) in my web application to log application specific logs in a few separate log files. The log4j2.xml file is bundled with in the WAR file.
Log files are generated and logs are generating fine to start with. After some time, it stops writing logs to the existing file and fails creating new folders/files too.
On restart everything resumes working and that is for some time only.
Tried monitoring, couldn't figure out any specific pattern or steps to simulate it.
<Configuration status="error" name="logger">
<Properties>
<Property name="logpath">path_to_log_file</Property>
</Properties>
<Appenders>
<Routing name="RoutingUserLogFile">
<Routes pattern="$${ctx:user}/">
<Route>
<RollingFile name="UserLogFile" fileName="${logpath}/${ctx:user}/MyLogFile.log" filePattern="${logpath}/${ctx:user}/%d{dd-MM-yyyy}-MyLogFile-%i.log.gz">
<PatternLayout>
<Pattern>%d %p %-40C{1.} %m%n</Pattern>
</PatternLayout>
<Policies>
<TimeBasedTriggeringPolicy interval="1" modulate="true" />
<SizeBasedTriggeringPolicy size="4 MB" />
</Policies>
</RollingFile>
</Route>
</Routes>
</Routing>
</Appenders>
<Loggers>
<Root>
<level value="debug" />
<AppenderRef ref="RoutingUserLogFile" level="debug" />
</Root>
</Loggers>
</Configuration>
I need to use log4j to append logs to a socket with UDP. However, I cannot find much on the internet about how to do so. In Log4J, the socketappender uses TCP. So I got log4j 2 beta, but I can't find any examples/documentation on how to use the socketappender, especially for UDP. I would really appreciate it if anyone could give me an example/show me how to use Log4j for UDP. Thanks.
I have been working with log4j 2.0-beta8 and got the UDP appender working with following log4j2.xml file (but note the 2 in the name of the file!):
<?xml version="1.0" encoding="UTF-8"?>
<configuration>
<appenders>
<Socket name="UDP" host="myhostname.com" port="3333" protocol="UDP">
<PatternLayout pattern="%d{yyyy-MM-dd HH:mm:ss,SSS} %-5p %c{1} - %m%n"/>
</Socket>
</appenders>
<loggers>
<root level="info">
<appender-ref ref="UDP"/>
</root>
</loggers>
</configuration>
I have developed a financial data distribution server with Akka, and I want to set logging level for the application. The documentation at akka.io is sketchy at the best; they say there is no more "logging" in Akka and logging is defined through event handlers now. There is also an example of event handler configuration, including logging level:
akka {
event-handlers = ["akka.event.EventHandler$DefaultListener"]
event-handler-level = "INFO"
}
I did that, but though akka.conf is successfully loaded, logging still appears to be at "DEBUG" level. What can be the problem there?
It appears that Akka uses slf4j/logback logging with default configuration. So the (never documented) solution would be to put e.g. the following logback.xml in your classpath:
<?xml version="1.0" encoding="UTF-8"?>
<configuration scan="false" debug="false">
<appender name="stdout" class="ch.qos.logback.core.ConsoleAppender">
<encoder>
<pattern>[%4p] [%d{ISO8601}] [%t] %c{1}: %m%n</pattern>
</encoder>
</appender>
<!-- you can also drop it completely -->
<logger name="se.scalablesolutions" level="DEBUG"/>
<root level="INFO">
<appender-ref ref="stdout"/>
</root>
</configuration>