cancel
Showing results for 
Search instead for 
Did you mean: 

EFM Status Issue

Level 3 Adventurer

EFM Status Issue

Hi Guys,

We have configured EFM CLuster with 1 master,2 slave and a witness.The whole setup was wotking as expected.But not,in a while,if we check the efm status ,the servers are not visible.But the replication is happening and the servers are in sync(checked in pg_stat_replication).The efm and edb services are running good.

 

Upon check the efm logs,we found as below.Kindly suggest if any solution.

Witness log:

12/19/18 6:00:48 PM org.jgroups.blocks.cs.TcpServer$Acceptor run WARNING: JGRP000006: failed accepting connection from peer
java.io.EOFException
at java.io.DataInputStream.readFully(DataInputStream.java:197)
at org.jgroups.blocks.cs.TcpConnection.readPeerAddress(TcpConnection.java:235)
at org.jgroups.blocks.cs.TcpConnection.<init>(TcpConnection.java:51)
at org.jgroups.blocks.cs.TcpServer$Acceptor.handleAccept(TcpServer.java:126)
at org.jgroups.blocks.cs.TcpServer$Acceptor.run(TcpServer.java:111)
at java.lang.Thread.run(Thread.java:748)

12/19/18 6:01:07 PM org.jgroups.blocks.cs.TcpServer$Acceptor run WARNING: JGRP000006: failed accepting connection from peer
java.io.EOFException
at java.io.DataInputStream.readFully(DataInputStream.java:197)
at org.jgroups.blocks.cs.TcpConnection.readPeerAddress(TcpConnection.java:235)
at org.jgroups.blocks.cs.TcpConnection.<init>(TcpConnection.java:51)
at org.jgroups.blocks.cs.TcpServer$Acceptor.handleAccept(TcpServer.java:126)
at org.jgroups.blocks.cs.TcpServer$Acceptor.run(TcpServer.java:111)
at java.lang.Thread.run(Thread.java:748)

[root@awsgpbdfemnpr01 efm-3.2]#

 

Master log:

[root@awsgpbdfemdpr01 efm-3.2]# more efm.log
12/19/18 5:58:07 PM com.enterprisedb.efm.admin.AdminServerThread processRequest INFO: Admin server received status request.

12/19/18 5:58:07 PM com.enterprisedb.efm.admin.AdminServerThread processRequest INFO: Received param: efm

12/19/18 5:58:07 PM com.enterprisedb.efm.nodes.EfmNode checkClusterStatus INFO: Sending status message to all nodes: node_status

12/19/18 5:58:07 PM com.enterprisedb.efm.nodes.EfmAgent handleStatusCall INFO: responding to db status request with status: Idle%%192.168.216.100%%true%%false%% %%fals
e

12/19/18 5:58:08 PM org.jgroups.JChannel up SEVERE: JGRP000018: failed passing message to up-handler
java.lang.OutOfMemoryError: Java heap space

12/19/18 5:58:18 PM com.enterprisedb.efm.exec.LoggerStream out WARNING: Exception in thread "AdminServerThread"

12/19/18 5:58:18 PM com.enterprisedb.efm.exec.LoggerStream out WARNING: java.lang.OutOfMemoryError: Java heap space

 

Slave log:

[root@awsgpbdfemdpr03 efm-3.2]# tail -200 efm.log
12/19/18 5:56:58 PM com.enterprisedb.efm.exec.LoggerStream out WARNING: Exception in thread "AdminServerThread"

12/19/18 5:56:58 PM com.enterprisedb.efm.exec.LoggerStream out WARNING: java.lang.OutOfMemoryError: Java heap space

 

Thanks and Regards,

Manisha

 

3 REPLIES
Highlighted
EDB Team Member

Re: EFM Status Issue

Hi Mahisha,

 

This issue seems to be regarding to the java heap size as per the log messages.

For understanding and invetsigating more on this, can you share the following :

 

1. EFM properties file.

2. Steps you used for starting the EFM services.

3. Cluster status.

EDB Team Member

Re: EFM Status Issue

We're looking into the memory issue now.

 

I recommend you stop the agents, change this property in the properties files, and restart the agents:

jvm.options=-Xmx128m

 

If you're not using OpenJDK for your Java, you might want to switch to that. It's what we use for testing, including trying to reproduce a memory leak.

 

Cheers,

Bobby

 

EDB Team Member

Re: EFM Status Issue

Oh, and to be really clear about it -- stopping the efm agents won't affect the databases in any way.

 

Cheers,

Bobby