I’m not an expert on Java EE, but from what I’ve been reading on the Internet, some versions of Glassfish happen exactly the same problem, but nobody knows why. So I’ll summarize the steps I found on the internet and make sure it works in your case.
01) Update your Glassfish, if it is the last version, go back to a previous one.
02) Check if you have more than 2gb RAM, some people report 2gb RAM problem with Glassfish on Windows 7/Server.
03) Start using Tomcat.
04) Check the routines of the software if it is your development, there are good practices that avoid this problem, for example, do not consult the bank in Gets and Sets.
That’s all I found on the Internet, as I said before, I’m not an expert on Java, I’m just trying to help.
I’m not an expert on Java EE. But you could put some more information on the system. For example, if you use some framework, which database, Glassfish version... One more thing, you’ve tried it with Tomcat?
– Pedro Soares
good, Firebird database , use Hibernate jpa and jsf
– Felipe
What version of Glassflish? Have you tried with Tomcat?
– Pedro Soares
to using glassfish 4, not tested with Tomcat yet
– Felipe
What is the configuration of the machine used?
– Pedro Soares
is an i7 with 16 GB of memory
– Felipe
Dude, if @Pedrosoares' suggestion doesn’t work, I believe you’ll need to gather some more data about this problem. I know little about Glassfish but, if I remember correctly, it has an administration console with monitoring tools. I think it might be a cool tool to understand what’s going on.
– Miguel Fontes
The first step is for you to monitor what is happening in the application during the crash. Is it a JVM crash, that is, is the process finished from scratch? Does the application simply stop responding? How much memory is it consuming at the time of locking? You can use tools like Jvisualvm that come in JDK to monitor the application in real time. Extract a thread dump, that is, a picture of the Traces stack of all threads running at the time of locking.
– utluiz
Anyway, you need to start an investigation and not keep kicking the problem. This can be caused by any upgrade someone has made to the system, updating in some library, abrupt increase of data loaded in memory (maybe some user has imported thousands of new records into the database) and so on. This occurs when someone performs a specific function?
– utluiz