Shulou(Shulou.com)06/03 Report--

How to answer the six major questions of WebLogic and JVM, many novices are not very clear about this, in order to help you solve this problem, the following editor will explain in detail for you, people with this need can come to learn, I hope you can gain something.

Under all kinds of JVM, the standard JAVA operation-Xms-Xmx parameter is provided. When you set-Xms512m-Xmx1024m in weblogic and observe from the backend console, weblogic only uses the initial heap size of 512m. When there is a memory leak in the application or when a large number of users use it, 512mheap cannot meet the needs, but weblogic will not expand the heap size in 512m-1024m. In this case, weblogic will crash due to memory exhaustion. What's the use of setting-Xmx1024m?

Six questions about WebLogic and JVM

◆ question 1: memory parameter setting

Under all kinds of JVM, the standard JAVA operation-Xms-Xmx parameter is provided. When you set-Xms512m-Xmx1024m in weblogic and observe from the backend console, weblogic only uses the initial heap size of 512m. When there is a memory leak in the application or when a large number of users use it, 512mheap cannot meet the needs, but weblogic will not expand the heap size in 512m-1024m. In this case, weblogic will crash due to memory exhaustion. What's the use of setting-Xmx1024m? It has always been strange that when searching on the Internet, it is always said that-Xms and-Xmx should be set to the same size. I wonder if it is because-Xmx can't be used at all.

1.-Xmx sets the upper memory value of the JVM heap. Of course, your-Xms512m-Xmx1024m is only 512m at the beginning. If your installation and patches are normal-Xmx will definitely work.

As for the practice of setting the minimum and * * values the same, it depends on the JVM of the specific vendor, because the GC algorithm for each different JVM and some other JVM structures are different.

◆ question 2: after opening verbosegc, AF,memoryallocatefailure will appear at the beginning. I don't know how to explain it. What is even more evil is that need216k will also fail to allocate, but there are many free, why?

Answer: it's very normal to have AF, and sometimes it's even possible to just how many bytes of need. The key is to analyze the frequency of AF and the ratio before and after free, as well as the number of AF operations to action, and so on. You can check the documentation of the corresponding JVM manufacturer.

Question 3 of ◆: when all the threads of weblogic are busy, new requests enter the queue and in many cases can no longer be used. How can you know the status of weblogic applications and let them bootstrap?

Answer: the setting of thread pool size depends on many factors, such as your machine hardware configuration, user usage concurrency, and so on, as you can see in performancemonitor.

In addition, if you have a large enough thread pool but still not enough, you need to consider whether the application's connection release is reasonable.

◆ question 4: there are two admin threads in weblogic, and when memory is exhausted, console will not get a response? Why not save a little for admin? Software development network

Answer: I don't know exactly what "memory exhaustion" means; if console fails to respond, it may be that weblogic is not working properly.

◆ question 5: when using cluster, an exception occurs on a managed node, such as thread or memory exhaustion. When a new request comes in, how can it be loaded to this node in the exception? Isn't it possible to cross between overload and HA? BMW has X6, BEA doesn't know when to release weblogicX1, but I think websphere will have the same problem.

Answer: you have two concepts of cluster and HA here. I don't know exactly what you are doing. However, according to our previous usage, normally configured weblogiccluster does not send requests to nodes in a failed state.

◆ question 6: WEBLOGIC installation

Installation, you have to look at the documentation while installing, there are a large number of system parameters check, including patches, etc., in the weblogic installation package, there is no automatic check on all kinds of platforms, engineers install, most can be installed smoothly, but the problem is to run for a period of time, found problems constantly, after careful examination, found that the system did not hit a patch. Is it so troublesome to install 51 job's weblogic for manufacturers, customers and application companies on a platform? If you find that there is something wrong with the installation environment, you need to provide a parameter table for the supplier to check, at least you have a good idea. If something goes wrong, you don't have to go to the online GOOGLE.

Answer: this problem is easy to solve. BEA has a supportedconfiguration document, which you can refer to according to your actual environment. This is the standard installation method. The installation of the specification is very beneficial to deal with the problems after the application system is put online.

Is it helpful for you to read the above content? If you want to know more about the relevant knowledge or read more related articles, please follow the industry information channel, thank you for your support.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.