How to use exceptions correctly in microservice architecture
醉折花枝作酒筹
Release： 2023-03-10 17:12:02
forward
1810 people have browsed it
This article will introduce to you the correct use of exceptions in microservice architecture. It has certain reference value. Friends in need can refer to it. I hope it will be helpful to everyone.
The correct use of exceptions ranks among the top three in importance in the microservice architecture. No comments.
The correct use of exceptions It ranks among the top three in importance in the microservice architecture. No comments.
Curdboys, long time no see. I wish you all a happy Dragon Boat Festival. I want to talk about exceptions recently. My thinking seems to have formed a closed loop. I hope this combination can be helpful to your business code.
The following will only discuss the best languages in the world and the most ecologically complete languages. I don’t have any opinions.
Similarities and Differences of Exceptions
PHP The design of exceptions in PHP7 is consistent with Java's Exception extends Throwable, but there are still some subtle differences in historical reasons and design concepts. . For example, exceptions in PHP have code attributes, so there are multiple exceptions clustered into the same exception, and then different business logic codes are written according to code in the catch block.
But Java exceptions have no code and cannot be designed like this. Different exceptions can only be used for different situations. Therefore, we are accustomed to encapsulating services through packaging classes when exposed to the outside world, instead of directly relying on the transparent transmission of exceptions.
Uniform exception handling
In Java code, the most criticized thing is the numerous try catches. I don’t have any objections. Just grab a piece of code
@Override public DataResult> getAds(Integer liveId) { try { List adsDTO = new ArrayList<>(); //...业务逻辑省略 DataResult.success(adsDTO); } catch (Exception e) { log.error("getAds has Exception:{}", e.getMessage(), e); DataResult.failure(ResultCode.CODE_INTERNAL_ERROR, e.getMessage()); // 将异常信息返回给服务端调用方 } return dataResult; }
Copy after login
Many times I just write a try catch without thinking, regardless of whether there are non-runtime exceptions in it. A better way is to use aop to intercept all service method calls, uniformly take over exceptions and handle them.
There is a small problem. If the exception information of service A is returned directly to caller B, there may be some potential risks. The caller can never be trusted, even if he is a third-generation poor peasant. . Because it is not certain how the caller will handle the error message, it may be returned directly to the front end as json.
RuntimeException
Exceptions in Java can be divided into runtime exceptions and non-runtime exceptions. Runtime exceptions do not need to be caught, nor do they need to be caught in methods. Mark throw Exception. For example, if we use the Preconditions tool class in the guava package in the method, the IllegalArgumentException thrown is also a runtime exception.
@Override public DataResult> getAds(Integer liveId) { Preconditions.checkArgument(null != liveId, "liveIds not be null"); List adsDTOS = new ArrayList<>(); //...业务逻辑省略 return DataResult.success(adsDTOS); }
Copy after login
We can also use this feature to customize our own business exception class to inherit RuntimeException
For situations that do not comply with business logic, XXServiceRuntimeException will be thrown directly
@Override public DataResult> getAds(Integer liveId) { if (null == liveId) { throw new XXServiceRuntimeException("liveId can't be null"); } List adsDTOS = new ArrayList<>(); //...业务逻辑省略 return DataResult.success(adsDTOS); }
Copy after login
and then in aop performs unified processing and makes corresponding optimizations. For the previous rough approach, exceptions other than XXServiceRuntimeException and IllegalArgumentException should be recorded internally and no longer exposed to the outside world. However, you must remember to string together distributed links through requestId. In DataResult Return in to facilitate troubleshooting.
As for the closed loop, after using the custom exception class, the threshold for monitoring and alarming the abnormal log can be reduced a lot, and the alarm To be more precise, take the monitoring of Alibaba Cloud SLS as an example
* and ERROR not XXServiceRuntimeException not IllegalArgumentException|SELECT COUNT(*) AS count
Copy after login
What is monitored here is the log recording exception log ①
Exception in PHP
The problems mentioned above in Java also exist in PHP. If you don't use 3 methods to simulate aop, you can't reflect that PHP is the best language in the world.
Similar to the above architectural logic, we will not repeat the pseudo code. , basically consistent. It is also possible to customize your own business exception class to inherit RuntimeException, and then perform external output processing.
However, there is some historical baggage in PHP. When originally designed, many runtime exceptions were output as Notice and Warning errors, but the error output lacked a call stack, which was not conducive to troubleshooting
function foo(){ return boo("xxx"); } function boo($a){ return explode($a); } foo();
Copy after login
Warning: explode() expects at least 2 parameters, 1 given in /Users/mengkang/Downloads/ab.php on line 8
Copy after login
You can't see the specific parameters, nor can you see the call stack. If you use set_error_handler ErrorException, it will be very clear.
Fatal error: Uncaught ErrorException: explode() expects at least 2 parameters, 1 given in /Users/mengkang/Downloads/ab.php:12 Stack trace: #0 [internal function]: {closure}(2, 'explode() expec...', '/Users/mengkang...', 12, Array) #1 /Users/mengkang/Downloads/ab.php(12): explode('xxx') #2 /Users/mengkang/Downloads/ab.php(8): boo('xxx') #3 /Users/mengkang/Downloads/ab.php(15): foo() #4 {main} thrown in /Users/mengkang/Downloads/ab.php on line 12
Copy after login
Modify the above function
function boo(array $a){ return implode(",", $a); }
Copy after login
and it will not be captured, because the PHP Fatal error: Uncaught TypeError is thrown, PHP7 With the new class Error implements Throwable, there will be Stack in the PHP system error log, but it cannot be connected in series with the entire business system. Here we have to talk about the design of the log. We expect to use a traceId like Java. All logs are concatenated, from Nginx logs to normal info level logs in PHP and these Uncaught TypeErrors, so the default output is taken over to the system error log and recorded to a unified place in the catch code block. Then simply modify it here to
But set_error_handler cannot handle some errors, such as E_PARSE errors. You can use register_shutdown_function to cover up.
值得注意的是register_shutdown_function的用意是在脚本正常退出或显示调用exit时，执行注册的函数。
是脚本运行(run-time not parse-time)出错退出时，才能使用。如果在调用register_shutdown_function的同一文件的里面有语法错误，是无法注册的，但是我们项目一般都是分多个文件的，这样就其他文件里有语法错误，也能捕获了
register_shutdown_function(function(){ $e = error_get_last(); if ($e){ throw new \ErrorException($e["message"], 10002, E_ERROR, $e["file"], $e["line"]); } });
The above is the detailed content of How to use exceptions correctly in microservice architecture. For more information, please follow other related articles on the PHP Chinese website!
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
