Python函數(shù)調(diào)用追蹤實現(xiàn)代碼
更新時間：2020年11月27日 12:02:30 作者：-零
這篇文章主要介紹了Python函數(shù)調(diào)用追蹤實現(xiàn)代碼,文中通過示例代碼介紹的非常詳細(xì)，對大家的學(xué)習(xí)或者工作具有一定的參考學(xué)習(xí)價值,需要的朋友可以參考下
對于分布式追蹤，主要有以下的幾個概念：
- 追蹤 Trace：就是由分布的微服務(wù)協(xié)作所支撐的一個事務(wù)。一個追蹤，包含為該事務(wù)提供服務(wù)的各個服務(wù)請求。
- 跨度 Span：Span是事務(wù)中的一個工作流，一個Span包含了時間戳，日志和標(biāo)簽信息。Span之間包含父子關(guān)系，或者主從(Followup)關(guān)系。
- 跨度上下文 Span Context：跨度上下文是支撐分布式追蹤的關(guān)鍵，它可以在調(diào)用的服務(wù)之間傳遞，上下文的內(nèi)容包括諸如：從一個服務(wù)傳遞到另一個服務(wù)的時間，追蹤的ID，Span的ID還有其它需要從上游服務(wù)傳遞到下游服務(wù)的信息。
我實現(xiàn)了一種簡單的調(diào)用追蹤。
import uuid
import os
import time
l = []
class Recorder(object):
def __init__(self,servername,root_span = None):
if root_span == None:
self.__span = Span(servername)
else:
self.__span = Span(servername,root_span)
#上下文管理器
def __enter__(self):
return self.__span
# 退出方法中，用來實現(xiàn)善后處理工作
def __exit__(self, exc_type, exc_val, exc_tb):
self.__span.record()
self.__span.record_save(self.__span.span)
class Span(object):
def __init__(self,servername,root_span = None):
self.servername = servername
self.span = self.newspan()
if root_span != None:
root_span.dic['child_span'] = self.span
self.span["root_span_flag"] = False
def newspan(self):
self.dic = {
"spanid": uuid.uuid4().int,
"servername": self.servername,
"location": "",
"ip": "",
"durationtime": 0,
"starttime": time.time(),
"endtime":0,
"tag": "",
"log": "",
"root_span_flag":True,
"child_span": ""
}
return self.dic
def record_save(self, span):
currenttracer = {"id": span["spanid"], "data": span}
print(currenttracer)
l.append(currenttracer)
def record(self):
self.span["servername"] = self.servername
self.span["location"] = os.getcwd() + "." + self.servername
self.span["endtime"] = time.time()
self.span["durationtime"] = self.span["endtime"] - self.span["starttime"]
def setspantag(self,tag):
self.span["tag"] = tag
def setspanlog(self,log):
self.span["log"] = log
# 連續(xù)調(diào)用
with Recorder('server1') as span:
time.sleep(1)
span.setspantag("test")
#調(diào)用server1方法
print("server1")
with Recorder('server2',span) as span1:
time.sleep(2)
# 調(diào)用server2方法
print("server2")
with Recorder('server3',span1) as span2:
time.sleep(0.5)
# 調(diào)用server3方法
print("server3")
#單獨調(diào)用記錄
with Recorder('server4') as span_test1:
time.sleep(1.5)
print("server4")
with Recorder('server5') as span_test2:
time.sleep(2)
print("server5")
運(yùn)行結(jié)果：
server1
server2
server3
{'id': 224716339449765695394515303164364012192, 'data': {'spanid': 224716339449765695394515303164364012192, 'servername': 'server3', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server3', 'ip': '', 'durationtime': 0.5004403591156006, 'starttime': 1598947338.0551107, 'endtime': 1598947338.555551, 'tag': '', 'log': '', 'root_span_flag': False, 'child_span': None}}
{'id': 254736847532758359233387151739984206570, 'data': {'spanid': 254736847532758359233387151739984206570, 'servername': 'server2', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server2', 'ip': '', 'durationtime': 2.501264810562134, 'starttime': 1598947336.0542862, 'endtime': 1598947338.555551, 'tag': '', 'log': '', 'root_span_flag': False, 'child_span': {'spanid': 224716339449765695394515303164364012192, 'servername': 'server3', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server3', 'ip': '', 'durationtime': 0.5004403591156006, 'starttime': 1598947338.0551107, 'endtime': 1598947338.555551, 'tag': '', 'log': '', 'root_span_flag': False, 'child_span': None}}}
{'id': 91028031631192607088457781914309166266, 'data': {'spanid': 91028031631192607088457781914309166266, 'servername': 'server1', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server1', 'ip': '', 'durationtime': 3.5021069049835205, 'starttime': 1598947335.0534441, 'endtime': 1598947338.555551, 'tag': 'test', 'log': '', 'root_span_flag': True, 'child_span': {'spanid': 254736847532758359233387151739984206570, 'servername': 'server2', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server2', 'ip': '', 'durationtime': 2.501264810562134, 'starttime': 1598947336.0542862, 'endtime': 1598947338.555551, 'tag': '', 'log': '', 'root_span_flag': False, 'child_span': {'spanid': 224716339449765695394515303164364012192, 'servername': 'server3', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server3', 'ip': '', 'durationtime': 0.5004403591156006, 'starttime': 1598947338.0551107, 'endtime': 1598947338.555551, 'tag': '', 'log': '', 'root_span_flag': False, 'child_span': None}}}}
server4
{'id': 103171729522922437998918618387133480096, 'data': {'spanid': 103171729522922437998918618387133480096, 'servername': 'server4', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server4', 'ip': '', 'durationtime': 1.5001769065856934, 'starttime': 1598947338.555551, 'endtime': 1598947340.055728, 'tag': '', 'log': '', 'root_span_flag': True, 'child_span': None}}
server5
{'id': 320091321623887285825256878422834254741, 'data': {'spanid': 320091321623887285825256878422834254741, 'servername': 'server5', 'location': 'D:\\python_protest\\protest\\trace\\trace_main.server5', 'ip': '', 'durationtime': 2.0005736351013184, 'starttime': 1598947340.055728, 'endtime': 1598947342.0563016, 'tag': '', 'log': '', 'root_span_flag': True, 'child_span': None}}
關(guān)于下一步，會使用redis存儲其結(jié)果，并進(jìn)行相應(yīng)的輸出分析。
修過不能追蹤同級調(diào)用的問題
import uuid
import os
import time
l = []
class Recorder(object):
def __init__(self,servername,root_span = None):
if root_span == None:
self.__span = Span(servername)
else:
self.__span = Span(servername,root_span)
#上下文管理器
def __enter__(self):
return self.__span
# 退出方法中，用來實現(xiàn)善后處理工作
def __exit__(self, exc_type, exc_val, exc_tb):
self.__span.record()
self.__span.record_save(self.__span.span)
class Span(object):
def __init__(self,servername,root_span = None):
self.servername = servername
self.span = self.newspan()
if root_span != None:
root_span.dic['child_span'].append(self.span)
self.span["root_span_flag"] = False
def newspan(self):
self.dic = {
"spanid": uuid.uuid4().int,
"servername": self.servername,
"location": "",
"ip": "",
"durationtime": 0,
"starttime": time.time(),
"endtime":0,
"tag": "",
"log": "",
"root_span_flag":True,
"child_span": []
}
return self.dic
def record_save(self, span):
currenttracer = {"id": span["spanid"], "data": span}
print(currenttracer)
l.append(currenttracer)
def record(self):
self.span["servername"] = self.servername
self.span["location"] = os.getcwd() + "." + self.servername
self.span["endtime"] = time.time()
self.span["durationtime"] = self.span["endtime"] - self.span["starttime"]
def setspantag(self,tag):
self.span["tag"] = tag
def setspanlog(self,log):
self.span["log"] = log
# 連續(xù)調(diào)用
with Recorder('server1') as span:
time.sleep(1)
span.setspantag("test")
#調(diào)用server1方法
print("server1")
with Recorder('server2',span) as span1:
time.sleep(2)
# 調(diào)用server2方法
print("server2")
with Recorder('server3',span1) as span2:
time.sleep(0.5)
# 調(diào)用server3方法
print("server3")
# 并級連續(xù)調(diào)用
# with Recorder('server1') as span1:
# time.sleep(1)
# span1.setspantag("test")
# #調(diào)用server1方法
# print("server1")
# with Recorder('server2',span1) as span2:
# time.sleep(2)
# # 調(diào)用server2方法
# print("server2")
# with Recorder('server3',span1) as span3:
# time.sleep(0.5)
# # 調(diào)用server3方法
# print("server3")
# #單獨調(diào)用記錄
# with Recorder('server4') as span_test1:
# time.sleep(1.5)
# print("server4")
#
# with Recorder('server5') as span_test2:
# time.sleep(2)
# print("server5")
以上就是本文的全部內(nèi)容，希望對大家的學(xué)習(xí)有所幫助，也希望大家多多支持腳本之家。
您可能感興趣的文章:
- python通過函數(shù)名調(diào)用函數(shù)的幾種場景
- Python根據(jù)字符串調(diào)用函數(shù)過程解析
- Python基于callable函數(shù)檢測對象是否可被調(diào)用
- Python函數(shù)遞歸調(diào)用實現(xiàn)原理實例解析
- Python基于字典實現(xiàn)switch case函數(shù)調(diào)用
- python怎么調(diào)用自己的函數(shù)
- Python如何在main中調(diào)用函數(shù)內(nèi)的函數(shù)方式
- 解決python調(diào)用自己文件函數(shù)/執(zhí)行函數(shù)找不到包問題
- python通過函數(shù)名調(diào)用函數(shù)的幾種方法總結(jié)
相關(guān)文章
Python實現(xiàn)連接MySql數(shù)據(jù)庫及增刪改查操作詳解
這篇文章主要介紹了Python實現(xiàn)連接MySql數(shù)據(jù)庫及增刪改查操作,結(jié)合實例形式詳細(xì)分析了Python安裝PyMySQL模塊及使用PyMySQL模塊進(jìn)行mysql數(shù)據(jù)庫的連接、增刪改查等相關(guān)操作技巧,需要的朋友可以參考下2019-04-04
本文詳細(xì)介紹了使用Python讀取Excel文件的多種方法,介紹了pandas、openpyxl、xlrd、xlwt、pyexcel、Tablib、LibreOffice、DuckDB和Calamine等庫的用法和特點,每種方法適用于不同的場景,文中通過代碼介紹的非常詳細(xì),需要的朋友可以參考下2024-10-10
打包FlaskAdmin程序時關(guān)于static路徑問題的解決
近期寫了個基于Flask-admin的數(shù)據(jù)庫管理程序，通過pyinstaller打包，給別人用，經(jīng)過幾次嘗試，打包的數(shù)據(jù)一直找不到static里面的樣式文件，查閱資料后，最總把問題搞定了。寫下處理流程，供后來人參考2021-09-09
這篇文章主要介紹了python3.x?zip用法詳解,通過一個簡單例子給大家詳細(xì)講解zip使用，結(jié)合示例代碼給大家介紹的非常詳細(xì)，需要的朋友可以參考下2022-11-11
這篇文章主要為大家詳細(xì)介紹了如何使用Python編寫一個簡易的抽獎系統(tǒng),順便幫助大家鞏固一下對Python語法和框架的理解,感興趣的小伙伴可以了解下2023-12-12
這篇文章主要介紹了Python面向?qū)ο缶幊袒A(chǔ)解析，具有一定參考價值，需要的朋友可以了解下。2017-10-10
什么是好用的代碼呢？其實就是代碼質(zhì)量比較高，如何評價代碼質(zhì)量的高低呢？最常用的、最重要的評價標(biāo)準(zhǔn)，就是代碼的可維護(hù)性、可讀性、可擴(kuò)展性、靈活性、簡潔性、可復(fù)用性、可測試性2021-09-09
這篇文章主要介紹了fastcgi文件讀取漏洞之python掃描腳本,需要的朋友可以參考下2017-04-04
Python深度學(xué)習(xí)理解pytorch神經(jīng)網(wǎng)絡(luò)批量歸一化
這篇文章主要是Python深度學(xué)習(xí)篇，通過示例的詳解讓大家更好的理解pytorch神經(jīng)網(wǎng)絡(luò)批量歸一化，有需要的的朋友可以借鑒參考下，希望能夠有所幫助2021-10-10
最近由于公司的發(fā)展安排本人實現(xiàn)公司項目的自動化測試,下面這篇文章主要給大家介紹了關(guān)于如何Django?+?Pytest搭建在線自動化測試平臺的相關(guān)資料,文中通過實例代碼介紹的非常詳細(xì),需要的朋友可以參考下2022-07-07