walRetention.py 14.6 KB
Newer Older
A
Alex Duan 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
###################################################################
#           Copyright (c) 2016 by TAOS Technologies, Inc.
#                     All rights reserved.
#
#  This file is proprietary and confidential to TAOS Technologies.
#  No part of this file may be reproduced, stored, transmitted,
#  disclosed or used in any form or by any means other than as
#  expressly provided by the written permission from Jianhui Tao
#
###################################################################

# -*- coding: utf-8 -*-

#
#  The option for wal_retetion_period and wal_retention_size is work well
#

import taos
A
Alex Duan 已提交
19
from taos.tmq import Consumer
A
Alex Duan 已提交
20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330

from util.log import *
from util.cases import *
from util.sql import *
from util.common import *
from util.sqlset import *


import os
import threading
import json
import time
from datetime import date
from datetime import datetime
from datetime import timedelta
from os       import path


#
# --------------    util   --------------------------
#
def pathSize(path):

    total_size = 0
    for dirpath, dirnames, filenames in os.walk(path):
        for i in filenames:
            # use join to concatenate all the components of path
            f = os.path.join(dirpath, i)
            # use getsize to generate size in bytes and add it to the total size
            total_size += os.path.getsize(f)
            # print(dirpath)

    print(" %s  %.02f MB" % (path, total_size/1024/1024))
    return total_size


# load json from file
def jsonFromFile(jsonFile):
    fp = open(jsonFile)
    return json.load(fp)


#
# ----------------- class ------------------
#

# wal file object
class WalFile:
    def __init__(self, pathFile, fileName):
        self.mtime = os.path.getmtime(pathFile)
        self.startVer = int(fileName)
        self.fsize = os.path.getsize(pathFile)
        self.endVer = -1
        self.pathFile = pathFile

    def needDelete(self, delTsLine):
        return True    

# VNode object
class VNode :
    # init
    def __init__(self, dnodeId, path, walPeriod, walSize):
        self.path = path
        self.dnodeId = dnodeId
        self.vgId = 0
        self.snapVer = 0
        self.walPeriod = walPeriod
        self.walSize   = walSize
        self.walFiles = []
        self.load(path)

    # load
    def load(self, path):
        # load wal
        walPath = os.path.join(path, "wal")
        metaFile = ""
        with os.scandir(walPath) as items:
            for item in items:
                if item.is_file():
                    fileName, fileExt = os.path.splitext(item.name)
                    pathFile = os.path.join(walPath, item)
                    if fileExt == ".log":
                        self.walFiles.append(WalFile(pathFile, fileName))
                    elif fileExt == "":
                        if fileName[:8] == "meta-ver":
                            metaFile = pathFile
        # load config
        tdLog.info(f' meta-ver file={metaFile}')
        if metaFile != "":
            jsonVer = jsonFromFile(metaFile)
            metaNode = jsonVer["meta"]
            self.snapVer = int(metaNode["snapshotVer"])

        # sort with startVer
        self.walFiles = sorted(self.walFiles, key=lambda x : x.startVer, reverse=True)
        # set endVer
        startVer = -1
        for walFile in self.walFiles:
            if startVer == -1:
                startVer = walFile.startVer
                continue
            walFile.endVer = startVer - 1
            startVer = walFile.startVer

        # print total
        tdLog.info(f" ----  dnode{self.dnodeId} snapVer={self.snapVer} {self.path}  --------")
        for walFile in self.walFiles:
            mt = datetime.fromtimestamp(walFile.mtime)
            tdLog.info(f" {walFile.pathFile} {mt} startVer={walFile.startVer} endVer={walFile.endVer}")

    # snapVer compare
    def canDelete(self, walFile):
        if walFile.endVer == -1:
            # end file
            return False

        if  self.snapVer > walFile.endVer:
            return True
        return False
    
    # get log size
    def getWalsSize(self):
        size = 0
        for walFile in self.walFiles:
            size += walFile.fsize
        
        return size
    
    # vnode
    def check_retention(self):
        #
        # check period
        #
        delta = self.walPeriod
        if self.walPeriod == 0:
            delta += 1 * 60  # delete after 1 minutes
        elif self.walPeriod < 3600: 
            delta += 3 * 60  # 5 minutes
        else:
            delta += 5 * 60 # 10 minutes

        delTsLine = datetime.now() - timedelta(seconds = delta)
        delTs = delTsLine.timestamp()
        for walFile in self.walFiles:
            mt = datetime.fromtimestamp(walFile.mtime)
            info = f" {walFile.pathFile} mt={mt} line={delTsLine}  start={walFile.startVer} snap={self.snapVer} end= {walFile.endVer}"
            tdLog.info(info) 
            if walFile.mtime < delTs and self.canDelete(walFile):
                # wait a moment then check file exist
                time.sleep(1) 
                if os.path.exists(walFile.pathFile):
                    #report error
                    tdLog.exit(f" wal file expired need delete. \n   {walFile.pathFile} \n   modify time={mt} \n   delTsLine={delTsLine}\n   start={walFile.startVer} snap={self.snapVer} end= {walFile.endVer}")
                    return False            

        #
        #  check size
        # 
        if self.walSize == 0:
            return True
        
        vnodeSize = self.getWalsSize()
        if vnodeSize < self.walSize:
            tdLog.info(f" wal size valid. {self.path} real = {vnodeSize} set = {self.walSize} ")
            return True
        
        # check valid
        tdLog.info(f" wal size over set. {self.path} real = {vnodeSize} set = {self.walSize} ")
        for walFile in self.walFiles:
            if self.canDelete(walFile):
                # wait a moment then check file exist
                time.sleep(1) 
                if os.path.exists(walFile.pathFile):
                    tdLog.exit(f"  wal file size over .\
                           \n   wal file = {walFile.pathFile}\
                           \n   snapVer  = {self.snapVer}\
                           \n   real     = {vnodeSize} bytes\
                           \n   set      = {self.walSize} bytes")
                return False
        return True


# insert by async
def thread_insert(testCase, tbname, rows):
    print(f"start thread... {tbname} - {rows} \n")
    new_conn = testCase.new_connect()
    testCase.insert_data(tbname, rows, new_conn)
    new_conn.close()
    print("end thread\n")

# case
class TDTestCase:
    def init(self, conn, logSql, replicaVar=1):
        self.ts = 1670000000000
        self.replicaVar = int(replicaVar)
        tdLog.debug("start to execute %s" % __file__)
        tdSql.init(conn.cursor())
        self.setsql = TDSetSql()
        self.conn = conn

        # init cluster path
        projDir =  __file__
        pos = projDir.find("tests")
        if pos != -1:
            self.projDir = projDir[:pos]
        else:
            self.projDir = __file__
        self.projDir += "sim/"

        # udf path
        self.udf_path = os.path.dirname(os.path.realpath(__file__)) + "/udfpy"
        self.column_dict = {
            'ts': 'timestamp',
            'col1': 'tinyint',
            'col2': 'smallint',
            'col3': 'int',
            'col4': 'bigint',
            'col5': 'tinyint unsigned',
            'col6': 'smallint unsigned',
            'col7': 'int unsigned',
            'col8': 'bigint unsigned',
            'col9': 'float',
            'col10': 'double',
            'col11': 'bool',
            'col12': 'varchar(120)',
            'col13': 'nchar(100)',
        }
        self.tag_dict = {
            't1': 'tinyint',
            't2': 'smallint',
            't3': 'int',
            't4': 'bigint',
            't5': 'tinyint unsigned',
            't6': 'smallint unsigned',
            't7': 'int unsigned',
            't8': 'bigint unsigned',
            't9': 'float',
            't10': 'double',
            't11': 'bool',
            't12': 'varchar(120)',
            't13': 'nchar(100)',         
        }

    # malloc new connect
    def new_connect(self):
        return taos.connect(host     = self.conn._host, 
                            user     = self.conn._user, 
                            password = self.conn._password, 
                            database = self.dbname,
                            port     = self.conn._port, 
                            config   = self.conn._config)

    def set_stb_sql(self,stbname,column_dict,tag_dict):
        column_sql = ''
        tag_sql = ''
        for k,v in column_dict.items():
            column_sql += f"{k} {v}, "
        for k,v in tag_dict.items():
            tag_sql += f"{k} {v}, "
        create_stb_sql = f'create stable {stbname} ({column_sql[:-2]}) tags ({tag_sql[:-2]})'
        return create_stb_sql
    
    def create_database(self, dbname, wal_period, wal_size_kb, vgroups):
        self.wal_period = wal_period
        self.wal_size = wal_size_kb * 1024
        self.vgroups = vgroups
        self.dbname = dbname
        tdSql.execute(f"create database {dbname} wal_retention_period {wal_period} wal_retention_size {wal_size_kb} vgroups {vgroups} replica 3")
        tdSql.execute(f'use {dbname}')
    
    # create stable and child tables
    def create_table(self, stbname, tbname, count):
        self.child_count = count
        self.stbname = stbname
        self.tbname  = tbname
        
        # create stable
        create_table_sql = self.set_stb_sql(stbname, self.column_dict, self.tag_dict)
        tdSql.execute(create_table_sql)

        batch_size = 1000
        # create child table
        for i in range(count):
            ti = i % 128
            tags = f'{ti},{ti},{i},{i},{ti},{ti},{i},{i},{i}.000{i},{i}.000{i},true,"var{i}","nch{i}"'
            sql  = f'create table {tbname}{i} using {stbname} tags({tags});'
            tdSql.execute(sql)            
            if i % batch_size == 0:
               tdLog.info(f" create child table {i} ...")

        tdLog.info(f" create {count} child tables ok.")


    # insert to child table d1 data
    def insert_data(self, tbname, insertTime):
        start = time.time()
        values = ""
        child_name = ""
        cnt = 0
        rows = 10000000000
        for j in range(rows):
            for i in range(self.child_count):
                tj = j % 128
                cols = f'{tj},{tj},{j},{j},{tj},{tj},{j},{j},{j}.000{j},{j}.000{j},true,"var{j}","nch{j}涛思数据codepage is utf_32_le"'
                sql = f'insert into {tbname}{i} values ({self.ts},{cols});' 
                tdSql.execute(sql)
                self.ts += 1
                #tdLog.info(f" child table={i} rows={j} insert data.")
            cost = time.time() - start
            if j % 100 == 0:
                tdSql.execute(f"flush database {self.dbname}")
331
                tdLog.info("   insert row cost time = %ds rows = %d"%(cost, j))
A
Alex Duan 已提交
332
                self.consume_topic("topic1", 5)
A
Alex Duan 已提交
333

A
Alex Duan 已提交
334
            if cost > insertTime and j > 100:
335 336
                tdLog.info(" insert finished. cost time = %ds rows = %d"%(cost, j))
                return
A
Alex Duan 已提交
337 338 339
   
    # create tmq
    def create_tmq(self):
A
Alex Duan 已提交
340
        sql = f"create topic topic1 as select ts, col1, concat(col12,t12) from {self.stbname};" 
A
Alex Duan 已提交
341
        tdSql.execute(sql)
A
Alex Duan 已提交
342
        sql = f"create topic topic2 as select * from {self.stbname};" 
A
Alex Duan 已提交
343 344 345 346 347 348 349 350 351 352 353 354 355
        tdSql.execute(sql)
        #tdLog.info(sql)

    def check_retention(self):
        # flash database
        tdSql.execute(f"flush database {self.dbname}")
        time.sleep(0.5)

        vnodes = []
        # put all vnode to list
        for dnode in os.listdir(self.projDir):
            vnodeDir = self.projDir + f"{dnode}/data/vnode/"
            print(f"vnodeDir={vnodeDir}")
A
Alex Duan 已提交
356
            if os.path.isdir(vnodeDir) == False or dnode[:5] != "dnode":
357
                continue
A
Alex Duan 已提交
358 359 360
            # enum all vnode
            for entry in os.listdir(vnodeDir):
                entryPath = path.join(vnodeDir, entry)
A
Alex Duan 已提交
361
                
A
Alex Duan 已提交
362 363
                if os.path.isdir(entryPath):
                    if path.exists(path.join(entryPath, "vnode.json")):
A
Alex Duan 已提交
364 365
                        vnode = VNode(int(dnode[5:]), entryPath, self.wal_period, self.wal_size)
                        vnodes.append(vnode)
A
Alex Duan 已提交
366 367 368 369 370
        
        # do check
        for vnode in vnodes:
            vnode.check_retention()

A
Alex Duan 已提交
371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402
    # consume topic 
    def consume_topic(self, topic_name, consume_cnt):
        print("start consume...")
        consumer = Consumer(
            {
                "group.id": "tg2",
                "td.connect.user": "root",
                "td.connect.pass": "taosdata",
                "enable.auto.commit": "true",
            }
        )
        print("start subscrite...")
        consumer.subscribe([topic_name])

        cnt = 0
        try:
            while True and cnt < consume_cnt:
                res = consumer.poll(1)
                if not res:
                    break
                err = res.error()
                if err is not None:
                    raise err
                val = res.value()
                cnt += 1
                print(f" consume {cnt} ")
                for block in val:
                    print(block.fetchall())
        finally:
            consumer.unsubscribe()
            consumer.close()

A
Alex Duan 已提交
403 404 405 406 407 408

    # test db1
    def test_db(self, dbname, checkTime ,wal_period, wal_size_kb):
        # var        
        stable = "meters"
        tbname = "d"
409
        vgroups = 6
A
Alex Duan 已提交
410
        count = 10
A
Alex Duan 已提交
411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434

        # do 
        self.create_database(dbname, wal_period, wal_size_kb, vgroups)
        self.create_table(stable, tbname, count)

        # create tmq
        self.create_tmq()

        # insert data
        self.insert_data(tbname, checkTime)

        #stopInsert = False
        #tobj = threading.Thread(target = thread_insert, args=(self, tbname, rows))
        #tobj.start()

        # check retention 
        tdLog.info(f" -------------- do check retention ---------------")
        self.check_retention()


        # stop insert and wait exit
        tdLog.info(f" {dbname} stop insert ...")
        tdLog.info(f" {dbname} test_db end.")

A
Alex Duan 已提交
435

A
Alex Duan 已提交
436 437 438 439 440 441
    # run
    def run(self):
        # period
        #self.test_db("db1", 10, 60, 0)
        # size
        #self.test_db("db2", 5, 10*24*3600, 2*1024) # 2M size
442
        
A
Alex Duan 已提交
443
        # period + size        
A
Alex Duan 已提交
444 445
        #self.test_db("db", checkTime = 5*60, wal_period = 60, wal_size_kb=10)
        self.test_db("db", checkTime = 1*60, wal_period = 0, wal_size_kb=0)
A
Alex Duan 已提交
446 447 448 449 450 451 452 453


    def stop(self):
        tdSql.close()
        tdLog.success("%s successfully executed" % __file__)

tdCases.addWindows(__file__, TDTestCase())
tdCases.addLinux(__file__, TDTestCase())