各位用户为了找寻关于PostgreSQL实战之启动恢复读取checkpoint记录失败的条件详解的资料费劲了很多周折。这里教程网为您整理了关于PostgreSQL实战之启动恢复读取checkpoint记录失败的条件详解的相关资料,仅供查阅,以下为您介绍关于PostgreSQL实战之启动恢复读取checkpoint记录失败的条件详解的详细内容

1、首先读取ControlFile->checkPoint指向的checkpoint

2、如果读取失败,slave直接abort退出,master再次读取ControlFile->prevCheckPoint指向的checkpoint

? 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 StartupXLOG->  |--checkPointLoc = ControlFile->checkPoint;  |--record = ReadCheckpointRecord(xlogreader, checkPointLoc, 1, true):  |-- if (record != NULL){    ...   }else if (StandbyMode){    ereport(PANIC,(errmsg("could not locate a valid checkpoint record")));   }else{    checkPointLoc = ControlFile->prevCheckPoint;    record = ReadCheckpointRecord(xlogreader, checkPointLoc, 2, true);    if (record != NULL){     InRecovery = true;//标记下面进入recovery    }else{     ereport(PANIC,(errmsg("could not locate a valid checkpoint record")));    }   }

一、那么什么条件下读取的checkpoint记录record==NULL?

1、ControlFile->checkPoint % XLOG_BLCKSZ < SizeOfXLogShortPHD 2、ReadRecord(xlogreader, ControlFile->checkPoint, LOG, true)返回NULL 3、ReadRecord读到的record!=NULL && record->xl_rmid != RM_XLOG_ID 4、ReadRecord读到的record!=NULL && info != XLOG_CHECKPOINT_SHUTDOWN && info != XLOG_CHECKPOINT_ONLINE 5、ReadRecord读到的record!=NULL && record->xl_tot_len != SizeOfXLogRecord + SizeOfXLogRecordDataHeaderShort + sizeof(CheckPoint)

二、ReadRecord函数返回NULL的条件

? 1 2 3 4 5 6 7 8 9 ReadRecord(xlogreader, ControlFile->checkPoint, LOG, true)  |--record = XLogReadRecord(xlogreader, ControlFile->checkPoint, &errormsg);  |-- 2.1 record==NULL && !StandbyMode  |-- 2.2 record!=NULL && !tliInHistory(xlogreader->latestPageTLI, expectedTLEs)  /*-----  note:只要读取了一页xlog,就会赋值为该页第一个记录的时间线  XLogReaderValidatePageHeader   -->xlogreader->latestPageTLI=hdr->xlp_tli;  ------*/

三、XlogReadRecord读取checkpoint返回NULL的条件?

XLogReadRecord(xlogreader, ControlFile->checkPoint, &errormsg)     targetPagePtr = ControlFile->checkPoint - (ControlFile->checkPoint % XLOG_BLCKSZ);     targetRecOff = ControlFile->checkPoint % XLOG_BLCKSZ;     readOff = ReadPageInternal(state,targetPagePtr, Min(targetRecOff + SizeOfXLogRecord, XLOG_BLCKSZ));     pageHeaderSize = XLogPageHeaderSize((XLogPageHeader) state->readBuf);     record = (XLogRecord *) (state->readBuf + RecPtr % XLOG_BLCKSZ);     total_len = record->xl_tot_len;     -------------     1、readOff < 0     2、0< targetRecOff < pageHeaderSize     3、(((XLogPageHeader) state->readBuf)->xlp_info & XLP_FIRST_IS_CONTRECORD) && targetRecOff == pageHeaderSize        page头有跨页的record并且checkpoint定位的偏移正好在页头尾部     4、targetRecOff <= XLOG_BLCKSZ - SizeOfXLogRecord &&        !ValidXLogRecordHeader(state, ControlFile->checkPoint, state->ReadRecPtr, record,randAccess)        ---(record->xl_tot_len < SizeOfXLogRecord || record->xl_rmid > RM_MAX_ID || record->xl_prev != state->ReadRecPtr)     5、targetRecOff > XLOG_BLCKSZ - SizeOfXLogRecord && total_len < SizeOfXLogRecord     6、total_len > state->readRecordBufSize && !allocate_recordbuf(state, total_len)        一旦该记录损坏,total_len的长度非常大的话,就需要allocate_recordbuf扩展state->readbuf,可能因此分配失败abort        记录的checksum需要等待全部读取完整记录后才校验     -------------

三、ReadPageInternal返回的readOff返回小于0的条件

? 1 ReadPageInternal(state,targetPagePtr, Min(targetRecOff + SizeOfXLogRecord, XLOG_BLCKSZ))

    1、第一次read wal文件,readLen = state->read_page:读取第一页。readLen < 0

    2、readLen>0 && !XLogReaderValidatePageHeader(state, targetSegmentPtr, state->readBuf)     --

    3、读取checkpoint所在页readLen = state->read_page: readLen < 0

    4、readLen > 0 && readLen <= SizeOfXLogShortPHD

    5、!XLogReaderValidatePageHeader(state, pageptr, (char *) hdr)

四、XLogPageRead何时返回值<0 ?

? 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 /*  1、WaitForWALToBecomeAvailable open失败  2、lseek 失败 && !StandbyMode  3、read失败 && !StandbyMode  4、校验page头失败 && !StandbyMode  如果是StandbyMode,则会重新retry->WaitForWALToBecomeAvailable,切换日志源进行open  */  !WaitForWALToBecomeAvailable(targetPagePtr + reqLen,private->randAccess,1,targetRecPtr)//open  |-- return -1  readOff = targetPageOff;  if (lseek(readFile, (off_t) readOff, SEEK_SET) < 0){   !StandbyMode:: return -1  }  if (read(readFile, readBuf, XLOG_BLCKSZ) != XLOG_BLCKSZ){   !StandbyMode:: return -1  }  XLogReaderValidatePageHeader(xlogreader, targetPagePtr, readBuf)  !StandbyMode:: return -1

五、WaitForWALToBecomeAvailable何时返回false?

--XLOG_FROM_ARCHIVE | XLOG_FROM_PG_WAL     1、先XLogFileReadAnyTLI open日志:         1、遍历时间线列表里的每一个时间线,从最新的开始         2、当读取checkpoint的时候,source是XLOG_FROM_ANY         3、先找归档的日志进行open;如果open失败再找WAL日志进行open         4、如果都没有open成功,则向前找时间线,open前一个时间线segno和文件号相同的文件进行open         5、open成功后expectedTLEs被赋值为当前时间线列表的所有值     2、如果open失败,则切换日志源:XLOG_FROM_ARCHIVE | XLOG_FROM_PG_WAL -> XLOG_FROM_STREAM     3、切换日志源后,XLOG_FROM_ARCHIVE | XLOG_FROM_PG_WAL 则:        slave && promote :return false        !StandbyMode:return false     --XLOG_FROM_STREAM     1、!WalRcvStreaming()即receiver进程挂了,切换日志源     2、CheckForStandbyTrigger()切换日志源     3、XLOG_FROM_STREAM->XLOG_FROM_ARCHIVE

总结

以上就是这篇文章的全部内容了,希望本文的内容对大家的学习或者工作具有一定的参考学习价值,如果有疑问大家可以留言交流,谢谢大家对的支持。

原文链接:http://blog.51cto.com/yanzongshuai/2154969