PostgreSQL問題分析1：時間線不一致

一、問題：requested timeline %u does not contain minimum recovery point %X/%X on timeline %u

該日志在代碼中的位置如下：

StartupXLOG:
    if (!XLogRecPtrIsInvalid(ControlFile->minRecoveryPoint) &&
        tliOfPointInHistory(ControlFile->minRecoveryPoint - 1, expectedTLEs) !=
        ControlFile->minRecoveryPointTLI)
        ereport(FATAL,
                (errmsg("requested timeline %u does not contain minimum recovery point %X/%X on timeline %u",
                        recoveryTargetTLI,
                        (uint32) (ControlFile->minRecoveryPoint >> 32),
                        (uint32) ControlFile->minRecoveryPoint,
                        ControlFile->minRecoveryPointTLI)));

二、分析

recoveryTargetTLI這個值是什么呢？

1、改值在啟動恢復過程中變化如下：

StartupXLOG:
    ...
    if (ControlFile->minRecoveryPointTLI > ControlFile->checkPointCopy.ThisTimeLineID)
        recoveryTargetTLI = ControlFile->minRecoveryPointTLI;
    else
        recoveryTargetTLI = ControlFile->checkPointCopy.ThisTimeLineID;
    readRecoveryCommandFile();
    |-- for (item = head; item; item = item->next){
    |       ...
    |       else if (strcmp(item->name, "recovery_target_timeline") == 0){
    |           rtliGiven = true;
    |           if (strcmp(item->value, "latest") == 0)
    |               rtli = 0;
    |           else
    |               rtli = (TimeLineID) strtoul(item->value, NULL, 0);
    |       }
    |       ...
    |   }
    |   ...
    |   if (rtliGiven){
    |       if (rtli){
    |           recoveryTargetTLI = rtli;
    |           recoveryTargetIsLatest = false;
    |       }else{
    |           recoveryTargetTLI = findNewestTimeLine(recoveryTargetTLI);
    |           recoveryTargetIsLatest = true;
    |       }
    |-- }

首先，pg_control文件記錄的最小恢復點時間線ControlFile->minRecoveryPointTLI會和pg_control文件記錄的checkpoint的的時間線?ControlFile->checkPointCopy.ThisTimeLineID比較，recoveryTargetTLI取兩者之間較大值。

然后readRecoveryCommandFile會讀取recovery.cnf文件，如果是主機則沒有recovery.cnf文件。recovery.cnf文件中recovery_target_timeline的值指定恢復到的時間線，如果是latest則時間線rtli為0否則為其指定值；如果rtli是指定值，則recoveryTargetTLI為指定值，否則從第一步獲取的recoveryTargetTLI值+1開始，看是否有其對于的history文件，找到最大時間線對于的history文件，recoveryTargetTLI為該時間線。