0 前言

1、2部分是对XLOG生成和清理逻辑的分析，XLOG暴涨的处理直接看第3部分。

1 WAL归档

# 在自动的WAL检查点之间的日志文件段的最大数量checkpoint_segments = # 在自动WAL检查点之间的最长时间checkpoint_timeout = # 缓解io压力checkpoint_completion_target = # 日志文件段的保存最小数量，为了备库保留更多段wal_keep_segments = # 已完成的WAL段通过archive_command发送到归档存储archive_mode = # 强制timeout切换到新的wal段文件archive_timeout = max_wal_size = min_wal_size =

1.1 不开启归档时

文件数量受下面几个参数控制，通常不超过

(2 + checkpoint_completion_target) * checkpoint_segments + 1

或

checkpoint_segments + wal_keep_segments + 1个文件。

如果一个旧段文件不再需要了会重命名然后继续覆盖使用，如果由于短期的日志输出高峰导致了超过

3 * checkpoint_segments + 1个文件，直接删除文件。

1.2 开启归档时

文件数量：删除归档成功的段文件

抽象来看一个运行的PG生成一个无限长的WAL日志序列。每段16M，这些段文件的名字是数值命名的，反映在WAL序列中的位置。在不用WAL归档的时候，系统通常只是创建几个段文件然后循环使用，方法是把不再使用的段文件重命名为更高的段编号。

当且仅当归档命令成功时，归档命令返回零。 在得到一个零值结果之后，PostgreSQL将假设该WAL段文件已经成功归档，稍后将删除段文件。一个非零值告诉PostgreSQL该文件没有被归档，会周期性的重试直到成功。

2 PG源码分析

2.1 删除逻辑

触发删除动作

RemoveOldXlogFiles> CreateCheckPoint> CreateRestartPoint

wal_keep_segments判断（调用这个函数修改_logSegNo，然后再传入RemoveOldXlogFiles）

static voidKeepLogSeg(XLogRecPtr recptr, XLogSegNo *logSegNo){ XLogSegNo segno; XLogRecPtr keep; XLByteToSeg(recptr, segno); keep = XLogGetReplicationSlotMinimumLSN(); if (wal_keep_segments > 0) { if (segno <= wal_keep_segments) segno = 1; else segno = segno - wal_keep_segments; } if (max_replication_slots > 0 && keep != InvalidXLogRecPtr) { XLogSegNo slotSegNo; XLByteToSeg(keep, slotSegNo); if (slotSegNo <= 0) segno = 1; else if (slotSegNo < segno) segno = slotSegNo; } if (segno < *logSegNo) *logSegNo = segno;}

删除逻辑

static voidRemoveOldXlogFiles(XLogSegNo segno, XLogRecPtr endptr){ ... ... while ((xlde = ReadDir(xldir, XLOGDIR)) != NULL) { if (strlen(xlde->d_name) != 24 || strspn(xlde->d_name, "0123456789ABCDEF") != 24) continue; /* * We ignore the timeline part of the XLOG segment identifiers in * deciding whether a segment is still needed. This ensures that we * won't prematurely remove a segment from a parent timeline. We could * probably be a little more proactive about removing segments of * non-parent timelines, but that would be a whole lot more * complicated. * * We use the alphanumeric sorting property of the filenames to decide * which ones are earlier than the lastoff segment. */ if (strcmp(xlde->d_name + 8, lastoff + 8) <= 0) { if (XLogArchiveCheckDone(xlde->d_name)) # 归档关闭返回真 # 存在done文件返回真 # 存在.ready返回假 # recheck存在done文件返回真 # 重建.ready文件返回假 { UpdateLastRemovedPtr(xlde->d_name); # 回收 或者 直接删除，清理.done和.ready文件 RemoveXlogFile(xlde->d_name, endptr); } } } ... ...}

2.2 归档逻辑

static voidpgarch_ArchiverCopyLoop(void){ char xlog[MAX_XFN_CHARS + 1]; # 拿到最老那个没有被归档的xlog文件名 while (pgarch_readyXlog(xlog)) { int failures = 0; for (;;) { /* * Do not initiate any more archive commands after receiving * SIGTERM, nor after the postmaster has died unexpectedly. The * first condition is to try to keep from having init SIGKILL the * command, and the second is to avoid conflicts with another * archiver spawned by a newer postmaster. */ if (got_SIGTERM || !PostmasterIsAlive()) return; /* * Check for config update. This is so that we'll adopt a new * setting for archive_command as soon as possible, even if there * is a backlog of files to be archived. */ if (got_SIGHUP) { got_SIGHUP = false; ProcessConfigFile(PGC_SIGHUP); } # archive_command没设的话不再执行 # 我们的command没有设置，走的是这个分支 if (!XLogArchiveCommandSet()) { /* * Change WARNING to DEBUG1, since we will left archive_command empty to * let external tools to manage archive */ ereport(DEBUG1, (errmsg("archive_mode enabled, yet archive_command is not set"))); return; } # 执行归档命令！ if (pgarch_archiveXlog(xlog)) { # 成功了，把.ready改名为.done pgarch_archiveDone(xlog); /* * Tell the collector about the WAL file that we successfully * archived */ pgstat_send_archiver(xlog, false); break; } else { /* * Tell the collector about the WAL file that we failed to * archive */ pgstat_send_archiver(xlog, true); if (++failures >= NUM_ARCHIVE_RETRIES) { ereport(WARNING, (errmsg("archiving transaction log file \"%s\" failed too many times, will try again later", xlog))); return; } pg_usleep(1000000L); } } }}

2.3 ready生成逻辑

static voidXLogWrite(XLogwrtRqst WriteRqst, bool flexible){... if (finishing_seg) { issue_xlog_fsync(openLogFile, openLogSegNo); WalSndWakeupRequest(); LogwrtResult.Flush = LogwrtResult.Write; # 归档打开 && wal_level >= archive if (XLogArchivingActive()) # 生成ready文件 XLogArchiveNotifySeg(openLogSegNo); XLogCtl->lastSegSwitchTime = (pg_time_t) time(NULL);...

2.4 总结

ready文件只要满足archive_mode=on和wal_lever>=archive，就总会生成（XLogWrite函数调用生成）

因为archive_command设置空，所以ready文件的消费完全由外部程序控制

done文件的处理由PG完成，两个地方会触发done文件处理，检查点和重启点

处理多少done文件受wal_keep_segments和replication_slot控制（KeepLogSeg函数）

3 WAL段累积的原因（长求总?）

注意：无论如何注意不要手动删除xlog文件

注意：checkpoint产生的日志回不立即生成ready文件，是在下一个xlog后一块生成的

3.1 ReplicationSlot

打开流了复制槽

-- 流复制插槽-- 如果restart_lsn和当前XLOG相差非常大的字节数, 需要排查slot的订阅者是否能正常接收XLOG, -- 或者订阅者是否正常. 长时间不将slot的数据取走, pg_xlog目录可能会撑爆select pg_xlog_location_diff(pg_current_xlog_location(),restart_lsn), * from pg_replication_slots;

删除

select pg_drop_replication_slot('xxx');

删除后PG会在下一个checkpoint清理xlog

3.2 较大的wal_keep_segments

检查参数配置，注意打开这个参数会使xlog和ready有一定延迟

3.3 回收出现问题

如果不使用PG自动回收机制，数据库依赖外部程序修改.ready文件，需要检测回收进程

（archive_mode=on archive_command=''）

3.4 检查点间隔过长

检查参数配置

以上为个人经验，希望能给大家一个参考，也希望大家多多支持。如有错误或未考虑完全的地方，望不吝赐教。