1. 報(bào)錯(cuò)詳情

啟動時(shí)報(bào)一下錯(cuò)誤：

20210929:14:24:06:008555 gpstart:tcloud:gpadmin-[INFO]: -Starting Master instance tcloud directory /home/greenplum/gpdata/master/gpseg-1 20210929:14:24:08:008555 gpstart:tcloud:gpadmin-[CRITICAL]:-Error occurred: non-zero rc: 1Command was: 'env GPSESSID=0000000000 GPERA=10738c01b86c1f89_210929142350 $GPHOME/bin/pg_ctl -D /home/greenplum/gpdata/master/gpseg-1 -l /home/greenplum/gpdata/master/gpseg-1/pg_log/startup.log -w -t 600 -o " -p 2345 -E " start' rc=1, stdout='waiting for server to start.... stopped waiting ', stderr='pg_ctl: could not start server Examine the log output. '

查詢報(bào)錯(cuò)日志：

# 讀取報(bào)錯(cuò)日志 [gpadmin@tcloud pg_log]$ cat /home/greenplum/gpdata/master/gpseg-1/pg_log/startup.log# 以下是報(bào)錯(cuò)的詳細(xì)信息 2021-09-29 14:23:50.116208 CST,,,p8565,th276797568,,,,0,,,seg-1,,,,,"LOG","00000","registering background worker ""sweeper process""",,,,,,,,"RegisterBackgroundWorker","bgworker.c",774, 2021-09-29 14:23:50.249829 CST,,,p8565,th276797568,,,,0,,,seg-1,,,,,"LOG","00000","redirecting log output to logging collector process",,"Future log output will appear in directory ""pg_log"".",,,,,,"SysLogger_Start","syslogger.c",989, 2021-09-29 14:24:06.500708 CST,,,p8695,th814495872,,,,0,,,seg-1,,,,,"LOG","00000","registering background worker ""ftsprobe process""",,,,,,,,"RegisterBackgroundWorker","bgworker.c",774, 2021-09-29 14:24:06.500992 CST,,,p8695,th814495872,,,,0,,,seg-1,,,,,"LOG","00000","registering background worker ""dtx recovery process""",,,,,,,,"RegisterBackgroundWorker","bgworker.c",774, 2021-09-29 14:24:06.501225 CST,,,p8695,th814495872,,,,0,,,seg-1,,,,,"LOG","00000","registering background worker ""sweeper process""",,,,,,,,"RegisterBackgroundWorker","bgworker.c",774, 2021-09-29 14:24:06.524429 CST,,,p8695,th814495872,,,,0,,,seg-1,,,,,"FATAL","XX000","could not map anonymous shared memory: Cannot allocate memory",,"This error usually means that PostgreSQL's request for a shared memory segment exceeded available memory, swap space, or huge pages. To reduce the request size (currently 234802044 bytes), reduce PostgreSQL's shared memory usage, perhaps by reducing shared_buffers or max_connections.",,,,,,"CreateAnonymousSegment","pg_shmem.c",587,1 0xbf0dac postgres errstart (elog.c:557) 2 0x9fd25e postgres PGSharedMemoryCreate (pg_shmem.c:578) 3 0xa5fe2c postgres CreateSharedMemoryAndSemaphores (ipci.c:214) 4 0xa10d6b postgres PostmasterMain (postmaster.c:1337) 5 0x6b5f21 postgres main (main.c:205) 6 0x7f022d49d555 libc.so.6 __libc_start_main + 0xf5 7 0x6c1c7c postgres <symbol not found> + 0x6c1c7c# 重點(diǎn) "FATAL","XX000","could not map anonymous shared memory: Cannot allocate memory"

2. 問題處理

對共享內(nèi)存段的請求超出了可用內(nèi)存【也就是內(nèi)存不足】給出的解決辦法是：減少請求大小（當(dāng)前為234802044字節(jié)），也可以通過減少 shared_buffers 或 max_connections 來減少 PostgreSQL 的共享內(nèi)存使用。

方法一【釋放內(nèi)存】

# 首先讀取 /proc/sys/vm/drop_caches 的值【默認(rèn)為0】 cat /proc/sys/vm/drop_caches 0# 手動執(zhí)行sync命令 # 描述：sync 命令運(yùn)行 sync 子例程。如果必須停止系統(tǒng)，則運(yùn)行sync 命令以確保文件系統(tǒng)的完整性。 # sync 命令將所有未寫的系統(tǒng)緩沖區(qū)寫到磁盤中，包含已修改的 i-node、已延遲的塊 I/O 和讀寫映射文件。 sync# 將 /proc/sys/vm/drop_caches 值設(shè)為 3 echo 3 > /proc/sys/vm/drop_caches cat /proc/sys/vm/drop_caches 3

方法二【這個(gè)要慎重 內(nèi)存小的就殺掉不必要的程序】

vim /etc/sysctl.conf# 添加vm.overcommit_memory = 2 # 使修改生效 sysctl -p

• 0 — 默認(rèn)設(shè)置。內(nèi)核執(zhí)行啟發(fā)式內(nèi)存過量使用處理，方法是估算可用內(nèi)存量，并拒絕明顯無效的請求。遺憾的是因?yàn)閮?nèi)存是使用啟發(fā)式而非準(zhǔn)確算法計(jì)算進(jìn)行部署，這個(gè)設(shè)置有時(shí)可能會造成系統(tǒng)中的可用內(nèi)存超載。
• 1 — 內(nèi)核執(zhí)行無內(nèi)存過量使用處理。使用這個(gè)設(shè)置會增大內(nèi)存超載的可能性，但也可以增強(qiáng)大量使用內(nèi)存任務(wù)的性能。
• 2 — 內(nèi)存拒絕等于或者大于總可用 swap 大小以及 overcommit_ratio 指定的物理 RAM 比例的內(nèi)存請求。如果您希望減小內(nèi)存過度使用的風(fēng)險(xiǎn)，這個(gè)設(shè)置就是最好的。

總結(jié)

以上是生活随笔為你收集整理的【Linux部署】Greenplum数据库6.13.0单机版 [CRITICAL]:-Error occurred: non-zero rc: 1（报错详情+问题处理：内存释放）的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。