1.問題描述

設(shè)置資源組后，可以正常啟動(dòng)數(shù)據(jù)庫，但是無法連接，cat /home/gpadmin/data/master/gpseg-1/pg_log/startup.log查看啟動(dòng)日志，報(bào)錯(cuò)信息如下：

2022-04-01 08:32:49.220252 CST,,,p7386,th-917768064,,,,0,,,seg-1,,,,,"LOG","00000","gpdb dir for cgroup component ""cpu"": /sys/fs/cgroup/cpu/gpdb/",,,,,,,,"dumpCompDirs","resgroup-ops-linux.c",471, 2022-04-01 08:32:49.220553 CST,,,p7386,th-917768064,,,,0,,,seg-1,,,,,"LOG","00000","gpdb dir for cgroup component ""cpuacct"": /sys/fs/cgroup/cpuacct/gpdb/",,,,,,,,"dumpCompDirs","resgroup-ops-linux.c",471, 2022-04-01 08:32:49.220801 CST,,,p7386,th-917768064,,,,0,,,seg-1,,,,,"LOG","00000","gpdb dir for cgroup component ""memory"": /sys/fs/cgroup/memory/gpdb/",,,,,,,,"dumpCompDirs","resgroup-ops-linux.c",471, 2022-04-01 08:32:49.221044 CST,,,p7386,th-917768064,,,,0,,,seg-1,,,,,"LOG","00000","gpdb dir for cgroup component ""cpuset"": /sys/fs/cgroup/cpuset/gpdb/",,,,,,,,"dumpCompDirs","resgroup-ops-linux.c",471, 2022-04-01 08:32:49.353172 CST,,,p7386,th-917768064,,,,0,,,seg-1,,,,,"FATAL","XX000","could not create semaphores: No space left on device","Failed system call was semget(5432001, 17, 03600).","This error does *not* mean that you have run out of disk space. It occurs when either the system limit for the maximum number of semaphore sets (SEMMNI), or the system wide maximum number of semaphores (SEMMNS), would be exceeded. You need to raise the respective kernel parameter. Alternatively, reduce PostgreSQL's consumption of semaphores by reducing its max_connections parameter. The PostgreSQL documentation contains more information about configuring your system for PostgreSQL.",,,,,,"InternalIpcSemaphoreCreate","pg_sema.c",126,1 0xbf0dac postgres errstart (elog.c:557) 2 0x9fc6c8 postgres PGSemaphoreCreate (pg_sema.c:113) 3 0xa74291 postgres InitProcGlobal (proc.c:259) 4 0xa60005 postgres CreateSharedMemoryAndSemaphores (ipci.c:290) 5 0xa10d6b postgres PostmasterMain (postmaster.c:1337) 6 0x6b5f21 postgres main (main.c:205) 7 0x7f39c6098555 libc.so.6 __libc_start_main + 0xf5 8 0x6c1c7c postgres <symbol not found> + 0x6c1c7c

修改參數(shù)：

kernel.sem = 2000 4096000 2000 2048

GPCC啟動(dòng)報(bào)錯(cuò)：

20220401:09:02:12:003108 gpstart:tcloud:gpadmin-[WARNING]:-error 'ERROR: Canceling query because of high VMEM usage. current group id is 6438, group memory usage 10 MB, group shared memory quota is 9 MB, slot memory quota is 0 MB, global freechunks memory is 0 MB, global safe memory threshold is 0 MB (runaway_cleaner.c:197) ' in 'SELECT pg_catalog.set_config('search_path', '', false)'

2.原因說明

原因是：由于GPCC的角色gpmon分到了組admin_group，而admin_group只有5%的cpu和不到10%的內(nèi)存，導(dǎo)致無法啟動(dòng)。

# 1.修改默認(rèn)資源組所占資源（設(shè)置默認(rèn)組和管理員組cpu占用為5%，默認(rèn)的為10%） ALTER RESOURCE GROUP default_group SET cpu_rate_limit 5; ALTER RESOURCE GROUP admin_group SET cpu_rate_limit 5; # 2.創(chuàng)建 dsjgroup 資源組（創(chuàng)建 dsjgroup 使用cpu為90% 內(nèi)存為90%） CREATE RESOURCE GROUP dsjgroup WITH (CPU_RATE_LIMIT=90, MEMORY_LIMIT=90, MEMORY_SPILL_RATIO=0);# 3.查看角色和分組 SELECT rolname, rsgname FROM pg_roles, pg_resgroup WHERE pg_roles.rolresgroup=pg_resgroup.oid;rolname | rsgname ---------------------+---------------gpcc_operator_basic | default_groupgpcc_operator | default_groupgpcc_basic | default_groupgpmon | admin_groupgpadmin | admin_group

3.問題處理

將gpmon調(diào)整到資源較多的組。

總結(jié)

以上是生活随笔為你收集整理的Greenplum【问题记录 01】资源组配置后GPCC用户无法认证登录的全部內(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。