Results 1 to 2 of 2
  1. #1
    Join Date
    Oct 2012
    Posts
    2

    Unanswered: Veritas SF 5.1 RHEL 6x64 postgres writer 100%Cpu

    Hi
    I've got:
    4 disks created with vxfs mounted.
    postgresql 9.1.3 build on vxfs filesystem (RHEL6x64)
    database data is on vxfs filesyetm too
    with default postgresql.conf database is runing no problem
    but changing 1 parameter in postgresql.conf: max_connections = 200
    database is starting but cant create database "createdatabase TEST -p 5444" and it hangs, no messeges in postgres log

    UID PID PPID C STIME TTY TIME CMD
    postgres 19680 19679 0 12:09 pts/1 00:00:00 -bash
    postgres 19736 1 0 12:10 pts/1 00:00:00 /app/pgsql8/bin/postgres -D /db/postgres8
    postgres 19738 19736 72 12:10 ? 00:00:14 postgres: writer process
    postgres 19739 19736 0 12:10 ? 00:00:00 postgres: wal writer process
    postgres 19740 19736 0 12:10 ? 00:00:00 postgres: autovacuum launcher process
    postgres 19741 19736 0 12:10 ? 00:00:00 postgres: stats collector process
    postgres 19746 19736 0 12:10 ? 00:00:00 postgres: postgres postgres [local] CREATE DATABASE
    postgres 19751 19680 1 12:10 pts/1 00:00:00 ps -fu postgres




    after runing comand: pg_ctl -D /db/postgres stop -m immediate
    database is stoping but postgres writer is using 100% CPU and can't be stoped.

    ./stop_postgres
    waiting for server to shut down.... done
    server stopped
    -bash-4.1$ ps x
    PID TTY STAT TIME COMMAND
    19680 pts/1 S 0:00 -bash
    19738 ? Rs 1:33 postgres: writer process
    19823 pts/1 R+ 0:00 ps x
    -bash-4.1$ ps x
    PID TTY STAT TIME COMMAND
    19680 pts/1 S 0:00 -bash
    19738 ? Rs 1:37 postgres: writer process
    19824 pts/1 R+ 0:00 ps x

    ps -fu postgres
    UID PID PPID C STIME TTY TIME CMD
    postgres 19738 1 96 12:10 ? 00:02:16 postgres: writer process


    kill -9 not killing proces

    after database shutdown in log there is message:

    [LOG: received immediate shutdown request
    WARNING: terminating connection because of crash of another server process
    DETAIL: The postmaster has commanded this server process to roll back the current transaction and exit, because another server process exited abnormally and possibly corrupted shared memory.


    so i made a test
    i created lvm with ext4 fs
    moved postgresql binaries and database files, changed config file with new patchs and started the databsae
    same postgresql.conf -> max_connections = 200
    and database is working, i can create, drop restor databases...

    kernel settings:
    kernel.shmmax = 1922048000
    kernel.shmall = 4194304
    kernel.shmmni = 4096



    any sugestions?
    THX

  2. #2
    Join Date
    Oct 2012
    Posts
    2
    Hi I've solved problem...
    Veritas SF was the problem. After applying patch 5.1SP1PR2RP3 the problem is gone so far...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •