From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752714AbaEaRyl (ORCPT ); Sat, 31 May 2014 13:54:41 -0400 Received: from mail-we0-f176.google.com ([74.125.82.176]:52518 "EHLO mail-we0-f176.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752202AbaEaRyj (ORCPT ); Sat, 31 May 2014 13:54:39 -0400 Message-ID: <538A175C.80705@colorfullife.com> Date: Sat, 31 May 2014 19:54:36 +0200 From: Manfred Spraul User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:24.0) Gecko/20100101 Thunderbird/24.5.0 MIME-Version: 1.0 To: Linux API , Linux Kernel Mailing List , Michael Kerrisk , Davidlohr Bueso Subject: MSGMNB and MSGMAX: good default values? Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi all, until now, every sysadmin/distro had to update the sysv limits. For shm, the new proposal is to increase the limits to (nearly) ULONG_MAX. Right now, I try to create patches that also increase the limits for sysv msg, but I got stuck: - MSGMNI is trivial, just increase it to nearly IPCMNI. - MSGMNB is not an upper limit, it is actually the default amount of data that can be stored in a message queue before senders must sleep. Since Linux-0.99.10, it always was 16384. A websearch shows that for both db2 and oracle, the recommendation is 65536. It seems that redhat has increased the value to 65536: https://bugzilla.redhat.com/show_bug.cgi?id=862597 Tuxedo seems to use sysv messages and the documentation warns that too large values are harmfull: http://read.pudn.com/downloads142/doc/618733/tux-a11-80/manual/13-ADM-PERF-Notes.pdf - MSGMAX: Same issue as above: it was slightly below 4096 and since 2.6 it is 8192. Any proposals? Right now, I would leave MSGMAX and MSGMNB as they are. -- Manfred