2009/1/15 Matt Sealey <matt@genesi-usa.com>:
On Thu, Jan 15, 2009 at 10:42 AM, Rob OpenSuSE <rob.opensuse.linux@googlemail.com> wrote:
Increasing the swappiness factor, as per Andrew Morton's view point, will save you some memory. Not huge amounts, but it'll be more signicant than whether the kernel loaded device mapper or not.
1) device-mapper was an example - a VERY simple example - of something which is loaded which has NO explicity dependency set in any other loaded service, module or toolkit running on the system after an
But none of the things you suggested explain the performance problem, you see, which is caused by a system having insufficient RAM.
2) swappiness is STILL not desired on our system because I keep saying that the disk is so goddamned slow that it actually severely impacts the responsiveness of the system. It is NOT a solution. How.. many.. times.. do... I .... need.. to... point.. .this.. out?
You're wrong. It'll allow anonymous pages to be saved to disk, once that you don't use because of these idle daemons, that you don't want to configure out. That'll happen at a slow rate, as part of the page freeing, done by the VM. Once those pages are saved on disk, you have more RAM pages for useful things like running programs and caching data.
3) you are not running an Efika, nor a PowerPC, so your "I don't see it on my 256MB x86 system so I don't believe your claims"-style statements are invalid anyway. Even with 256MB and a much faster system (personally I have a Via EPIA, Pegasos 1GHz G4, MPC8641D dual core 1.5GHz, a 2.4GHz Pentium 4 and a bunch of VirtualBox installations), resource usage after boot on SUSE 11.1 compared to 10.3 is far, far higher - by something like 70MB just booting GNOME.
This says more about you. I looked and made comparisons, I report that I do not see general bloat increase from 10.3 to 11.1 and give some expamples of lowered memory consumption. My message is you need to look elsewhere, not these general statements for solutions. I do believe you have problems, but they don't affect me, nor well specified systems that have more RAM.
modules (boot.lvm or boot.dmraid) and/or tools. Cups and HPLIP only *need* to be started as explicit dependencies on *actually having a printer* (it does not affect udev hotplug of USB printers that cups is
Desktop OS, server most ppl need to print. HPLIP as I told you before are not installed on my system, why don't you ask yourself why it is on yours?
not running!). rpcbind only *NEEDS* to be started if a service using rpcbind/portmap is started too.
Agreed, but it's not causing performance problems.
These are very very simple dependency issues which are not solved at some fundamental problem in the system of designing init scripts and their order and status. I do not install systems with LVM2 yet every system has boot.device-mapper active. This is the simplest to explain scenario. *WHY* is this the case?
And Rob, I don't even want an answer from you, I want someone who actually knows what they are doing, actually handles this subsystem in SUSE, and is paying attention.
Good luck! -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org For additional commands, e-mail: opensuse-factory+help@opensuse.org