Re: nagios and freebsd threads issue : help please ...

2005-08-23 Thread Warner Losh
Thanks for forwarding the reply. First, I'll note that the response contains highly charged, inflamitory language. I've tried my best to ignore that language and focus on the actual issues. Second, this posting demonstrates a fundamental confusion between thread-safe and async-safe. That is the

Re: nagios and freebsd threads issue : help please ...

2005-08-23 Thread Daniel Eischen
On Wed, 24 Aug 2005, Christophe Yayon wrote: > Hi all, > > Here is my copy/paste from freebsd-hackers to nagios-devel list, and the > answer from a Nagios developper. Enough, please. The POSIX standard says what it says for a reason. Other OS's may appear to work, but there may in fact be race c

Re: Help with console scrolling problem (RELENG54)

2005-08-23 Thread Enache Adrian
On Tue, Aug 23, 2005 at 11:54:35PM -0400, Mike Adewole wrote: > Is there anyway of writing to the last (row, column) without causing the > screen to scroll ? Looking at function scterm_puts at (1) there doesn't seem > to be any way to avoid scrolling. Yes. Look at how ncurses is doing it. (PutChar

Re: nagios and freebsd threads issue : help please ...

2005-08-23 Thread Christophe Yayon
Hi all, Here is my copy/paste from freebsd-hackers to nagios-devel list, and the answer from a Nagios developper. Christophe Yayon wrote: >> Hi again, >> >> After some discussions on freebsd-hackers mailling list, here is a resume : >> >> 1. There a recommendation (or a suggestion) for wha

Help with console scrolling problem (RELENG54)

2005-08-23 Thread Mike Adewole
Is there anyway of writing to the last (row, column) without causing the screen to scroll ? Looking at function scterm_puts at (1) there doesn't seem to be any way to avoid scrolling. Ideally, I'll like to avoid the call to sc_vtb_delete from sc_term_gen_scroll so that the cursor position is prope

Re: IBM Active Protection System Approach

2005-08-23 Thread David Scheidt
On Tue, Aug 23, 2005 at 01:06:18AM +0200, Stefan Bethke wrote: > > Early reports from Mac enthusiast sites (and I believe similar > reports from IBM users) indicate that the hysteresis is so small that > gently pounding the table the notebook is sitting on will make the > drive park the head

ffs_balloc() ordering deadlock against vm_fault

2005-08-23 Thread Matthew Dillon
I think this applies to FreeBSD, too and I could use some help in figuring out how best to solve the problem. I have a user reporting an interesting filesystem/VM deadlock in DragonFly when running rtorrent. rtorrent has the particular effect of issuing socket read()'s direct

Re: IBM Active Protection System Approach

2005-08-23 Thread Peter Jeremy
On Tue, 2005-Aug-23 15:37:49 +0200, Borja Marcos wrote: >>mmm.. There is a static 1G force on the laptop while it is on your >>desk. When >>it falls it goes to 0G as it is in free fall. >> >>Still.. "delta G == park laptop heads" :) > >It's not a "force meter", but an acelerometer. It measure

Re: Global txpower in ath

2005-08-23 Thread Sam Pierson
On 8/10/05, Sam Leffler <[EMAIL PROTECTED]> wrote: > Sam Pierson wrote: > > I noticed that when I control the signal strength through > > ifconfig, I can effectively reduce the signal when I set it > > as something like: ifconfig ath0 txpower 1. I have read > > that this input is device driver de

Re: IBM Active Protection System Approach

2005-08-23 Thread Borja Marcos
mmm.. There is a static 1G force on the laptop while it is on your desk. When it falls it goes to 0G as it is in free fall. Still.. "delta G == park laptop heads" :) It's not a "force meter", but an acelerometer. It measures acceleration. If the computer is sitting on your desk, as it h

Re: IBM Active Protection System Approach

2005-08-23 Thread Daniel O'Connor
On Tuesday 23 August 2005 22:27, Erich Dollansky wrote: > > mmm.. There is a static 1G force on the laptop while it is on your desk. > > When it falls it goes to 0G as it is in free fall. > > Do not mix up acceleration and the force which leads to it. Ah well, I only did physics in first year and

Re: IBM Active Protection System Approach

2005-08-23 Thread Erich Dollansky
Hi, Daniel O'Connor wrote: On Tuesday 23 August 2005 21:38, Erich Dollansky wrote: The sensor will notice already the start of the move. The notebook will then normally be accalerated with 1G as it simply goes into a free fall. If the heads are then moved away from the disks and the notebook

Re: IBM Active Protection System Approach

2005-08-23 Thread Daniel O'Connor
On Tuesday 23 August 2005 21:38, Erich Dollansky wrote: > The sensor will notice already the start of the move. The notebook will > then normally be accalerated with 1G as it simply goes into a free fall. > > If the heads are then moved away from the disks and the notebook hits > some ms later grou

Re: IBM Active Protection System Approach

2005-08-23 Thread Erich Dollansky
Hi, Gary Jennejohn wrote: Borja Marcos writes: Søren Schmidt writes: It seems to me that the worst-case scenario, dropping the laptop, would result in extremely high G-loading in microsecomds rather than milliseconds. Not much can be done to save the disk in such a short time. The sensor w

Re: IBM Active Protection System Approach

2005-08-23 Thread Eric Anderson
Gary Jennejohn wrote: Borja Marcos writes: Søren Schmidt writes: I think this will need to be tailored to the exact type of "mishap" one wants to protect against. I think that the main purpose of the shock detection system is to allow data to be recovered from the disk in case the lap

Re: IBM Active Protection System Approach

2005-08-23 Thread Gary Jennejohn
Borja Marcos writes: > > Søren Schmidt writes: > > I think this will need to be tailored to the exact type of "mishap" > > one wants to protect against. > > I think that the main purpose of the shock detection system is > to allow data to be recovered from the disk in case the laptop is

[4.11] weekly panics - hints needed

2005-08-23 Thread inten.pl
Hello, we have problem with one of your 4.11 machine. About once a week it has a kernel panic. I've tried many different kernel configs and a few different -STABLE builds for different months. It always gets panic one or two times each week, but in random time. This server is used mainly as a web/

Making several custom boot configurations?

2005-08-23 Thread sd
I'm trying to make several different configurated systems on one FreeBSD box: different kernel parameters for each configuration, different hostname, startup scripts, network configurations, etc. Can it be done by adding some custom points to boot manager menu or altering existent ones? Thanks,

Re: number of simultanously opened files

2005-08-23 Thread Divacky Roman
On Mon, Aug 22, 2005 at 02:06:45PM +0200, Dag-Erling Sm?rgrav wrote: > Divacky Roman <[EMAIL PROTECTED]> writes: > > witten ~# sysctl kern.maxfiles=10 > > kern.maxfiles: 10 -> 10 > > witten ~# sysctl kern.maxfilesperproc=10 > > kern.maxfilesperproc: 10 -> 10 > > witten ~# >

Re: IBM Active Protection System Approach

2005-08-23 Thread Julian Elischer
M. Warner Losh wrote: In message: <[EMAIL PROTECTED]> Julian Elischer <[EMAIL PROTECTED]> writes: : You have about 1/4 a second to get the heads in once you detect that : you are falling. Yes. For a 4' fall, we know it takes 1/2 a second to reach the ground: s = 1/2 a t^2 +