Re: UP kernel on SMP machine?
-On [2927 07:55], Nick Sayer ([EMAIL PROTECTED]) wrote: >Does anyone think that setting up a non-SMP kernel on an SMP motherboard >(but with only one processor installed) might have anything to do with >this? UP kernels should work on a SMP box. And a few others agree with me on this. Unfortunately I don't have a 4-STABLE SMP box to test this on. Have you tried enabling DDB and trying to break to DDB when it is just blank? Did you actually see FreeBSD reporting: buffers flushed rebooting or something to that nature and the system attempting to reboot? -- Jeroen Ruigrok van der Werven Network- and systemadministrator <[EMAIL PROTECTED]>VIA Net.Works The Netherlands BSD: Technical excellence at its best http://www.via-net-works.nl Grant me the serenity to accept the things I cannot change, courage to change the things I can, and wisdom to know the difference... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: max. amount of physical memory in FBSD 4.1?
Guten tag, -On [2927 13:35], O. Hartmann ([EMAIL PROTECTED]) wrote: [snip] >And what is about FreeBSD 4.1? We use now TYAN's Thunder 2500 maonboard >which is capable to hold two CPUs and get equipted with max. 8GB main >memory. This is a nice option because we plan to solve some memory >intensive environtmental research calculations. So, my question seems >to be stupid for those who understand reading kernel code, but for me, >a kind of "normalo", please tell me: how much memory and how many CPUs >is FreeBSD 4.1.1-STABLE capable to work with in a stable fashion? I have been running FreeBSD 4-STABLE and higher on SMP boxes reliable. Those were 2 CPU boxen. For more CPU's see LINT's options: # Mandatory: options SMP # Symmetric MultiProcessor Kernel options APIC_IO # Symmetric (APIC) I/O # Optional, these are the defaults plus 1: options NCPU=5 # number of CPUs options NBUS=5 # number of busses options NAPIC=2 # number of IO APICs options NINTR=25# number of INTs 4 Gigabyte is currently the limit. Peter Wemm [[EMAIL PROTECTED]] is current working on getting 8 Gigabyte supported and working. -- Jeroen Ruigrok van der Werven Network- and systemadministrator <[EMAIL PROTECTED]>VIA Net.Works The Netherlands BSD: Technical excellence at its best http://www.via-net-works.nl Grant me the serenity to accept the things I cannot change, courage to change the things I can, and wisdom to know the difference... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Random net errors: any guru out there?
I've some random net problems using my FreeBSD 4.1 STABLE (i'm using dial-up connection). A lot of site seems to be unreachable. I get a lot of timeout errors when i try to wget, to connect to irc server, to use www, etcetc but: -ping, nslookup & traceroute give me a positive response: the site is up and running, and i'm unable to reach it -the errors are random: in the first connection www.slashdot.org is unreachable. So i disconnect and redial, now i can see slashdot but another one site that previously was ok, now is unavailable, and so again...i'm really sad...=( -all is ok under windows, so it's not an error of my isp, it's the bsd side that is wrong I hope that anyone can help me... Paolo p.s. apologize me for my bad english To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: negative bytes/sec?
-On [2927 01:40], Roman Shterenzon ([EMAIL PROTECTED]) wrote: >i586_bzero() bandwidth = -1704293721 bytes/sec >bzero() bandwidth = 132802124 bytes/sec There are apparantly some issues with calculating this. As Maxim Sobolev said to me: it is archived in the current mailinglist if you want to know the how and why. It is nothing to be worried about though AFAIUI. -- Jeroen Ruigrok van der Werven Network- and systemadministrator <[EMAIL PROTECTED]>VIA Net.Works The Netherlands BSD: Technical excellence at its best http://www.via-net-works.nl Grant me the serenity to accept the things I cannot change, courage to change the things I can, and wisdom to know the difference... To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: negative bytes/sec?
-On [2927 01:40], Roman Shterenzon ([EMAIL PROTECTED]) wrote:
>i586_bzero() bandwidth = -1704293721 bytes/sec
>bzero() bandwidth = 132802124 bytes/sec
Try this,
it worked for Maxim Sobolev, apparantly the data type is too small to
adequately store the calculation on newer systems:
--- npx.c.orig Sat May 15 01:14:13 1999
+++ npx.c Sat May 15 02:01:51 1999
@@ -696,8 +696,8 @@
if (usec <= 0)
usec = 1;
if (bootverbose)
- printf("%s bandwidth = %ld bytes/sec\n",
-funcname, (long)(BUFSIZE * (int64_t)100 / usec));
+ printf("%s bandwidth = %ld Kbytes/sec\n",
+funcname, (long)(BUFSIZE * (int64_t)100 / (1024*usec)));
free(buf, M_TEMP);
return (usec);
}
--
Jeroen Ruigrok van der Werven Network- and systemadministrator
<[EMAIL PROTECTED]>VIA Net.Works The Netherlands
BSD: Technical excellence at its best http://www.via-net-works.nl
Grant me the serenity to accept the things I cannot change, courage to
change the things I can, and wisdom to know the difference...
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message
Re: negative bytes/sec?
On Wed, 27 Sep 2000, Jeroen Ruigrok van der Werven wrote:
> -On [2927 01:40], Roman Shterenzon ([EMAIL PROTECTED]) wrote:
> >i586_bzero() bandwidth = -1704293721 bytes/sec
> >bzero() bandwidth = 132802124 bytes/sec
>
> Try this,
>
> it worked for Maxim Sobolev, apparantly the data type is too small to
> adequately store the calculation on newer systems:
>
> --- npx.c.orig Sat May 15 01:14:13 1999
> +++ npx.c Sat May 15 02:01:51 1999
> @@ -696,8 +696,8 @@
> if (usec <= 0)
>usec = 1;
> if (bootverbose)
> - printf("%s bandwidth = %ld bytes/sec\n",
> - funcname, (long)(BUFSIZE * (int64_t)100 / usec));
> + printf("%s bandwidth = %ld Kbytes/sec\n",
> + funcname, (long)(BUFSIZE * (int64_t)100 / (1024*usec)));
> free(buf, M_TEMP);
> return (usec);
> }
Yes, that's the obvious solution, but, it's for printf only, isn't it used
somewhere else?
--Roman Shterenzon, UNIX System Administrator and Consultant
[ Xpert UNIX Systems Ltd., Herzlia, Israel. Tel: +972-9-9522361 ]
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-stable" in the body of the message
Re: Userland ppp
Quoting [EMAIL PROTECTED]: > > Are all the clients running on the FreeBSD box or are you routing for > other machines? Is it possible its just FreeBSD's higher performing TCP > stack? It can transfer things much faster than windows... > > Andrew > No, it's my home computer, however the same goes to other computer which is attached to it (I run ppp -nat), when I d/l something big, the network is unreachable. --Roman Shterenzon, UNIX System Administrator and Consultant [ Xpert UNIX Systems Ltd., Herzlia, Israel. Tel: +972-9-9522361 ] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
comapq 221 smartaddry slow down the machine
hi all... i have following problem i use a SMP mchine with 2 800 MHZ piii and 1 GB ram with FreeBSD 4.1.1 the controller ist a compaq smartarray 221 if i write a big file on harddisk the complete machine freeze for some seconds. the harddisk led have very high load and the system performance (top) say he have nothing to do but the machine slow down extremly on any file operations any sugesstion idea´s ? -- To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
[stable] Re: Freebsd vs. UPS
> "Matt" == Matt Heckaman <[EMAIL PROTECTED]> writes: Matt> I have to agree with you totally. We have a Powerware 5119RM Matt> here and as far as Price:BackupTime goes, it beats APC hands Matt> down. I'm also very happy with it's quality. I have not found Matt> software to talk to it yet, but we still like it much more than Matt> APC. Matt> Note: The Poweware 5119RM was a replacement to an APC that broke Matt> after 1.5 years of usage. I've got a few of the Powerware... both RackMount and regular. I've found them to be dependable and cost effective. They come with software for Linux that I havn't been able to get running. I contacted Powerware's technical support and they agreed to release their communications protocol under NDA to me, but then they never followed up with sending me the NDA (and I got busy with other things). The 9 pin connector on the back is _not_ serial. It does provide serial ... but not on the normal pins. There are also pins in that connector that provide simple information through contacts. I havn't tried to interface to that yet, but hooking it up to a serial port's ring signal may give some basic information for cheap. I also considered firing up the linux version on a linux box (or inside VMWare) and watching the output, but again... i've been busy. Dave. -- |David Gilbert, Velocet Communications. | Two things can only be | |Mail: [EMAIL PROTECTED] | equal if and only if they | |http://www.velocet.net/~dgilbert | are precisely opposite. | =GLO To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: UP kernel on SMP machine?
Jeroen Ruigrok van der Werven wrote: > > -On [20000927 07:55], Nick Sayer ([EMAIL PROTECTED]) wrote: > >Does anyone think that setting up a non-SMP kernel on an SMP motherboard > >(but with only one processor installed) might have anything to do with > >this? > > UP kernels should work on a SMP box. And a few others agree with me on > this. > > Unfortunately I don't have a 4-STABLE SMP box to test this on. > > Have you tried enabling DDB and trying to break to DDB when it is just > blank? > > Did you actually see FreeBSD reporting: > > buffers flushed > rebooting > > or something to that nature and the system attempting to reboot? I have never experienced any problems with running UP OS's on a SMP mobo with either one or 2 CPU's installed. This is both with Pentium 200 CPUs on a Tyan Tomcat mobo... and Intel Celeron 400A on a Tyan 440BX mobo. -- ANTONY T CURTIS Tel: +44 (1635) 36222 Abacus Polar Holdings Ltd Fax: +44 (1635) 38670 > You can't judge a book by the way it wears its hair. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
nfs errors
Wed Sep 27 09:31:00 EDT 2000 Hi, I have previously sent the message below to questions and net, but received no response. If anyone here has any ideas I would really appreciate it. We have recently been moving more and more to FreeBSD. The enjoyment of such a stable platform allows us to see that we are making the right decision. We do have one error message that shows up all the time. It does not affect us or our users, but it does make me curious. I have searched everywhere, and have only found one reference to this error, but no repies. The error message is as follows: /kernel: nfs send error 32 for server blah:/partitiona The above error message occurs sometimes several times a minute, a few times an hour, there is no preset frequency, but it is often. It also affects all nfs mounted paritions. Our set up is as follows: Solaris 7 fileserver x86 FreeBSD 4.1 clients. amd.home entry: partitiona -type:=nfs;rfs:=/partitiona;opts:=rw,intr,nosuid,quota \ rhost:=blah This also occurred with 4.0 clients. Any help or insight into this message would be great. Andrew. To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: max. amount of physical memory in FBSD 4.1?
On Wed, Sep 27, 2000 at 01:34:02PM +0200, "O. Hartmann" <[EMAIL PROTECTED]> wrote: >Dear Sirs. > >Every step done under Linux is well documented by the press, especially here >in Germany! When Microsoft found a bug in AMD's Athlon CPUs which causes >Win2k to crash, the 4GB physical memory limit came into play again. Some >discussions were made about PAE addressing modell and so on and although the >discussion started on a M$ <-> AMD problem, the migration towards Linux's >memory modell was done smoothly. Well, we now know, Linux has recently a >4GB physical memory limitation due its PAE address space model. And what is >about FreeBSD 4.1? We use now TYAN's Thunder 2500 maonboard which is capable >to hold two CPUs and get equipted with max. 8GB main memory. This is a nice >option because we plan to solve some memory intensive environtmental >research calculations. So, my question seems to be stupid for those who >understand reading kernel code, but for me, a kind of "normalo", please >tell me: how much memory and how many CPUs is FreeBSD 4.1.1-STABLE capable to >work with in a stable fashion? There are some fairly complex issues that need to be resolved before FreeBSD can support more than 4GB of physical memory (splitting vm_offset/vm_size into virtual and physical parts like netbsd, cleanups, drivers that do dma..). But 4GB works just fine. As for processors, IIRC, it supports any amount that the hardware can handle, but it won't scale very well with many due to the big lock. >Many thanks, > >Gruss O. Hartmann >--- >[EMAIL PROTECTED] > >Klimadatenserver des IPA, Universitaet Mainz >Netzwerk- und Systembetreuung -- Andreas Persson [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
Re: UP kernel on SMP machine?
Nick Sayer stated: : I recently set up a machine to act as a remote bridge for a remote : office. The machine has a couple of problems. I can workaround them, but : they're annoying. It occurs to me that the cause may be that it is an : SMP machine, by nature, but it only has 1 processor installed and the : kernel does not have SMP stuff turned on. : : The two problems are: : : 1. When I reboot the machine, it never actually reboots. The screen : clears after FreeBSD says it's going to reboot, but you never see the : BIOS again. Is this a Digital PC? I see this with all of my Digital-branded PCs (all uniprocessor though). I acutally have to do a shutdown -h or -p and power cycle it. It is a royal PITA (I cannot remote boot any of them). : Does anyone think that setting up a non-SMP kernel on an SMP motherboard : (but with only one processor installed) might have anything to do with : this? It should treat it like a UP machine (in theory). S -- 101-01010101010 Sean O'Connell [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-stable" in the body of the message
