man rcd missing
cd (4) talks about /dev/rcd, but if I do man rcd, I get man: no entry for rcd in the manual. I think rcd should symlink to cd. CL
specified device does not match mounted device
I tried to mount a CD-ROM twice: [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom mount_ffs: /dev/cd0c on /mnt/cdrom: specified device does not match mounted device [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom In first time I got an error message which doesn't make sense - there was only specified device, no mounted device (when you invoke the mount with intention to mount a device, the device is not mounted yet). Man mount doesn't mention the error message. I suggest the error message to be added and explained in the mount manual page. Could you please also tell me what the error message means and why I got it once and not second time? CL
Re: Looking for HowTo instructions ...
On Sun, Oct 01, 2006 at 11:46:48PM -0300, Marc G. Fournier wrote: http://www.openbsd.org/cgi-bin/man.cgi?query=periodicapropos=1 Perfect, now, what the man page doesnt' seem to indicate is where the best place for putting the 'config variables' ... under FreeBSD, this goes in /etc/periodic.conf ... where does OpenBSD put it? same, or /etc/rc.conf.local? it does indicate it. it says that these files are shell scripts, run by cron. so you can put whatever you normally put in a shell script in the shell script, and whatever you normally put in your cron config files in your cron config files. the page also indicates in various places if things need to be set in crontab, and so on. if there are any caveats, i'd like to know them. jmc
Re: Change ISAKMP udp port?
On Friday 29 September 2006 17:01, Joachim Schipper wrote: There also are some IP-over-DNS hacks available; take a look at them, if you want even more stealth. Also, IPsec might slip by some misconfigured firewalls. isakmpd has the -p option that sets the listening port. --- Lars
Re: strange hw.cpuspeed readings
On Sun, Oct 01, 2006 at 05:59:20PM -0400, [EMAIL PROTECTED] wrote: I have done some preliminary poking around, what I have learned so far is that the celeron processor can be paired with the ich southbridges that support speedstep, it however dosen't actually support this functionality. I have a patch in the works that will ensure that we only attempt to use ich speedstep on pentium 4 cpu's, in which case you wont have this setperf method, but you should still have p4tcc clock throttling. gwk to be hones, i'm aware that my cpu doesn't do speedstep, if it did i would probably see something like (epsecially the last line): cpu0: Intel(R) Pentium(R) M processor 1.40GHz (GenuineIntel 686-class) 1.40 GHz cpu0: FPU,V86,DE,PSE,TSC,MSR,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,TM,SBF,EST,TM2 cpu0: Enhanced SpeedStep 1400 MHz (1116 mV): speeds: 1400, 1300, 1200, 1100, 1000, 900, 800, 600 MHz in my dmesg.. so that's not the problem.. what i wanted to know is wheather the hw.cpuspeed is set correctly (even before playing with setperf), and if it's not (which looks like that way) wheather it is important enough to fill in a bug report.. the second thing (setperf issues) was purly informative. what you've written about it seems to explain why it might happen, thanks. i'd be happy to test your patch. -- Przemyslaw Nowaczyk [EMAIL PROTECTED] CS student @ Poznan University of Technology http://unixlab.cs.put.poznan.pl/~inf73015/
Re: Monitor not suspending? (Xorg, DPMS, OBSD 3.9) [solved]
Just thought I'd reply to the list, as this may serve
someone else in the future.
The problem was user configuration error (or it seems).
After some more googling and reading more man pages, I
wondered about the following suggestion in xorg.conf(5):
Options
Some Option flags that may be useful to include in
Monitor sections (when needed) include DPMS, and
SyncOnGreen.
So I thought I would give it a go since I had nothing
to lose. Well, putting 'Option DPMS' in the Monitor
section of my xorg.conf file and restarting X seems to
have done the trick.
I don't understand why though, since 'xset q' was
reporting that DPMS was enabled prior to this change.
Cheers
--- patrick ~ [EMAIL PROTECTED] wrote:
Greetings,
This may just be a problem with my video card, but
I thought I'd ask since I couldn't find a definitive
answer from googling.
I noticed that none of the DPMS settings (Standby,
Suspend nor Off) would take effect at their designated
time periods. In short, X wouldn't shut-off my monitor.
Now I used to have a Linux system connected to this
monitor, and I am 100% positive with DPMS enabled, my
monitor would shut-off at the proper/expected time.
The interesting bit is that even though the monitor
isn't shutting down, X seems to pause (for a lack
of better term). I can explain it this way:
For example, if I lock the terminal using:
$ xlock -mode marquee
Marquee uses fortune to put up text on the screen (if
you didn't know this). After some time passing, with
no interaction with the system via keyboard or mouse,
the marquee would get stuck, or pause mid-sentence:
e.g.,
The opposite of a profound truth may well be ano
However, if you were to move the mouse the fortune
would continue from where it had paused (and this
could be many hours after its original pause time):
...ther profound truth.
-- Bohr
I'd like to clarify that this is not an xlock issue,
since the screen doesn't blank even if I don't run
xlock.
Since I produce this on two different DPMS aware monitors,
I can only consider either of the following: OpenBSD's
Xorg release or the video card.
Anyone else experience similar issues? If not, does
anyone know whether an ATI Rage 128 Pro video card
is DPMS capable? I couldn't find my answer using google.
Since there are so many posts of X{,org}.0.log out on
the net, searching using the DPMS as a keyword isn't
very effective.
TIA
--patrick
ps., I also cron'ed a script that would print `date'
and `xset -q | grep -A 1 -i dpms' into a file every
5 minutes. The output indicates that X is in fact
thinking (or being lead to believe) the monitor is
in fact transitioning though each of the states,
while in reality, the monitor is still on :-)
[snip...]
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com
Re: specified device does not match mounted device
Karel Kulhavy wrote: I tried to mount a CD-ROM twice: [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom mount_ffs: /dev/cd0c on /mnt/cdrom: specified device does not match mounted device [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom In first time I got an error message which doesn't make sense - there was only specified device, no mounted device (when you invoke the mount with intention to mount a device, the device is not mounted yet). Man mount doesn't mention the error message. I suggest the error message to be added and explained in the mount manual page. Could you please also tell me what the error message means and why I got it once and not second time? CL The error message is explicit mount_ffs (mount a Berkeley Fast File System) does not match the file format on the CD-ROM - if you had used the -t switch to mount(8) or used mount_cd9660(8) you would have not received that error message. HTH Fred -- OpenBSD on the Zaurus C3200 http://www.crowsons.net/puters/zaurus.php
Re: man rcd missing
On 10/2/06, Karel Kulhavy [EMAIL PROTECTED] wrote: cd (4) talks about /dev/rcd, but if I do man rcd, I get man: no entry for rcd in the manual. I think rcd should symlink to cd. CL rcd should be the raw interface to cd(4)... like there's /dev/rwd0c -Nick
Open Source is not Just Linux - Intel needs to do more to be true your campagin about Openness
Dear Mr. Ketrenos, Mr. Awad. As one of your customers, using Open Source Operating Systems, for different purposes the following two materials http://developer.osdl.org/dev/opendrivers/summit2006/james_ketrenos.pdf http://developer.osdl.org/dev/opendrivers/summit2006/james_ketrenos.mp3 did provide me with information on how Intel is supporting Open Source. Unfortunately I found that what was said there is not really true. I would like to kindly bring to your notice the details of my observations. First of all let me bring to your notice that when you said about supporting Open Source you were just referring to Linux. There are many other Open Source Operating systems other than Linux and many of Intel's customers use them. So in order to call yourself as a company supporting Open Source you need to listen to developers of these projects and to your customers who use these other Operating Systems like OpenBSD, FreeBSD, NetBSD, DragonflyBSD, OpenSolaris etc. as well To give an example I am typing this from an HP laptop that uses OpenBSD 3.9 Operating System. It has the Intel PRO/Wireless 2200 BG device. Intel has no official driver for this device in OpenBSD. The OpenBSD driver was written by Damien Bergamini. http://damien.bergamini.free.fr/ipw/ Now I would like to bring your attention to the 6th page of your PDF which has 4 points about how Intel supports Open Source that seems to me is not true. I would like to bring them to your notice if you did not know it already. Point 1 - Enable the community to do as much as possible In the mp3 you said Intel really wants to take steps as much as you can to enable the community anywhere possible But I had to download firmware for the above mentioned device manually from Bergamini's website. That is a shame :-( The firmware had an un-free distribution licence which prevented them to be included in the OpenBSD Distribution. The OpenBSD OS already has a lot of firmwares of other products shipped with it because they have a *free* licence for distribution. If Intel really wants to take steps as much as it can to enable community anywhere possible what is preventing it from making the the firmwares *freely distributable*? Please refer to http://marc.theaimsgroup.com/?l=openbsd-miscm=115972630820403w=2 if you want to get right information on how your licence restricts *free* distribution of these firmwares. Would you consider taking steps and changing the licence to be consistent with your First point? Now the second and Third Points. 2) Only keep internal those things the community cannot contribute to. 3) ...document hardware sufficiently that the community can provide their own. The link below http://marc.theaimsgroup.com/?l=openbsd-miscm=115960486810982w=2 is a public mail from the author of the BSD driver who could not get documentation for the Intel PRO/Wireless products. Will Intel consider giving him and other developers the necessary documentation to be consistent with your public statements summarized in points 2 and 3? Now the fourth point. 4) Treat the community as a member of your internal team ( Listen, and respond to, their input and feedback ) It has been publicly brought to notice that the developers who has contacted Intel has been lied to about the facts. Is this the way Intel treats a member of its internal team? Please refer to the public statements below for details. http://marc.theaimsgroup.com/?l=openbsd-miscm=115971978209040w=2 I am sure that both of you would have got similar mails from your customers regarding these issues. I do send this mail to a public list so others may add their comments and clarify/add things regarding these issues. Since your presentation promises us a lot we are eagerly looking forward towards the steps that Intel will take to be consistent with Its declaration to support open source and their customers. Thankyou so much. Kind Regards Siju
List of OpenBSD CVS commiters
Hi all, Is there a way to get a list of the OpenBSD project CVS commiters per domain (kernel, userland, ports...) ? I think it could be usefull to have an idea of who's who... Best regards, Bruno.
Re: List of OpenBSD CVS commiters
On Mon, Oct 02, 2006 at 01:12:40PM +0400, Bruno Carnazzi wrote: Hi all, Is there a way to get a list of the OpenBSD project CVS commiters per domain (kernel, userland, ports...) ? I think it could be usefull to have an idea of who's who... http://www.oxide.org/cvs/ Best regards, Bruno. -- Alexander Yurchenko
Re: List of OpenBSD CVS commiters
2006/10/2, Alexander Yurchenko [EMAIL PROTECTED]: On Mon, Oct 02, 2006 at 01:12:40PM +0400, Bruno Carnazzi wrote: Hi all, Is there a way to get a list of the OpenBSD project CVS commiters per domain (kernel, userland, ports...) ? I think it could be usefull to have an idea of who's who... http://www.oxide.org/cvs/ Whooo ! It rocks ! This url is going to my del.icio.us account :) Thank you, Bruno. Best regards, Bruno. -- Alexander Yurchenko
Re: Looking for HowTo instructions ...
2006/10/2, Marc G. Fournier [EMAIL PROTECTED]: Can someone that has installed BSDstats on your server please email me instructions on *how* to install it for your flavor of BSD? I Usually through ports(7). Best Martin
Re: Serial ATA raid
LSI is cheaper anyway, so I'll steer away from the intel. Daniel Ouellet wrote: David Gwynne wrote: On 29/09/2006, at 11:09 PM, Francois Slabbert wrote: hi misc, i'm looking to purchase a sata raid controller, and have shortlisted it down to two models for no particular reason other than the controllers being supported by openbsd, being 'afordable',compatible with the equipment i already have and available in a third world country. the two options i have is the intel srcs16 and the lsi megaraid sata-6, is there a clear winner between the two with regards to using it on openbsd - the array will be used for the archiving 'valuable' data. The intel board is basically a rebadged lsi board. make your choice based on warranty and price. Nope, I would say make your choice based on witch company actually respect your as a customers and Intel is simply not it! Didn't you see the tread on misc@ lately! Why don't you make your choice known to Intel as well now is a good time as ever. If you really care about your choice of OS here you say OpenBSD, then make your voice eared with your money! If you don't make your voice count with your wallet, why should company listen then? Sometime company understand that much better then anything else! Current reference: http://marc.theaimsgroup.com/?l=openbsd-miscm=115961387624300w=2 Daniel -- This e-mail and its contents are subject to AfriGIS PTY Limited e-mail disclaimer at http://www.afrigis.co.za/eMailDisclaimer --
Re: Looking for HowTo instructions ...
On Sun, 1 Oct 2006, Jason LaRiviere wrote: $ ls -l /var/log/*.out -rw-r--r-- 1 root wheel 1693 Oct 1 01:31 /var/log/daily.out -rw-r--r-- 1 root wheel15 Oct 1 05:30 /var/log/monthly.out -rw-r--r-- 1 root wheel59 Sep 30 03:32 /var/log/weekly.out Hello everyone, can anyone tell me why this is information is (by default) 644 and not 600 or 640 for example? At least in daily log mailq cannot be run as regular user (with postfix from ports this appears to be possible too). Weekly and monthly logs doesn't seem to contain (I didn't check very thorougly) anything that normal user can't obtain, unless admin has added something to the .local scripts. -- Antti Harri
filenames with extra characters like é,è,ö ... with rsync
Hello, I'm trying to rsync via a netware and an openbsd machine. The problem is that files with filenames like idie.doc are stored on Openbsd like this id\#202e.doc. How can I tell Openbsd to use the correct codepage when using rsync? Thank you very much! - - Didier Wiroth CEDIES route d'Esch, 211 L-1471 Luxembourg Tel: (+352) 478-8669 Fax: (+352) 478-9-8669 Web: http://www.cedies.public.lu GPG Key ID: 9A8B2ACA GPG Fingerprint: 6FF8 4362 F880 F7A8 A708 9F0D 3DD2 0502 9A8B 2ACA
RE: filenames with extra characters like é,è,ö ... with rsync
Oups ... forgot to mention that I'm using: $ uname -a OpenBSD backup 4.0 GENERIC#1 i386 + $ rsync --version rsync version 2.6.8 protocol version 29 - - Didier Wiroth -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Didier Wiroth Sent: 02 October 2006 14:18 To: misc@openbsd.org Subject: filenames with extra characters like i,h,v ... with rsync Hello, I'm trying to rsync via a netware and an openbsd machine. The problem is that files with filenames like idie.doc are stored on Openbsd like this id\#202e.doc. How can I tell Openbsd to use the correct codepage when using rsync? Thank you very much! - - Didier Wiroth CEDIES route d'Esch, 211 L-1471 Luxembourg Tel: (+352) 478-8669 Fax: (+352) 478-9-8669 Web: http://www.cedies.public.lu GPG Key ID: 9A8B2ACA GPG Fingerprint: 6FF8 4362 F880 F7A8 A708 9F0D 3DD2 0502 9A8B 2ACA
Re: overwritten file recovery - how ?
On 10/1/06, Joachim Schipper [EMAIL PROTECTED] wrote: On Sat, Sep 30, 2006 at 07:24:43PM +0200, Bambero wrote: Hello I need to recovery overwritten txt file. Ex. echo my data testfile.txt echo testfile.txt I have partition image file creted using dd. Is it possible to dump it and search using grep for example ? Is it possible to recover overwritten data ? Well, let this teach you about the values of good backups. amrecover (AMANDA) is considerably friendlier than what you're about to go through... (and I can attest to both from personal experience. Ouch.) I only backup my large repositories and media once a month. 29 days of work is worth hunting for. You're quite lucky, though, to have deleted a plain text file. Provided you still know a couple of words, you could search for them. grep -A would work, but be careful to redirect it or it'll mess up your terminal. (I dont see how grep would help here) Tools like TCT (The Coroner's Toolkit, by Wietse Venema c) or The Sleuth Kit (more modern; apparently, Autopsy is something of a GUI for it) could help a lot, if you're desparate. Joachim hexedit works just fine for this purpose imo. http://archives.neohapsis.com/archives/openbsd/2005-01/1717.html It is very safe to use, and free (as in COST, it is probobly gnu, not meeting my own concept of 'free').
RE: filenames with extra characters like é,è,ö ... with rsync
Hello, Thanks, but I already did that, and I currently use this option. ;-) This options strips/removes the special characters, isn't it possible to to store the files with the the extra characters? - - Didier Wiroth -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Christian Weisgerber Sent: 02 October 2006 15:24 To: misc@openbsd.org Subject: Re: filenames with extra characters like i,h,v ... with rsync Didier Wiroth [EMAIL PROTECTED] wrote: I'm trying to rsync via a netware and an openbsd machine. The problem is that files with filenames like idie.doc are stored on Openbsd like this id\#202e.doc. How can I tell Openbsd to use the correct codepage when using rsync? You need to tell rsync with -8. (I didn't know that. When I saw your question I skimmed through the rsync man page. You could have done that, too.) -- Christian naddy Weisgerber [EMAIL PROTECTED]
Re: Looking for HowTo instructions ...
Cross-list addresses removed. Come on, is it so difficult to post the same message (or even lightly personalized message?) three or four times so we can minimize the cross-list trash that results from people hitting group reply mindlessly? Marc G. Fournier wrote: The point of using periodic, at least under FreeBSD, is that there is a 'report' that is issued at the end of the monthly periodic run letting the admin know the status of various things on their servers ... So, for instance, it would give them a monthly reminder that the script *is* running on their machine ... sounds like it should go in the /etc/monthly.local script. Whomever gets the rest of the system reports will get that. http://www.openbsd.org/cgi-bin/man.cgi?query=monthlysektion=8 however, this does run as root, so you will probably want to make sure that is delt with properly. Still...there is no magic here, the difference between this and cron is just the wrapper scripts. Cron is Unix. This periodic thingie is apparently FreeBSD. If this is a cross-platform app, it should be Unix, not FreeBSD. Watching how bsdstats.org has been acting mostly as a random number generator for at least the last week or so has been amusing, but doesn't do much for faith in the project. It can claim to be fair, as I think it has ranked each of the major BSDs as most popular by a huge margin at least once in the last week, but I wouldn't call that meaningful. I think you need to be providing some clear, accurate, front-page explanation of how you are compiling your numbers. If you are having temporary difficulties, the responsible thing to do is to not display garbage. Otherwise, you truly are acting as nothing better than a random number generator, and one that might be interpreted as meaningful by someone in the press or management. They do things like that, you know. Nick.
Re: filenames with extra characters like é,è,ö ... with rsync
Didier Wiroth [EMAIL PROTECTED] wrote: I'm trying to rsync via a netware and an openbsd machine. The problem is that files with filenames like idie.doc are stored on Openbsd like this id\#202e.doc. How can I tell Openbsd to use the correct codepage when using rsync? You need to tell rsync with -8. (I didn't know that. When I saw your question I skimmed through the rsync man page. You could have done that, too.) -- Christian naddy Weisgerber [EMAIL PROTECTED]
Intel and Open Source Licensing w/r/t the OpenBSD Operating System
Dear Mr. Awad: As a recent purchaser of a brand-new Intel 965 motherboard with Core 2 Duo chip I am very pleased with the product, but disappointed with news from my favorite Open Source community, the OpenBSD community (http://www.openbsd.org) that Intel's licensing schemes fall short of the needs of that community. I cannot pretend to be fully conversant with the matters at dispute, but as a longtime (20 yrs +) Open Source author and occasional contributor to OpenBSD, I hope that your organization will give full and careful consideration to any concerns raised by the very experienced, talented and trustworthy OpenBSD principals. Sincerely, Jack J. Woehr -- Jack J. Woehr Director of Development Absolute Performance, Inc. [EMAIL PROTECTED] 303-443-7000 ext. 527
Intel and Licensing
Dear Mr. Awad. It has come to my attention, yet again, that intel, despite its claims of being Open Source friendly, is again failing to produce pertient API information for its products and restrictive licencing, terms and conditions. This goes against the whole priciple of open source in all its forms and unfortunately, I no longer purchase your products or recommend them to anyone else and will continue to use other suppliers until you change this policy. As you are probably aware, there are several open source products e.g. Linux, OpenBSD, FreeBSD, NetBSD and others. Further, there are different licenses ie. BSD, Apache, GPL etc. Despite the different licencing policies, all the open source projects need the same thing. The key component is that source should be open. If you can't provide source then API's have to be open (no licencing, agreements, restrictions etc.) so they can write efficient and reliable drivers for your products, which I should note, is a free service to your company. In the case of OpenBSD, one of the most efficient and secure OS's, below is an outake from their policy page, which you should take the time to read in full. http://www.openbsd.org/policy.html Because the OpenBSD copyright imposes no conditions beyond those imposed by the Berkeley copyright, OpenBSD can hope to share the same wide distribution and applicability as the Berkeley distributions. It follows however, that OpenBSD cannot include material which includes copyrights which are more restrictive than the Berkeley copyright, or must relegate this material to a secondary status, i.e. OpenBSD as a whole is freely redistributable, but some optional components may not be. A number of applications have been culled from OpenBSD because of licensing issues. A lot of people on different projects do a lot of work getting intel products to work, for very little thanks and usually no money. Do NOT make it harder for them than it already is and do NOT squander the good will of the open source community as they are IT professionals with a large networking base and you will rapidly find your products being rejected at companies and data centers, which is something neither you, your management or your shareholders will appreciate in the long run. I hope you, as a company, will take the time to learn what the open source community needs and expects, and will create a consistent and open framework that meets ALL their needs. When this happens, I will gladly reconsider the purchase and recommendation of intel products. -- Regards...Martin
Re: Change ISAKMP udp port?
On Mon, Oct 02, 2006 at 03:28:15PM +0800, Lars Hansson wrote: On Friday 29 September 2006 17:01, Joachim Schipper wrote: There also are some IP-over-DNS hacks available; take a look at them, if you want even more stealth. Also, IPsec might slip by some misconfigured firewalls. isakmpd has the -p option that sets the listening port. Getting isakmpd out is most likely possible, and not necessary anyway; it's getting ESP or AH out that is the matter... Joachim
Re: overwritten file recovery - how ?
On Mon, Oct 02, 2006 at 08:56:57AM -0400, Jeff Quast wrote: On 10/1/06, Joachim Schipper [EMAIL PROTECTED] wrote: On Sat, Sep 30, 2006 at 07:24:43PM +0200, Bambero wrote: Hello I need to recovery overwritten txt file. Ex. echo my data testfile.txt echo testfile.txt I have partition image file creted using dd. Is it possible to dump it and search using grep for example ? Is it possible to recover overwritten data ? Well, let this teach you about the values of good backups. amrecover (AMANDA) is considerably friendlier than what you're about to go through... (and I can attest to both from personal experience. Ouch.) I only backup my large repositories and media once a month. 29 days of work is worth hunting for. My backups are semi-daily, and /home is synchronised regularly between my laptop and my desktop. A single hard disk crash is enough to get one truly paranoid [1]. You're quite lucky, though, to have deleted a plain text file. Provided you still know a couple of words, you could search for them. grep -A would work, but be careful to redirect it or it'll mess up your terminal. (I dont see how grep would help here) I was going to talk about how a text file contains easily-findable data, most of the time, but considering the link you posted below... grep -bA /dev/rwd0a might be a quick way to get some offsets, without using special tools (hexedit really isn't, but most people aren't familiar with it). Tools like TCT (The Coroner's Toolkit, by Wietse Venema c) or The Sleuth Kit (more modern; apparently, Autopsy is something of a GUI for it) could help a lot, if you're desparate. Joachim hexedit works just fine for this purpose imo. http://archives.neohapsis.com/archives/openbsd/2005-01/1717.html It is very safe to use, and free (as in COST, it is probobly gnu, not meeting my own concept of 'free'). That's another option, yes. It all depends on what one likes. Joachim [1] The student association's server. No data was lost, but we were *really* lucky that /var/mail was in the part that didn't get damaged, and /usr was. The other way round would have been thoroughly unpleasant. Also, it was a good thing we were doing backups, albeit to another hard disk, and that it wasn't the second hard disk that failed. Quite a scare, although it amounted to no more than a couple of hours of downtime in the end.
RAIDFrame parity rebuild: why so slow?
Hi all, I've been using RAIDFrame on OpenBSD since 3.1 and in 4 years I've never seen any performance improvement in getting the system to work any faster at rebuilding parity after a hard shutdown. I've tried RAID1, RAID5, SCSI drives, IDE drives, processors from PentiumII 400s to Athlon64 3200+ and it has *always* been ridiculously slow at rebuilding. Just a 9G RAID5 partition takes over 2 hours. A 60G RAID1 takes 11 hours. 11!!! Before flaming me to say, just go and edit the code, it's never been out of beta or whatever, explain why compared to other OSes it's always so slow, even to build the first time around. Linux's code in particular comes to mind. Cheers, Noth
Re: PXE capable NICs: non-intel chipsets
On 2006/10/02 10:54, Jacob Yocom-Piatt wrote: i am planning on grabbing some PXE capable NICs in the next few days and would like to know if any particular cards are better than others. in the vein of intel being crappy, i am actively avoiding any intel-based chipsets. fxp(4) are generally the most decent you'll find for little money. You can buy them second-hand and avoid lining Intel's coffers (the earliest ones won't have PXE, but you can probably look up the part number). i'm looking for something cheap and reliable, the 3com 3c905c pops up near the top of the list on froogle: Those are xl(4), they're ok but istr having some problem with them, may have been connected with vlans. If you look at the list archives they're not highly recommended. I think there are some cheaper dlink that do pxe (probably rl or re). You'll probably find some linksys, allied-telesyn, smc etc. Watch out for the difference between cards with a socket that can take an EPROM to do PXE, and cards that have it supplied. Depending on your combination of skills, curiosity, and ratio of time:money available, you may be able to find a suitable module from etherboot.org (or from a motherboard using a suitable onboard nic with PXE support) to add to your bios and re-flash it to support a nic you already have. http://www.etherboot.org/wiki/biosmodule
Re: Intel Firmware license analysis
On 10/1/06, bofh [EMAIL PROTECTED] wrote: On 10/1/06, Theo de Raadt [EMAIL PROTECTED] wrote: What if some vendor decides that they like OpenBSD so much that they want to turn it (or a part of it) into a commercial product, perhaps for a specialized market segment we do not reach? So they would take the OpenBSD source tree and mutate it to their needs? We fully endorse vendors taking parts of our code and doing so, since this is much better than having vendors invent their own insecure crud and then having the world use that. And this is not just words -- there are vendors doing just that. I know - I nmaped a slingbox, and to my surprise, it returned an OpenBSD 3.7(or something) Is that a misidentification or does anyone know if they are running OpenBSD? We have some Slingboxes here at work for guys in Central and South America that still work for us. Maybe I'll check one out, I haven't had a chance to. The latest toy I've poked at is some Isilon storage, it appears to be running FreeBSD with Isilon's OneFS filesystem. Greg
Re: Lenovo laptops on OpenBSD
On 10/1/06, Aaron Glenn [EMAIL PROTECTED] wrote: On 10/1/06, J Moore [EMAIL PROTECTED] wrote: The T60 or T60p look like reasonable units for my applications - anyone got any pros or cons they can share? I dont run OpenBSD on my T60p, so I'm of no real help there. They are about to release a T61 that's core-duo enabled (I wouldn't care from a performance standpoint, merely from an economical standpoint). Check out http://forum.thinkpads.com/ for more info on the Thinkpad series and some of their quirks. I absolutely love my T60p - especially after I swapped out the Intel wireless card for an atheros one (-: rgds, aaron.glenn I have had nothing but success with IBM/Lenovo laptops and BSD, Solaris, what have you. Frequency scaling, wireless, ethernet, sound, it all just works. Bear in mind that I've never used a newer Thinkpad, but the older (P3 + P4) Thinkpads that I have used have been literally indestructible as well. The T-Series is definately the Lenovo line you want.
comment /var mount
I plan to MFS swap the /var to ramdisk as the following line in fstab: Swap /var mfs rw,-P=/proto/var,-s=65535,noexec,nosuid,nodev 0 0 This effectively mounts /var for me. Is there any gotcha if comment out line 258 in /etc/rc to: # mount /var /dev/null 21 To avoid getting /var mounted twice? Thanks!
ral0 errors
I have OBSD setup as my wireless router and it works perfectly! However when I connect with my PowerBook, I start getting tons of output on the console. The PowerBook does connect and it can use the network. 1.) Is there anything I can do about this? 2.) If not can I send these messages to /dev/null somehow? I still want to get most messages on the console, but this one happens so often that I can't do anything else if I'm on the console. cat /etc/hostname.ral0 inet 192.168.2.254 255.255.255.0 NONE media autoselect mediaopt hostap mode 11g nwid Devious_WL_BackOff chan 11 I'm also using OpenVPN for this and have several other machines connected wirelessly no problem, but whenever the PowerBook connects I get Sep 28 12:47:49 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:49 mbsd /bsd: ral0: sending data frame failed 0x0099aafa Sep 28 12:47:49 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:49 mbsd /bsd: ral0: sending data frame failed 0x0099aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x00c9aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:50 mbsd /bsd: ral0: sending data frame failed 0x0319aafa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x001003fa Sep 28 12:47:51 mbsd /bsd: ral0: sending data frame failed 0x05d1aafa Thanks B
Re: specified device does not match mounted device
This question was actually fun for learning more about mount - and about the kernel code involved. =:c) Fred Crowson wrote on Mon, Oct 02, 2006 at 10:31:24AM +0200: Karel Kulhavy wrote: I tried to mount a CD-ROM twice: [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom mount_ffs: /dev/cd0c on /mnt/cdrom: \ specified device does not match mounted device [EMAIL PROTECTED]:~$ mount /dev/cd0c /mnt/cdrom If i understand correctly, the second mount command succeeded. In first time I got an error message which doesn't make sense - there was only specified device, no mounted device (when you invoke the mount with intention to mount a device, the device is not mounted yet). Yes, *hopefully*. On the other hand, *if* you invoke mount(8) on a device which is already mounted, specifying the -u (change status) flag, you *do* get the same errno(2) in case you give the wrong mount point: [EMAIL PROTECTED] # mount | grep ftp /dev/wd0l on /ftp type ffs (local, nodev, noexec) [EMAIL PROTECTED] # mount /dev/wd0l /mnt mount_ffs: /dev/wd0l on /mnt: Device busy [EMAIL PROTECTED] # mount -u /dev/wd0l /mnt mount_ffs: /dev/wd0l on /mnt: \ specified device does not match mounted device Man mount doesn't mention the error message. When citing man pages, please do not omit the section number. The error message is not explained in mount(8), but in mount(2): [EINVAL] An argument given was invalid. OK, i admit this is not at all obvious. I suggest the error message to be added and explained in the mount manual page. Maybe the mount_ffs(8) man page could be improved; i will perhaps think about it... Could you please also tell me what the error message means and why I got it once and not second time? The error message is explicit mount_ffs (mount a Berkeley Fast File System) does not match the file format on the CD-ROM - if you had used the -t switch to mount(8) or used mount_cd9660(8) you would have not received that error message. In one respect, this is almost certainly correct - clearly, if mount_ffs(8) is invoked on a cd9660 file system, it will fail just like Karel reported. But the point is - why did mount(8) invoke mount_ffs(8) at all? Usually, mount(8) is able to autodetect cd9660 file systems. It uses readlabelfs(3) from /usr/src/lib/libutil/readlabel.c to accomplish that. Thus, why did readlabelfs(3) fail the first time, and why did it succeed the second time? To find out, i just tried the following: I opened my CD drive, put a CD into the tray, typed the command below, hit return and only *then* pressed the close tray button on the CD drive. The first time i tried, the message specified device does not match mounted device did NOT show up. So i unmounted and removed the CD and started over. Now look at the result i got from the second try: [EMAIL PROTECTED] # \ ( while true; do mount /dev/cd0c /mnt echo DONE break done; mount /dev/cd0c /mnt ) mount.out 21; \ tail -n 6 mount.out mount_ffs: /dev/cd0c on /mnt: Operation not supported by device mount_ffs: /dev/cd0c on /mnt: Operation not supported by device mount_ffs: /dev/cd0c on /mnt: Operation not supported by device mount_ffs: /dev/cd0c on /mnt: \ specified device does not match mounted device DONE mount_cd9660: /dev/cd0c on /mnt: Device busy I read this as follows: Before the CD is accessible, readlabelfs(3) will fail, so mount(8) will use its default which happens to be -t ffs. As soon as the CD becomes accessible, readlabelfs(3) will of course return the correct value cd9660. Before the CD is accessible, mount(2) called by mount_ffs(8) will fail with ENODEV, translated to Operation not supported by device by strerror(3). As soon as the CD becomes accessible, mount(2) will notice the file system type mismatch, returning EINVAL, translated to specified device does not match mounted device by mount_ffs(8). Apparently, there is a race condition. Consider the following timeline: 1. the user starts mount(8) 2. mount(8) calls readlabelfs(3) *** ASSUME THE CD IS STILL UNAVAILABLE AT THIS TIME *** 3. readlabelfs(3) returns NULL to mount(8) (note: if the CD would already be readable now, the return value would be cd9660 and the mount would succeed) 4. mount(8) does fork(2) and exec(3) to start mount_ffs(8) 5. mount_ffs(8) calls mount(2) with option MOUNT_FFS 6. switch to kernel space, sys_mount from kern/vfs_syscalls.c 7. sys_mount uses copyinstr(9) to get fstypename from user space 8. sys_mount calls ffs_mount from ufs/ffs/ffs_vfsops.c 9. ffs_mount uses copyinstr(9) to get mount point and device name 10. ffs_mount calls ffs_mountfs from ufs/ffs/ffs_vfsops.c *** ASSUME THE CD HAS NOW BECOME AVAILABLE *** 11. ffs_mountfs successfully opens the cd(4) device for reading by calling VOP_OPEN from kern/vnode_if.c = cdopen from scsi/cd.c 12. ffs_mountfs searches for an ffs superblock on the CD - of
Serial control of LCD display
I am trying to get a CrystalFontz 632 serial display to work with an OpenBSD box. Under Windows I can just connect the display to a com port, run Hyperterminal and send text directly to it, so I assumed that I could just send a data stream to /dev/tty00 under OpenBSD and make it work as well. Unfortunately it is not turning out to be anywhere that simple. If I use cu or tip and connect to /dev/tty00 and 19200 then I can send data to the display, but eventually I need to be able to send data to it from a shell script. Any attempt I make to send data to it (such as cat test /dev/tty00) results in an error of sh: Cannot create /dev/tty00: Interrupted system call. I've tried to mess with the stty command to setup the serial port (open it up, set the speed, etc), but no luck, that error always comes up. Can anyone point me to the right direction on this? Thanks, Peter
Re: comment /var mount
On 10/2/06, Ray [EMAIL PROTECTED] wrote: I plan to MFS swap the /var to ramdisk as the following line in fstab: Is there any gotcha if comment out line 258 in /etc/rc to: # mount /var /dev/null 21 Works for me. Haven't had any problems. Riley -- Education: The ability to listen to almost anything without losing your temper or self confidence. - -- Robert Frost
Re: Looking for HowTo instructions ...
On Mon, Oct 02, 2006 at 08:21:30PM +0200, Manuel Bouyer wrote: On Mon, Oct 02, 2006 at 12:02:34AM -0300, Marc G. Fournier wrote: The point of using periodic, at least under FreeBSD, is that there is a 'report' that is issued at the end of the monthly periodic run letting the admin know the status of various things on their servers ... So, for instance, it would give them a monthly reminder that the script *is* running on their machine ... The standard output and errors of cron jobs is mailed to the owner of the cron tab. I'm not sure what periodic can do more in this area. -- Manuel Bouyer, LIP6, Universite Paris VI. [EMAIL PROTECTED] NetBSD: 26 ans d'experience feront toujours la difference -- Suppose your cron jobs don't emit output, which any good job shouldn't do. -Damian
Re: avoiding INTEL
I 2ND THAT! -Original Message- Gilles Chehade Sent: Sunday, October 01, 2006 4:17 PM Mr. Majid Awad, I have recently been provided with a DELL laptop at work which runs Windows and OpenBSD. I have recently bought a VAIO laptop for personal use which runs OpenBSD. These two laptops have a point in common. ALL of the hardware is recognized and works ... except for the Intel wifi chipsets. About four days ago, at an airport, I had written code that I needed to synchronize with the current code base at work. This forced me to keep booting between Windows and my working operating system so that I could actually work and make the changes visible for the people at my company by ... using the network. You can't imagine how irritable one can get when spending time waiting for a couple systems to boot about twenty times within a few hours. This has wasted both me and my company a lot of time and money. And worst of all, this is how we are treated for trusting your company and SPENDING MONEY on its hardware. I just got back from my business trip and first thing I'll do tomorrow is discuss with the decision takers at work so that they contact dell and tell them the Intel based hardware is not working and we need it replaced (oh don't worry that's just about a hundred laptops, it won't affect your sales that much). And for my personal use, this is the last time I buy a laptop or desktop with Intel hardware in it, I have wasted enough time with your integrated graphic chipsets and your integrated wifi chipsets, I can't even recall how many times I got angry at some of your hardware not being working properly (or at all) because you refuse to cooperate with ... your customers. I'd rather spend my money on something that works and keep a smile on my face while working. I am angry and you lost a customer.
Re: comment /var mount
On 10/2/06, Riley McIntire [EMAIL PROTECTED] wrote: On 10/2/06, Ray [EMAIL PROTECTED] wrote: I plan to MFS swap the /var to ramdisk as the following line in fstab: Is there any gotcha if comment out line 258 in /etc/rc to: # mount /var /dev/null 21 Works for me. Haven't had any problems. Don't hack /etc/rc set the noauto flag on /var. that will prevent it from being mounted by mount -a but mount /var will mount it anyway. -- GDB has a 'break' feature; why doesn't it have 'fix' too?
Re: comment /var mount
Works for me. Haven't had any problems. (Riley) Don't hack /etc/rc set the noauto flag on /var. that will prevent it from being mounted by mount -a but mount /var will mount it anyway. Thanks Riley! I would rather leave rc alone - but found out that using noauto option may stop the device from giving out DHCP leases after reboot - some sort of conflict, but haven't figured it out yet - or proven that it's the noauto option. It seems healthier with the comment in /etc/rc - but testing more...
Re: Intel policy wrt OSS [was: Re: cvs.openbsd.org: src]
On Sat, Sep 30, 2006 at 03:03:57AM -0600, Theo de Raadt wrote: [snip] Majid Awad at Intel has stated to developers that he is the current person who is responsible for this particular area. So go ahead, let him know how you feel about this. Again, his email address is [EMAIL PROTECTED] So let's win back the rights to run the hardware we purchased. Please feel free to let other open source communities know about this matter. Thank you. Does anyone happen to have a snail-mail address for Majid? -Damian
Re: Intel's Open Source Policy Doesn't Make Sense
On Sun, Oct 01, 2006 at 11:14:37AM -0700, Brian wrote: [snip] What does Intel gain by not being open? I am puzzled. I am not an engineer, so is there something that I am overlooking? Cheers, Brian I can think of a few possibilities: a) Intel doesn't own the technology, but licensed it from another vendor. The licensing terms don't allow Intel to release full details. b) Intel has agreements with other customers/vendors to not release information about a particular piece of hardware. c) Intel doesn't feel that it's worth the cost to provide information for driver developers. I suspect that in most cases it's a matter of will rather than any technical or legal obstacles. -Damian
Re: OpenBSD Paypal used against User Agreement?
On Sat, Sep 30, 2006 at 10:40:40AM +0200, viq wrote: [snip] I read some not-really-nice comments about paypal, and as one of alternatives listed were moneybookers (.com) Can't say i tried either, but comments seemed positive. -- viq Google has a payment service, but it's restricted to pre-approved sellers: http://checkout.google.com/ From what I've heard, PayPal tends to simply lock accounts and ban people versus investigate allegations. For example, a former co-worker had his account banned because one of his buyers included a comment in a transaction about illicit drugs. Needless to say the buyer was joking, but PayPal refused to reinstate the seller's account. -Damian
Re: Lenovo laptops on OpenBSD
On 10/1/06, J Moore [EMAIL PROTECTED] wrote: I've got to buy a couple of laptops, and want to get something that's as open source friendly as possible. I know at one time, there were a number of OpenBSD users that were enthusiastic about ThinkPads. thinkpads are still the favorite. work is always being done to add more hardware support. when I first got my z60t I didn't have support for sound, wireless, acpi or video. Now there's support for video and sound. acpi is still in the pipe. Atheros has seen lots of work this month and I've been tracking -current and providing feedback like many of us eager atheros hopefuls.
Re: Intel Firmware license analysis
On 10/2/06, Greg Thomas [EMAIL PROTECTED] wrote: On 10/1/06, bofh [EMAIL PROTECTED] wrote: I know - I nmaped a slingbox, and to my surprise, it returned an OpenBSD 3.7(or something) Is that a misidentification or does anyone know if they are running OpenBSD? We have some Slingboxes here at work for guys in Central and No idea. I just poked at it again (this is after a firmware upgrade months ago), and now I get some other result.
Re: Looking for HowTo instructions ...
Damian Wiest [EMAIL PROTECTED] wrote: Suppose your cron jobs don't emit output, which any good job shouldn't do. Huh? If you want a task to run on a schedule, and then mail you the results, then cron is exactly what you want. Any good job does what its author wants it to. If they want it to emit output, then having it be silent for no reason does not make it a good job. Adam
Was= Re: glib2 fails to build on ppc-current: pthreads issue?
Hello, glib2 builds fine now but it gives me an error about glib2-docs. I did a fresh checkout of the ports tree and I still get the same error. The error is: make: don't know how to make /usr/ports/devel/glib2/w-glib2-2.10.3/pkg/DESCR-docs. Stop in /usr/ports/devel/glib2. A little more context is below. I attached the config.log as well. Please let me know if you need anymore information. When I set SUBPACKAGES= it still gave me an error. -- Installing ./html/index.sgml /bin/sh ../../../mkinstalldirs /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/man/man1 install -c -o root -g bin -m 444 ./glib-mkenums.1 /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/man/man1/glib-mkenums.1 install -c -o root -g bin -m 444 ./glib-genmarshal.1 /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/man/man1/glib-genmarshal.1 install -c -o root -g bin -m 444 ./gobject-query.1 /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/man/man1/gobject-query.1 gmake[5]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference/gobject' gmake[4]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference/gobject' gmake[4]: Entering directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference' gmake[5]: Entering directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference' gmake[5]: Nothing to be done for `install-exec-am'. gmake[5]: Nothing to be done for `install-data-am'. gmake[5]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference' gmake[4]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference' gmake[3]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs/reference' gmake[3]: Entering directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs' gmake[4]: Entering directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs' gmake[4]: Nothing to be done for `install-exec-am'. gmake[4]: Nothing to be done for `install-data-am'. gmake[4]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs' gmake[3]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs' gmake[2]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3/docs' gmake[1]: Leaving directory `/usr/ports/devel/glib2/w-glib2-2.10.3/glib-2.10.3' sed -e s/dependency_libs='/dependency_libs='-pthread/ /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/lib/libgthread-2.0.la /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/lib/libgthread-2.0.temp mv /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/lib/libgthread-2.0.temp /usr/ports/devel/glib2/w-glib2-2.10.3/fake-macppc/usr/local/lib/libgthread-2.0.la make: don't know how to make /usr/ports/devel/glib2/w-glib2-2.10.3/pkg/DESCR-docs. Stop in /usr/ports/devel/glib2. *** Error code 2 Stop in /usr/ports/devel/glib2 (line 1312 of /usr/ports/infrastructure/mk/bsd.port.mk). *** Error code 1 Stop in /usr/ports/devel/glib2 (line 1753 of /usr/ports/infrastructure/mk/bsd.port.mk). *** Error code 1 Stop in /usr/ports/devel/glib2 (line 2049 of /usr/ports/infrastructure/mk/bsd.port.mk). [EMAIL PROTECTED]:/usr/ports/devel/glib2 # On 9/30/06, Antoine Jacoutot [EMAIL PROTECTED] wrote: On Fri, 29 Sep 2006, Otto Moerbeek wrote: I just committed this fix; powerpc stacks are a bit weird in that they need access above to stack pointer; while the stack grows down. With the more strict mmap()ed stacks that kurt@ committed recently, this got broken. Allright. It seems better with your fix: [...] checking size of pthread_t... 4 checking for pthread_attr_setstacksize... yes checking for minimal/maximal thread priority... none found configure: WARNING: I can not find the minimal and maximal priorities for threads on your system. Thus threads can only have the default priority. If you happen to know these main/max priorities, please inform the GLib developers. checking for posix yield function... sched_yield configure: WARNING: the 'g_get_(user_name|real_name|home_dir|tmp_dir)' functions will not be MT-safe during their first call because there is no working 'getpwuid_r' on your system. checking size of pthread_mutex_t... 4 checking byte contents of PTHREAD_MUTEX_INITIALIZER... 0,0,0,0 checking whether to use assembler code for atomic operations... powerpc [...] Thanks! -- Antoine -- --dfc [EMAIL PROTECTED] [demime 1.01d removed an attachment of type application/octet-stream which had a name of config.log]
Re: Looking for HowTo instructions ...
On Mon, Oct 02, 2006 at 12:02:34AM -0300, Marc G. Fournier wrote: The point of using periodic, at least under FreeBSD, is that there is a 'report' that is issued at the end of the monthly periodic run letting the admin know the status of various things on their servers ... So, for instance, it would give them a monthly reminder that the script *is* running on their machine ... The standard output and errors of cron jobs is mailed to the owner of the cron tab. I'm not sure what periodic can do more in this area. -- Manuel Bouyer, LIP6, Universite Paris VI. [EMAIL PROTECTED] NetBSD: 26 ans d'experience feront toujours la difference --
Re: Looking for HowTo instructions ...
Marc G. Fournier [EMAIL PROTECTED] writes: Can someone that has installed BSDstats on your server please email me instructions on *how* to install it for your flavor of BSD? I've put together a very rough draft for a BSDstats getting started guide, available for digestion and criticism at http://bsdly.net/~peter/bsdstat/. And yes, the bits about OpenBSD are extremely similar to the instructions at bsdstats.org for a very good reason ;) -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://www.blug.linux.no/rfc1149/ http://www.datadok.no/ http://www.nuug.no/ First, we kill all the spammers The Usenet Bard, Twice-forwarded tales 20:11:56 delilah spamd[26905]: 146.151.48.74: disconnected after 36099 seconds
Re: Looking for HowTo instructions ...
On Mon, Oct 02, 2006 at 08:21:30PM +0200, Manuel Bouyer wrote: On Mon, Oct 02, 2006 at 12:02:34AM -0300, Marc G. Fournier wrote: The point of using periodic, at least under FreeBSD, is that there is a 'report' that is issued at the end of the monthly periodic run letting the admin know the status of various things on their servers ... So, for instance, it would give them a monthly reminder that the script *is* running on their machine ... The standard output and errors of cron jobs is mailed to the owner of the cron tab. I'm not sure what periodic can do more in this area. It just saves you from getting multiple messages. Putting a script in /etc/periodic/monthly is exactly the same as adding that script onto/into /etc/monthly.local. In fact, FreeBSD still has /etc/monthly.local, which is run by /etc/monthly/999.local. Part of the adding and removing scripts from directories is easier for the package management system than sed scripts theory, I suspect. Cheers, -- Andrew
annoying openbsd mutt package
I am using mutt with openbsd. I am getting annoyed by a message error i got just after i start it on command line: The message is the following: /var/mail/grios: No such file or directory (errno = 2) The strangest thing about it, it is that it only happens with my openbsd installed version even having the following in ~/.muttrc set mbox_type=Maildir set folder=~/.mail/ I don't know what i am supposed to do to prevent it from happening. Thanks in advance.
Re: annoying openbsd mutt package
On Mon, Oct 02, 2006 at 10:06:34PM -0300, Gustavo Rios wrote: I am using mutt with openbsd. I am getting annoyed by a message error i got just after i start it on command line: The message is the following: /var/mail/grios: No such file or directory (errno = 2) The strangest thing about it, it is that it only happens with my openbsd installed version even having the following in ~/.muttrc set mbox_type=Maildir set folder=~/.mail/ I don't know what i am supposed to do to prevent it from happening. Thanks in advance. There's a mutt mailing list, and they are very helpful. It would be more appropriate there, even if this only happens to you on OpenBSD. Have you tried set spoolfile=+whatever? Also, I prefer set folder=$HOME/.mail -- Darrin Chandler| Phoenix BSD Users Group [EMAIL PROTECTED] | http://bsd.phoenix.az.us/ http://www.stilyagin.com/ |
Re: PXE capable NICs: non-intel chipsets
Jacob Yocom-Piatt wrote: i am planning on grabbing some PXE capable NICs in the next few days and would like to know if any particular cards are better than others. in the vein of intel being crappy, i am actively avoiding any intel-based chipsets. i'm looking for something cheap and reliable, the 3com 3c905c pops up near the top of the list on froogle: I have a lot of old 3c905B and 3c905C cards with PXE boot ROMs on them. None of them work with OpenBSD's PXE booting. HOWEVER, I suspect it is just an issue of the age of the ROMs on these things (they all came out of the same batch of computers which are about six or seven years old now)... 3Com also has a boot floppy which will initiate a PXE boot from a machine with almost any ISA or PCI 3Com NIC in it (and supposedly PCMCIA, as well), and that works fine with OpenBSD (finding it on their website may be a challenge), so I think they understand how to do PXE now. However, if you are purchasing surplus cards...you might end up with the old ones. I have never found a flash update for the 3Com 3c905 cards I have on 3Com's website (may not be flash devices, of course). The older Intel PXE cards I have tried had problems similar to the 3Com cards, but they could be flashed with updates (for the most part...there appeared to be some that the update utility refused to touch, based on what was already in the ROM). Lesson there /might/ be: old PXE stuff is unreliable, new stuff is better. If you aren't sure how old it is, get something that can be updated. Nick.
Re: PXE capable NICs: non-intel chipsets
Nick Holland wrote: Lesson there /might/ be: old PXE stuff is unreliable, new stuff is better. If you aren't sure how old it is, get something that can be updated. Alternatively, get gigabit cards that support PXE, which should be new enough to avoid problems. I've never seen a gigabit card that didn't support PXE (although they probably exist), and most gigabit cards will perform better than a 100 card even at 100.
Re: avoiding INTEL
I completely agree with Mr. Chehade. Mr. Majid Awad, I do not work for a company that has hundreds of laptop. I only have 1 laptop, but I guarantee you that the next one will not run on Intel. Jean-Daniel Beaubien Gilles Chehade wrote: Mr. Majid Awad, I have recently been provided with a DELL laptop at work which runs Windows and OpenBSD. I have recently bought a VAIO laptop for personal use which runs OpenBSD. These two laptops have a point in common. ALL of the hardware is recognized and works ... except for the Intel wifi chipsets. About four days ago, at an airport, I had written code that I needed to synchronize with the current code base at work. This forced me to keep booting between Windows and my working operating system so that I could actually work and make the changes visible for the people at my company by ... using the network. You can't imagine how irritable one can get when spending time waiting for a couple systems to boot about twenty times within a few hours. This has wasted both me and my company a lot of time and money. And worst of all, this is how we are treated for trusting your company and SPENDING MONEY on its hardware. I just got back from my business trip and first thing I'll do tomorrow is discuss with the decision takers at work so that they contact dell and tell them the Intel based hardware is not working and we need it replaced (oh don't worry that's just about a hundred laptops, it won't affect your sales that much). And for my personal use, this is the last time I buy a laptop or desktop with Intel hardware in it, I have wasted enough time with your integrated graphic chipsets and your integrated wifi chipsets, I can't even recall how many times I got angry at some of your hardware not being working properly (or at all) because you refuse to cooperate with ... your customers. I'd rather spend my money on something that works and keep a smile on my face while working. I am angry and you lost a customer.
Re: PXE capable NICs: non-intel chipsets
Original message Date: Mon, 02 Oct 2006 21:55:38 -0400 From: Steve Shockley [EMAIL PROTECTED] Subject: Re: PXE capable NICs: non-intel chipsets To: misc misc@openbsd.org Nick Holland wrote: Lesson there /might/ be: old PXE stuff is unreliable, new stuff is better. If you aren't sure how old it is, get something that can be updated. Alternatively, get gigabit cards that support PXE, which should be new enough to avoid problems. I've never seen a gigabit card that didn't support PXE (although they probably exist), and most gigabit cards will perform better than a 100 card even at 100. steve, these PXE capable cards are intended for the myriad shitboxes i currently lord it over and i can't justify the additional expense of gigabit cards. additionally, a lot of these machines' network throughput will be limited by their slow disk I/O. nick, thx for the warning about the old ROMs being flaky. i'll order some old intel cards and see what happens. stuart, i'll try the etherboot stuff and post about the results in a week or so. i look forward to not having to reboot my machine twice to swap in a new kernel after a panic :). cheers, jake
ipsec vpn: freebsd and openbsd
Hi guys, I'm setting up ipsec/vpn on freebsd and openbsd. I try to read this how to http://www.securityfocus.com/infocus/1859 but this applies to 2 openbsd systems. could anyone help me on how to setup between two systems? cheers, kintaro0e
Re: PXE capable NICs: non-intel chipsets
Jacob Yocom-Piatt wrote: steve, these PXE capable cards are intended for the myriad shitboxes i currently lord it over and i can't justify the additional expense of gigabit cards. Depends what you call cheap. Newegg's got new gig cards starting at $7, but I guess you could get a box full or 3c905s for that price.
Re: ipsec vpn: freebsd and openbsd
ipsec between freebsd and openbsd didn't turn up anything on Google directly related to what you seem to want to do (at least for me), so I guess you'll have to look at the FreeBSD side of things: http://www.freebsd.org/doc/en_US.ISO8859-1/books/handbook/ipsec.html http://www.onlamp.com/pub/a/bsd/2002/12/26/FreeBSD_Basics.html They seem to mention the use of the gif(4) interface a lot for VPNs between two FreeBSD machines, but I believe you can still make a FreeBSD VPN without it, so I think you'll want to forgo its use if you'll be connecting to an OpenBSD box. I've done FreeBSD -- FreeBSD and OpenBSD -- OpenBSD VPNs in the past, but I don't recall ever doing FreeBSD -- OpenBSD. After seeing how easy it seems to be to make an IPsec VPN using OpenBSD's 'ipsecctl' from the article you mention, I'm guessing the harder part will be the configuration of the FreeBSD side with Racoon, setkey and such. Good luck, -Martin On 10/2/06, kintaro oe [EMAIL PROTECTED] wrote: Hi guys, I'm setting up ipsec/vpn on freebsd and openbsd. I try to read this how to http://www.securityfocus.com/infocus/1859 but this applies to 2 openbsd systems. could anyone help me on how to setup between two systems? -- What the world needs is more geniuses with humility; there are so few of us left. -- Oscar Levant
Re: ipsec vpn: freebsd and openbsd
kintaro oe wrote: I'm setting up ipsec/vpn on freebsd and openbsd. I try to read this how to http://www.securityfocus.com/infocus/1859 but this applies to 2 openbsd systems. could anyone help me on how to setup between two systems? Perhaps OpenVPN is a good alternative? I wrote a setupscript for OpenVPN: http://www.xs4all.nl/~hanb/software/braindead-rsa/ # Han
Re: Intel and Licensing
marrandy wrote on Mon, Oct 02, 2006 at 11:56:44AM -0400: [...] The key component is that source should be open. If you can't provide source then API's have to be open In similar arguments, it might even be better to argue just the other way round: Please provide hardware and firmware documentation. Just in case you have no docs for you product (ps?!?!), then, but only as a second best, please provide driver source code in order to at least simplify reverse engineering for the *BSD kernel developers. But don't take that as an excuse to refrain from properly documenting your product!
Re: ipsec vpn: freebsd and openbsd
I agree with you Han. If Kintaro finds that configuring an IPsec VPN between a FreeBSD and an OpenBSD machine is too complicated, OpenVPN installed on both machines may offer an easier alternative. -Martin On 10/2/06, Han Boetes [EMAIL PROTECTED] wrote: kintaro oe wrote: I'm setting up ipsec/vpn on freebsd and openbsd. I try to read this how to http://www.securityfocus.com/infocus/1859 but this applies to 2 openbsd systems. could anyone help me on how to setup between two systems? Perhaps OpenVPN is a good alternative? I wrote a setupscript for OpenVPN: http://www.xs4all.nl/~hanb/software/braindead-rsa/ -- What the world needs is more geniuses with humility; there are so few of us left. -- Oscar Levant
NIS server
I wonder if some here knows a NIS server (ypserv) that uses openldap as information source! If so, please, let me know. I am desperately searching for a nis server that uses ldap. Thanks in advance.
Intel's Support for Open Source
Mr. Awad and Mr. Ketreno: First, I offer my sincere thanks to Intel for their ongoing support for 965 graphics chipsets at http://intellinuxgraphics.org/ - as a Slackware Linux user, I appreciate having an open-source option for accelerated graphics. However, since I am also an OpenBSD user, I must admit that I am disappointed that Intel's overall commitment to open-source software is rather lacking. As indicated above, releasing source code is certainly appreciated; however, it is not enough. The community really needs accurate, up to date documentation which would allow them to write and/or maintain its own drivers. Another area that could use some work is the licensing terms of the firmware images for various Intel wireless chipsets. I can see no good reason to place any restrictions upon any vendor wishing to distribute unmodified firmware images with their operating system. For any operating system, being able to offer out of the box support for your hardware will result in more sales of your hardware. As it stands, I cannot in good conscience recommend Intel products to either my acquaintances or customers due to the issues mentioned above. Should these issues be addressed satisfactorily, I would have absolutely no qualms about recommending purchase of Intel hardware. Thanks in advance for your time and consideration. Cordially, Robert Workman -- http://rlworkman.net
Re: man rcd missing
Karel Kulhavy wrote on Mon, Oct 02, 2006 at 08:52:00AM +0200: cd (4) talks about /dev/rcd, but if I do man rcd, I get man: no entry for rcd in the manual. I think rcd should symlink to cd. [EMAIL PROTECTED] $ ls /dev/r* | sed 's/[01-9].*//' | uniq /dev/radio /dev/raid /dev/random /dev/rccd /dev/rcd /dev/rd /dev/rfd /dev/rmcd /dev/rmidi /dev/rraid /dev/rrd /dev/rsd /dev/rst /dev/rsvnd /dev/rvnd /dev/rwd /dev/rwt You really want symlinks for ccd cd fd mcd midi raid rd sd st svnd vnd wd wt? That's 13 symlinks for a rather trivial point. On the other hand, why not? This is not the first time people show up here who were confused by raw devices. Such trivial links do exist in other places, even in places where it is less likely that newbies stumble on it, e.g. [EMAIL PROTECTED] $ ls /usr/share/man/cat3/exec* /usr/share/man/cat3/exec.0 /usr/share/man/cat3/execlp.0 /usr/share/man/cat3/execl.0 /usr/share/man/cat3/execv.0 /usr/share/man/cat3/execle.0/usr/share/man/cat3/execvp.0 [EMAIL PROTECTED] $ man 3 LIST_INIT [ snip a long list ]
Re: avoiding INTEL
Dear Mr. Majid Awad, I am working here in Germany for a satellite based remote sensing corporation. Well, we won't go buy thousands of desktops or notebooks, as we do not need that many of them, but we are just in the phase of buying our image processing cluster hardware. I just wanted to let you know that we already, before I read all the discussions here at the list, decided to spend our money on AMD64 based machines instead of Intel based ones. These discussions here about your licensing policies are fortifying me that we made the right decision to not spend money on hardware where the hardware and its drivers are burdened with unfree licenses. I just wanted to let you know that. kind regards Sebastian Justin Blackmore [EMAIL PROTECTED] wrote: I 2ND THAT! -Original Message- Gilles Chehade Sent: Sunday, October 01, 2006 4:17 PM Mr. Majid Awad, I have recently been provided with a DELL laptop at work which runs Windows and OpenBSD. I have recently bought a VAIO laptop for personal use which runs OpenBSD. These two laptops have a point in common. ALL of the hardware is recognized and works ... except for the Intel wifi chipsets. About four days ago, at an airport, I had written code that I needed to synchronize with the current code base at work. This forced me to keep booting between Windows and my working operating system so that I could actually work and make the changes visible for the people at my company by ... using the network. You can't imagine how irritable one can get when spending time waiting for a couple systems to boot about twenty times within a few hours. This has wasted both me and my company a lot of time and money. And worst of all, this is how we are treated for trusting your company and SPENDING MONEY on its hardware. I just got back from my business trip and first thing I'll do tomorrow is discuss with the decision takers at work so that they contact dell and tell them the Intel based hardware is not working and we need it replaced (oh don't worry that's just about a hundred laptops, it won't affect your sales that much). And for my personal use, this is the last time I buy a laptop or desktop with Intel hardware in it, I have wasted enough time with your integrated graphic chipsets and your integrated wifi chipsets, I can't even recall how many times I got angry at some of your hardware not being working properly (or at all) because you refuse to cooperate with ... your customers. I'd rather spend my money on something that works and keep a smile on my face while working. I am angry and you lost a customer.
Re: Looking for HowTo instructions ...
Marc G. Fournier wrote on Sun, Oct 01, 2006 at 10:28:34PM -0300: Can someone that has installed BSDstats on your server please email me instructions on *how* to install it for your flavor of BSD? I doubt the project is worth the effort at all. Whatever numbers might result will be heavily biased. BSDstats is typical bloatware that lots of OpenBSD users will hate (not all, mind you, but many more than e.g. in Linuxland). Besides, the OpenBSD community tends to just not care about marketing. OpenBSD is about correctness, simplicity, freedom and security. Popularity is *not* among the project goals. Most of the developers work on it because they need good code themselves - and certainly not in order to become famous. Thus, i should expect the following attitude from typical OpenBSD users: A software for measuring popularity? How boring. What, it will even run cron scripts and open network connections? No way on my machine...
Re: Was= Re: glib2 fails to build on ppc-current: pthreads issue?
On Mon, 2 Oct 2006, Douglas F. Calvert wrote: glib2 builds fine now but it gives me an error about glib2-docs. I did a fresh checkout of the ports tree and I still get the same error. The error is: make: don't know how to make /usr/ports/devel/glib2/w-glib2-2.10.3/pkg/DESCR-docs. Stop in /usr/ports/devel/glib2. I cannot reproduce this here. -- Antoine
