Early kernel log collection
Hello! While investigating bug #5820 I was wondering why* my OpenBSD system reboots on the very early stage of bootup process (a half a second after kernel gets control over CPU). The problem is that I can't see dmesg log that is generated by kernel right before failure. On Linux, f.e., kernel panic just stops the execution of the kernel, and one can always read pre-panic log messages to see if there is something particular to his problem. Bu OpenBSD for me just reboots. Can I suppress such behavior (by UKC or something else)? If not, are there any ways of an early kernel log collection? I would like to get it by serial line but -- peaty -- I have no such port. Maybe, some sort of network logs? Or maybe I can force ddb entering on panic situation? Tnx for help guys. --- [*] The reason is an ACPI enabled if it matters somehow.
timezone anomalies
hi there, today i wanted to copy the pictures from my camera sd card to my openbsd notebook. after mounting the card i noticed that there are files with future dates... amaaq> ls -la /etc/localtime lrwxr-xr-x 1 root wheel 36 May 21 12:23 /etc/localtime@ -> /usr/share/zoneinfo/Pacific/Auckland amaaq> ls -la -rw-rw-rw- 1 f wheel906429 May 23 05:36 dscn9996.jpg -rw-rw-rw- 1 f wheel877491 May 23 05:37 dscn9997.jpg -rw-rw-rw- 1 f wheel915682 May 23 05:38 dscn9998.jpg amaaq> TZ= ls -la -rw-rw-rw- 1 f wheel906429 May 22 17:36 dscn9996.jpg -rw-rw-rw- 1 f wheel877491 May 22 17:37 dscn9997.jpg -rw-rw-rw- 1 f wheel915682 May 22 17:38 dscn9998.jpg amaaq> echo $TZ amaaq> ls prints the correct date/time when explicitly clearing $TZ. but there is nothing in it in the first place... i had different timezones before, but now i am in new zealand... this has never happened before as far as a i can tell.. ls(1) says that TZ is used to print the dates but it never occured to me that it would add $TZ hours to the file date, that's why i am having filedates in the future, obivously, that the files have correct dates on disk. is this the accepted behaviour? the mounts are ext2fs and msdos btw. 4.3-current as of may 19 -f -- how much can i get away with and still go to heaven?
Re: asus eee ethernet and 4.3
hmm, on Tue, May 20, 2008 at 08:00:20AM +0200, Rolf Sommerhalder said that > Just found that my previous analysis was flawed. The problem is not > related to the length of the patch cable. lii(4) comes up correctly if > the eeePC is connected to the switch at the time when the eeePC is > powered on. However, if the eeePC is not plugged in to the switch at > powerup, then there seems to be no way to bring the link up later. > Even a reboot does not help. Only cycling power appears to bring up > the link to active state, and independent of the length of the patch > cable. interesting. might be connected to the fact that in 4.3 current as of may 19 with acpi enabled and apm disabled (on the 910 bios) acpibat does not drop into ddb anymore and the battery level is reported. only when i pull the plug, the machine freezes. -f -- he's teflon brain (nothing sticks).
Re: timezone anomalies
dual booting with linux these days i am now totally lost. seems like the xandros distro picks up the how clock but the set /etc/localtime didn't do anything. date shows the same as the bios time... could the linux dualbooters help me set up the system so the two os do not fight over time? what is the proper setup? bios: UTC os: timezone or bios: localtime os: localtime and pretend i am in a timezone? (ntpd gets crazy this way) or bios: timezone os: timezone -f -- is that a banana in your pocket, or you happy to see me?
Re: timezone anomalies
On Thu, May 22, 2008 at 02:23:07PM +0200, frantisek holop wrote: | dual booting with linux these days i am now totally lost. | seems like the xandros distro picks up the how clock | but the set /etc/localtime didn't do anything. date | shows the same as the bios time... | | could the linux dualbooters help me set up the system | so the two os do not fight over time? | | what is the proper setup? | | bios: UTC | os: timezone | | or | | bios: localtime | os: localtime and pretend i am in a timezone? (ntpd gets crazy this way) | | or | | bios: timezone | os: timezone I don't quite understand these three options you give. Both OS and BIOS should run in UTC. You configure your environment with TZ which will default to /etc/localtime. That is, do not explicitly set TZ and you get the timezone pointed to by /etc/localtime (should be a symlink to /usr/share/zoneinfo/...), export TZ=Europe/Zurich and get the times as used in Switzerland. Kernel and NTPd just use UTC. Cheers, Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: timezone anomalies
2008/5/22 frantisek holop <[EMAIL PROTECTED]>: > bios: UTC > os: timezone This is how I setup all of my *strictly* *nix machines, be they GNU/Linux or *BSD. > bios: localtime > os: localtime and pretend i am in a timezone? (ntpd gets crazy this way) This is what I do for machines that dual boot MS Windows + *nix. If the machine were mine, dual booting GNU/Linux and OpenBSD, I'd go for the former. Of course, I'm no expert and YMMV. kmw
Re: timezone anomalies
> -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > Behalf Of Paul de Weerd > Sent: Thursday, May 22, 2008 8:40 AM > To: OpenBSD > Subject: Re: timezone anomalies > > On Thu, May 22, 2008 at 02:23:07PM +0200, frantisek holop wrote: > | dual booting with linux these days i am now totally lost. > | seems like the xandros distro picks up the how clock > | but the set /etc/localtime didn't do anything. date > | shows the same as the bios time... > | > | could the linux dualbooters help me set up the system > | so the two os do not fight over time? > | > | what is the proper setup? > | > | bios: UTC > | os: timezone > | > | or > | > | bios: localtime > | os: localtime and pretend i am in a timezone? (ntpd gets crazy > this way) > | > | or > | > | bios: timezone > | os: timezone > > I don't quite understand these three options you give. Both OS and > BIOS should run in UTC. You configure your environment with TZ > which > will default to /etc/localtime. That is, do not explicitly set TZ > and > you get the timezone pointed to by /etc/localtime (should be a > symlink > to /usr/share/zoneinfo/...), export TZ=Europe/Zurich and get the > times > as used in Switzerland. Kernel and NTPd just use UTC. > > Cheers, > > Paul 'WEiRD' de Weerd I know at least for windows it wants to set the BIOS time to local time. Not sure how the linux handles it. If you need/want to have the BIOS time set to local time, you can adjust OpenBSD to handle that. http://www.openbsd.org/faq/faq8.html#TimeZone
Re: Unbound: a validating, recursive, and caching DNS resolver
On Wednesday 21 May 2008, Andris wrote: > I just read about this project, might be of interest: > http://unbound.net/ Not a real world test (well, real in a really small environment - my personal home server) but after I saw your post I installed this on my server running side-by-side with bind (different port of course), both doing recursive resolution from the root servers and the times to resolve dig queries were much faster with unbound. Just a handful of tries, nothing scientific. Have it running full time currently, no issues yet. -- Chris
Why Perl for pkg_* tools ?
Hi list ! Reading through OpenBSD's codebase, I have noticed that the code living under src/usr.sbin/pkg_add is written in Perl. Perl is distributed under the Artistic license, though. The latter is not as permissive as the BSD license under which monst of OpenBSD is released. No doubt that is the reason why Perl lives in src/gnu. Why have such a tool using a non-BSD package when there was choice not to do so ? What technical reasons have lead the developers to elect this language ? I am just curious about the fact and didn't manage to find information in tech@ and mis@ archives. Thanks in advance. Hyjial. __ Do You Yahoo!? En finir avec le spam? Yahoo! Mail vous offre la meilleure protection possible contre les messages non sollicitis http://mail.yahoo.fr Yahoo! Mail
Re: Unbound: a validating, recursive, and caching DNS resolver
On Wed, May 21, 2008 at 10:09 AM, Andris <[EMAIL PROTECTED]> wrote: > I just read about this project, might be of interest: > http://unbound.net/ > > It's developed by Kirei, NLnet Labs, Nominet, and VeriSign; and > released under a permissive free software license: > http://unbound.net/svn/trunk/LICENSE > > I read about it at: > http://tech.slashdot.org/tech/08/05/21/0153201.shtml > > Original source for the article: > http://www.networkworld.com/news/2008/052008-open-source-dns-server.html > > Greetings. > > I wonder if anyone has reviewed or audited this code for bugs, besides the above mentioned developers. I guess time will tell.
NAT Rules
Hello there, We have two seperate datacentres, one using 172.16.1.0/24 and the other using 172.16.2.0/24. In front of both are NAT'ing OpenBSD firewalls, using something like: nat on $ext_if from -> ($ext_if:0) (Where prv_net contains the netblock of that datacentre). Now, I would like that NAT to be conditional on the destination address, such that if a packet from datacentre a (172.16.1.12) was heading to datacentre b (172.16.2.16), then it wouldn't get NAT'ed. Is that possible? How would I do that? Thanks -- joe. Excuse me? Is that your samosa?
Re: NAT Rules
On Thu, May 22, 2008 at 06:18:21PM +0100, Joe Warren-Meeks wrote: Hey there, > We have two seperate datacentres, one using 172.16.1.0/24 and the other > using 172.16.2.0/24. In front of both are NAT'ing OpenBSD firewalls, > using something like: > > nat on $ext_if from -> ($ext_if:0) Ignore me, I just found no nat. -- joe. I have a lot of time for David Pleat.
Re: Help: OpenBSD 4.2 setup VPN gateway for mobile users
Chiah Tong Kiat wrote: > Hi > > Could anyone give me some pointers in setting up a VPN gateway for mobile > users? > > All the current docs that I've seen are for site-to-site VPN. Existing > documents for mobiles uses certpatch to create a SubjectAltName which does > not exist anymore > > Could anyone please help? > > thanks > tongkiat > > I have found OpenVPN to be an easy solution in the past. I've got Linux, Windows, Mac clients all connecting fine. I have heard that IPSec on OpenBSD over the past few releases has gotten much easier to work with. Lots of doc's on the openvpn web site to help. I've also seen some Howto's for OpenBSD specifically. But with any of these, it is really important to understand why you are doing something.
Re: Why Perl for pkg_* tools ?
On Thu, 22 May 2008, hyjial wrote: I am just curious about the fact and didn't manage to find information in tech@ and mis@ archives. Thanks in advance. Hyjial. You didn't try very hard then. This has been discussed on many occasions. g.day
Re: Why Perl for pkg_* tools ?
hyjial wrote: What technical reasons have lead the developers to elect this language ? If you dislike it so much, why don't you rewrite pkg_* in C, and submit patches? -- chs
spamd sync protocol changed
A heads up about spamd. For those heavily using spamd in sync mode, the protocol has changed to fix a few bugs. The protocol has a version number and we incrememnted it as a result. You will need to update all your spamd sync boxes at the same time (or, older boxes and newer boxes will ignore each other's packets).
Re: using lynx to manage router
On Wed, 2008-05-21 at 00:36 +0200, ropers wrote: > s/EMCAScript/ECMAScript > > 2008/5/21 ropers <[EMAIL PROTECTED]>: > > 2008/5/20 Default User <[EMAIL PROTECTED]>: > >> Hello! > >> > >> I would like to use lynx to manage my local small lan router. I can > >> manage a broadband modem that way. But the router webpage expects to be > >> managed by a graphical browser, so the initial control webpage just > >> shows up as unintelligible garbage. > >> > >> Since I run command line only, I do not want to activate X, install a > >> graphical browser, and run X, with all the overhead and security issues, > >> just to manage a simple router. Is there another way text-only way to > >> accomplish this (ie, ssh etc.)? > >> > >> Thanks for any advice. > > > > Since you apparently *require* a text-only browser, have you tried these: > > ELinks > > Links > > w3m > > > > Wikipedia also lists edbrowse, but it doesn't appear to be in ports, > > so YMMV trying to get it to work on OpenBSD. > > > > If you *don't* really *require* a text-only/console browser, ie. if > > there is e.g. a chance to enable SSH on your modem (some of these run > > Linux...), then you'll have to give more details. > > > > Another solution that I could think of might be to use curl/wget to > > fetch the pages you want, and then to write a program/shell script to > > transmogrify the page to something you can use. Of course, in the > > extreme this might require partially implementing an EMCAScript > > interpreter -- assuming that that's what's really missing; not being > > able to see the colourful images should not be much of an issue, but > > most text-based browsers not grokking EMCAScript probably would be. > > > > Hope this helps, > > --ropers > > Thanks for the suggestions, but no luck. Unfortunately, none of the text browsers I tried (lynx, links, elinks, links+, w3m) worked. The router's internal webpage is . It seems to require javascript (ECMAscript), which may well be the problem. SSH to port 22 does not work (it just times out), and telnet replies "connection refused". I am not up to compiling external applications; I try to stick with what's in the OpenBSD packages collection. And of course, the manufacturer's website was absolutely clueless. So, it seems that I can either: 1) just manage the router from another computer with another OS. 2) activate X on the OpenBSD computer and install a graphical browser. If I choose option #2, what what graphical browser would have the least overhead, and above all, do the least damage to my security? I know it's not OpenBSD's fault that the router's control webpage requires javascript, but I am surprised that there doesn't seem to be a simpler, less insecure alternative. Oh, well - so much for security . . .
Re: Why Perl for pkg_* tools ?
On Thu, May 22, 2008 at 9:35 PM, Christer Solskogen <[EMAIL PROTECTED]> wrote: > hyjial wrote: >> >> What technical reasons have lead the >> developers to elect this >> language ? >> Her is an interview with Espie It contains many hints to research from I also thought Espie said that perl enabled them to do some stuff that other tools wouldnt and accompplish it faster, i actually thought it was this interview but couldnt find it with a quick glance. http://mongers.org/openbsd/interview-espie-ports
Re: Why Perl for pkg_* tools ?
On Thu, May 22, 2008 at 03:23:17PM +, hyjial wrote: | Hi list ! | Reading through OpenBSD's codebase, I have noticed that the code | living | under src/usr.sbin/pkg_add is written in Perl. Perl is distributed | under the Artistic license, though. The latter is not as permissive | as the BSD | license under which monst of OpenBSD is released. No doubt | that is the reason | why Perl lives in src/gnu. | Why have such a tool using a non-BSD package when | there was choice | not to do so ? | What technical reasons have lead the | developers to elect this | language ? | I am just curious about the fact and | didn't manage to find information | in tech@ and mis@ archives. So, first of .. your indenting could use some help... Anyway, perl is distributed under the artistic license, yet the pkg-tools are licensed under an ISC-style license. Compare, if you will, with most other tools in OpenBSD. They're C programs with an ISC or BSD-style license. However, GCC is distributed under the GPL. Boo-freakidy-hoo .. why make a problem of the perl license now, is bashing GCC's license not fun anymore ? You know, if you want, you could write an ISC-licensed perl interpreter. Go right ahead and feel free to send patches when you're done. I'll suggest a name for you : 'hurl'. If you're done, could you please write an ISC-licensed C-compiler in perl so I can finally shut up all the idiots that claim that a system without a compiler is more secure ? Don't worry, I can wait. Cheers, Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: using lynx to manage router
On May 22, 2008, at 4:09 PM, Default User wrote: On Wed, 2008-05-21 at 00:36 +0200, ropers wrote: s/EMCAScript/ECMAScript 2008/5/21 ropers <[EMAIL PROTECTED]>: 2008/5/20 Default User <[EMAIL PROTECTED]>: Hello! I would like to use lynx to manage my local small lan router. I can manage a broadband modem that way. But the router webpage expects to be managed by a graphical browser, so the initial control webpage just shows up as unintelligible garbage. Since I run command line only, I do not want to activate X, install a graphical browser, and run X, with all the overhead and security issues, just to manage a simple router. Is there another way text-only way to accomplish this (ie, ssh etc.)? Thanks for any advice. Since you apparently *require* a text-only browser, have you tried these: ELinks Links w3m Wikipedia also lists edbrowse, but it doesn't appear to be in ports, so YMMV trying to get it to work on OpenBSD. If you *don't* really *require* a text-only/console browser, ie. if there is e.g. a chance to enable SSH on your modem (some of these run Linux...), then you'll have to give more details. Another solution that I could think of might be to use curl/wget to fetch the pages you want, and then to write a program/shell script to transmogrify the page to something you can use. Of course, in the extreme this might require partially implementing an EMCAScript interpreter -- assuming that that's what's really missing; not being able to see the colourful images should not be much of an issue, but most text-based browsers not grokking EMCAScript probably would be. Hope this helps, --ropers Thanks for the suggestions, but no luck. Unfortunately, none of the text browsers I tried (lynx, links, elinks, links+, w3m) worked. The router's internal webpage is HTML 4.0 TANSITIONAL//EN">. It seems to require javascript (ECMAscript), which may well be the problem. SSH to port 22 does not work (it just times out), and telnet replies "connection refused". I am not up to compiling external applications; I try to stick with what's in the OpenBSD packages collection. And of course, the manufacturer's website was absolutely clueless. So, it seems that I can either: 1) just manage the router from another computer with another OS. 2) activate X on the OpenBSD computer and install a graphical browser. If I choose option #2, what what graphical browser would have the least overhead, and above all, do the least damage to my security? I know it's not OpenBSD's fault that the router's control webpage requires javascript, but I am surprised that there doesn't seem to be a simpler, less insecure alternative. Oh, well - so much for security . . . 3) Stop using closed crappy proprietary routers? You obviously have the acumen to install and use openbsd. Why not use OpenBSD as your gateway machine? for mild to moderate connections one of the cheap soekris running on a compact flash card works fantastically IMO. -Adam
Re: using lynx to manage router
On 5/22/08, Default User <[EMAIL PROTECTED]> wrote: > If I choose option #2, what what graphical browser would have the least > overhead, and above all, do the least damage to my security? > > I know it's not OpenBSD's fault that the router's control webpage > requires javascript, but I am surprised that there doesn't seem to be a > simpler, less insecure alternative. Oh, well - so much for > security . . . You are expecting your router to attack your browser? I think you need a new router.
Re: Why Perl for pkg_* tools ?
Paul de Weerd wrote: On Thu, May 22, 2008 at 03:23:17PM +, hyjial wrote: | Hi list ! | Reading through OpenBSD's codebase, I have noticed that the code | living | under src/usr.sbin/pkg_add is written in Perl. Perl is distributed | under the Artistic license, though. The latter is not as permissive | as the BSD | license under which monst of OpenBSD is released. No doubt | that is the reason | why Perl lives in src/gnu. | Why have such a tool using a non-BSD package when | there was choice | not to do so ? | What technical reasons have lead the | developers to elect this | language ? | I am just curious about the fact and | didn't manage to find information | in tech@ and mis@ archives. So, first of .. your indenting could use some help... Anyway, perl is distributed under the artistic license, yet the pkg-tools are licensed under an ISC-style license. Compare, if you will, with most other tools in OpenBSD. They're C programs with an ISC or BSD-style license. However, GCC is distributed under the GPL. Boo-freakidy-hoo .. why make a problem of the perl license now, is bashing GCC's license not fun anymore ? You know, if you want, you could write an ISC-licensed perl interpreter. Go right ahead and feel free to send patches when you're done. I'll suggest a name for you : 'hurl'. If you're done, could you please write an ISC-licensed C-compiler in perl so I can finally shut up all the idiots that claim that a system without a compiler is more secure ? Don't worry, I can wait. Cheers, Paul 'WEiRD' de Weerd Don't be so defensive. He said he didn't manage to find information on the mailing lists. Where did you want him to ask an honest question? "What technical reasons have lead the developers to elect this language ?" Since when is that a question provoking sarcasm and anger? Its curiosity. Same thing that got most of us here at some point or another. Everyone is so quick to be the first with a nasty response.
glxsb?
On Tue, May 20, 2008 at 4:34 PM, Paul de Weerd <[EMAIL PROTECTED]> wrote: > glxsb (4/i386) - Geode LX Security Block crypto accelerator > > In other words, there's onboard crypto support in these machines that > is supported in OpenBSD. You may not need a separate accelerator. Thanks for the reminder, I forgot the (slightly more expensive) Net5501 had this chip :) Does this just automagically accelerate anything using entropy or AES? Is there any way to temporarily disable acceleration to run benchmarks? Thanks, Kevin
Re: using lynx to manage router
On May 22, 2008, at 8:44 PM, Ted Unangst wrote: On 5/22/08, Default User <[EMAIL PROTECTED]> wrote: If I choose option #2, what what graphical browser would have the least overhead, and above all, do the least damage to my security? I know it's not OpenBSD's fault that the router's control webpage requires javascript, but I am surprised that there doesn't seem to be a simpler, less insecure alternative. Oh, well - so much for security . . . You are expecting your router to attack your browser? I think you need a new router. More like insecure javascripting leading to XSS attacks? Not something that can't occur sans javascript, but making the entire interface javascriptish definitely complicates things. Plus, it's "internal" only, so why does the web interface need secure handling? -Adam
Re: glxsb?
On May 22, 2008, at 9:27 PM, K K wrote: On Tue, May 20, 2008 at 4:34 PM, Paul de Weerd <[EMAIL PROTECTED]> wrote: glxsb (4/i386) - Geode LX Security Block crypto accelerator In other words, there's onboard crypto support in these machines that is supported in OpenBSD. You may not need a separate accelerator. Thanks for the reminder, I forgot the (slightly more expensive) Net5501 had this chip :) Does this just automagically accelerate anything using entropy or AES? Is there any way to temporarily disable acceleration to run benchmarks? Thanks, Kevin I was under the impression that kern.usercrypto did this. it seems to have a negligible affect on my net5501 I do have a glxsb -=[~]=- -=[Thu May 22]=- -=[21:58:20]=- [EMAIL PROTECTED] (set -ex;sysctl kern.usercrypto=1;openssl speed -evp aes-256-cbc;sysctl kern.usercrypto=0;openssl speed -evp aes-256-cbc) + sysctl kern.usercrypto=1 kern.usercrypto: 1 -> 1 + openssl speed -evp aes-256-cbc To get the most accurate results, try to run this program when this computer is idle. Doing aes-256-cbc for 3s on 16 size blocks: 721327 aes-256-cbc's in 2.71s Doing aes-256-cbc for 3s on 64 size blocks: 216391 aes-256-cbc's in 2.91s Doing aes-256-cbc for 3s on 256 size blocks: 54838 aes-256-cbc's in 2.85s Doing aes-256-cbc for 3s on 1024 size blocks: 13739 aes-256-cbc's in 2.86s Doing aes-256-cbc for 3s on 8192 size blocks: 1722 aes-256-cbc's in 2.94s OpenSSL 0.9.7j 04 May 2006 built on: date not available options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) blowfish(idx) compiler: information not available available timing options: USE_TOD HZ=100 [sysconf value] timing function used: getrusage The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes256 bytes 1024 bytes 8192 bytes aes-256-cbc 4257.29k 4765.26k 4923.10k 4920.21k 4802.25k + sysctl kern.usercrypto=0 kern.usercrypto: 1 -> 0 + openssl speed -evp aes-256-cbc To get the most accurate results, try to run this program when this computer is idle. Doing aes-256-cbc for 3s on 16 size blocks: 758660 aes-256-cbc's in 2.84s Doing aes-256-cbc for 3s on 64 size blocks: 212083 aes-256-cbc's in 2.84s Doing aes-256-cbc for 3s on 256 size blocks: 55383 aes-256-cbc's in 2.87s Doing aes-256-cbc for 3s on 1024 size blocks: 13931 aes-256-cbc's in 2.88s Doing aes-256-cbc for 3s on 8192 size blocks: 1749 aes-256-cbc's in 2.88s OpenSSL 0.9.7j 04 May 2006 built on: date not available options:bn(64,32) md2(int) rc4(idx,int) des(ptr,risc1,16,long) aes(partial) blowfish(idx) compiler: information not available available timing options: USE_TOD HZ=100 [sysconf value] timing function used: getrusage The 'numbers' are in 1000s of bytes per second processed. type 16 bytes 64 bytes256 bytes 1024 bytes 8192 bytes aes-256-cbc 4268.50k 4773.03k 4944.93k 4961.86k 4970.08k -=[~]=- -=[Thu May 22]=- -=[21:59:29]=- [EMAIL PROTECTED] grep glxsb /var/run/dmesg.boot glxsb0 at pci0 dev 1 function 2 "AMD Geode LX Crypto" rev 0x00: RNG AES
Re: Decipering "Understanding IP addressing"
ropers wrote: > 2008/5/21 ropers <[EMAIL PROTECTED]>: >>> On Wed, May 21, 2008 at 1:36 PM, Kendall Shaw <[EMAIL PROTECTED]> wrote: >>> ... I'm having a hard time understanding it. In many places they use 2 numbers, e.g. 2(21) or 232 (4,294,967,296). Can you understand what they are saying? >> >> I am really heartened to see how quickly everybody here has responded >> and pointed out the error and correction. >> >> I am less delighted that 3com, who I emailed about this probably over >> 2 years ago, and who said they were going to look into this, *still* >> haven't fixed their PDF. >> >> Maybe if everybody who responded to this thread were to email them as >> well, *maybe* that would help. > > Or we could just post some errata at > http://www.openbsd.org/faq/faq6.html#Intro , where the PDF is liked. > Would the FAQ maintainers be in favour of this? If so, then I could > probably write a diff for http://www.openbsd.org/faq/faq6.html . s/Or/And/ :) sometimes, the best thing one can do with a bug is document it, which may prompt people to say, "hey, that sucks!", and fix it. Another solution would be to find/write a replacement document/site. (HTML preferred over pdf). Show me a diff. I am not interested in the FAQ being a errata sheet for someone else's document, but an advisory that the typography is hosed would not be bad (a better resource would be..uh..better). Nick.
FuGhu - últimas fechas -
FuGhu zltimas fechas despuis de ser elegidos por Dream Theater para abrir su show en el Luna Park y antes de que Santiago B|rgi se vaya durante tres meses invitado por el Teatro Schvnbrunn de Austria a cantar Die Fledermaus -- Dos recitales totalmente diferentes para despedir a FuGhu hasta la primavera 5/06/2008 FuGhu - Temple of Infinity Liberarte (ctes 1555) a las 21:30hs entrada $15 - La calma antes de la tormenta - 13/06/2008 FuGhu - Acid Rain Teatro Colonial ( P.Colon y Belgrano) a las 21:30 entrada $15 - La Tormenta despuis de la calma - No te quedes afuera.Haci doblete y adquirm las entradas para los dos shows el dma el 5 en liberarte por $20 -- Es un sonido nuevo y potente. Es Rock fuerte con la profundidad de la mzsica clasica y ramces Argentinas asomando entre las grietas. Es una banda que siempre sorprende en vivo, que propone, que divierte, que incomoda, siempre impredecible. Son pesados, son profundos, raros, y a veces suaves y dulces como el veneno de FuGhu. Fotos del zltimo show www.angel-photos.com.ar/fughu Videos de Fughu www.myspace.com/fughuweb www.fughuweb.com.ar www.fotolog.com/fughuwww.acid-rain.com.ar
Re: OpenOSPFD warning
Redirecting to misc@, as that is more appropriate (although I have my doubts, see below) On Thu, May 22, 2008 at 05:50:48PM -0700, Jim Engeseth wrote: | I intalled openospf. What do you mean ? It comes with the base OS, so this seems like a weird statement to make. Do you mean that you've installed OpenBSD to be able to use ospf with OpenOSPFD ? Or are you, by any chance, not running OpenBSD ? | When I enter "ospfd" I received message: | ospfd: /usr/local/etc/ospfd.conf: group/world readable/writeable | | is this good or bad? the openospf process is not started. You will need to chmod g-rw /usr/local/etc/ospfd.conf, apparently. This is what the error is telling you. The fact that it's not started after you get that message should be an indicator if it's good or bad. It's weird though, since the configuration file for OpenOSPFD should live in /etc/ .. are you by any chance not using OpenBSD ? | When I enter "ospfctl" I received message: | ospfctl: connect: /var/run/ospfd.sock: No such file or directory That's probably because ospfd is not running. Once it's running, it'll open a unix domain socket in /var/run/ so it can be controlled with ospfctl(8). If it's not running, the socket should not be there since there's nothing to control. | This message does not look good | | any ideas on what I'm missing Yeah, I have an idea. I think you're missing some details about your setup in this e-mail. How about telling us what OS you're running. This is an OpenBSD mailinglist, you know. Cheers, Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: Why Perl for pkg_* tools ?
On Thu, May 22, 2008 at 07:55:48PM -0500, Adam Patterson wrote: > Paul de Weerd wrote: >> On Thu, May 22, 2008 at 03:23:17PM +, hyjial wrote: >> | Hi list ! >> | Reading through OpenBSD's codebase, I have noticed that the code >> | living >> | under src/usr.sbin/pkg_add is written in Perl. Perl is distributed >> | under the Artistic license, though. The latter is not as permissive >> | as the BSD >> | license under which monst of OpenBSD is released. No doubt >> | that is the reason >> | why Perl lives in src/gnu. >> | Why have such a tool using a non-BSD package when >> | there was choice >> | not to do so ? >> | What technical reasons have lead the >> | developers to elect this >> | language ? >> | I am just curious about the fact and >> | didn't manage to find information >> | in tech@ and mis@ archives. >> >> So, first of .. your indenting could use some help... >> >> Anyway, perl is distributed under the artistic license, yet the >> pkg-tools are licensed under an ISC-style license. >> >> Compare, if you will, with most other tools in OpenBSD. They're C >> programs with an ISC or BSD-style license. However, GCC is distributed >> under the GPL. Boo-freakidy-hoo .. why make a problem of the perl >> license now, is bashing GCC's license not fun anymore ? >> >> You know, if you want, you could write an ISC-licensed perl >> interpreter. Go right ahead and feel free to send patches when you're >> done. I'll suggest a name for you : 'hurl'. If you're done, could you >> please write an ISC-licensed C-compiler in perl so I can finally shut >> up all the idiots that claim that a system without a compiler is more >> secure ? Don't worry, I can wait. >> >> Cheers, >> >> Paul 'WEiRD' de Weerd >> >> > Don't be so defensive. He said he didn't manage to find information on the > mailing lists. Where did you want him to ask an honest question? I don't know. If you come here I'd expect informed questions. What's the use of discussing the license of the interpreter of the software when talking about the software ? > "What technical reasons have lead the developers to elect this language ?" > Since when is that a question provoking sarcasm and anger? Its curiosity. > Same thing that got most of us here at some point or another. > Everyone is so quick to be the first with a nasty response. The useless discussion on licenses beyond the control of the developers coupled with the poor formatting provoked some sarcasm, yes. Marc Espie, who wrote most of the code, gave us the pkg-tools under an ISC license. The reasons for his choice of language have been documented on the OpenBSD mailinglists. I was not 'quick' or 'trying to be the first' (a useless effort when you're replying to a mail that has already been replied to, by the way), just pointing out (in a sarcastic way, I will grant you that) that it's mostly a fruitless discussion. There's a difference between : "What technical reasons have lead the developers to elect this language ?" and "[Perl is not BSD licensed] What technical reasons have lead the developers to elect this language ?" The first is asking a technical question, the second is bringing politics into your techincal question. What do you want, a technical discussion or a political discussion ? As had been pointed out, the technical question had already been answered, the political discussion (I think) merits a sarcastic answer, as this has definitely been discussed over and over and over again. If you don't like the license on perl, you are free to implement the language on your own and license the result any way you like. I just don't see how its license is of any relevance to the software you write in it. OpenBSD comes with perl. It's not going away. Why not use it ? How is it different to using GPL'd GCC to compile ISC'd code ? In the latter case, everybody seems to understand that the license of the compiler has little to do with the license of the code it compiles. The political discussion about using GPL'd GCC and the technical discussion about using C for the base OS have so far been completely separate. The intent of my sarcastic mail was to point out that these two are best kept separate. Obviously, I failed. Paul 'WEiRD' de Weerd [arguing because I'm Dutch] -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: taskjuggler problems
On Wed, May 21, 2008 at 2:12 PM, Vijay Sankar <[EMAIL PROTECTED]> wrote: ... > On OpenBSD 4.3 (i386) I am not able to run TaskJugglerUI 2.3.1p2. My previous > OpenBSD 4.1 and 4.2 desktops had TaskJuggler 2.3.1 and it worked without any > problems. > > TaskJugglerUI:/usr/local/lib/libqt-mt.so.31.1: undefined > symbol 'pthread_mutexattr_init' > lazy binding failed! > Segmentation fault (core dumped) What's the output of ldd /usr/local/lib/libqt-mt.so.31.1 ? How about nm -u /usr/local/lib/libqt-mt.so.31.1 | grep pthread_mutexattr_init ? If the latter shows anything, but the former _doesn't_ mention "libpthread.so.9.0" then the qt library wasn't built correctly. If this is indeed the case, I suppose it would be possible to work around by creating a stub libqt-mt.so.31.1 shared library that just has two dependencies: the real libqt-mt.so and libpthread.so... Philip Guenther
