Problems with PPPoE, VLAN, 5.5 (amd64)
Under 5.4-stable, following configuration sets up a working connection for me: /etc/hostname.pppoe0: inet 0.0.0.0 255.255.255.255 NONE \ pppoedev vlan10 authproto pap \ authname 'XXX' authkey 'YYY' up dest 0.0.0.1 !/sbin/route add default -ifp pppoe0 0.0.0.1 #eof /etc/hostname.vlan10: vlan 10 vlandev rl0 up #eof /etc/hostname.rl0: up lladdr 12:34:56:78:90:ab #needed for assignment of 'static' IP #eof Having installed OpenBSD-5.5 amd64 the exact same configuration does not succeed in setting up a working network connection. I set the 'debug' flag for pppoe0 so I could see what happened: pppoe(8863) state=1, session=0x0 output -> ff:ff:ff:ff:ff:ff, len=18 pppoe(8863) state=2, session=0x0 output -> aa:aa:aa:aa:aa:aa, len=42 pppoe0: timeout [...] aa:aa:aa:aa:aa:aa is a MAC I do not know. To further check for errors, I booted a working 5.4 installation and used tcpdump on vlan10 before restarting the pppoe interface. As expected, I get only one PADO after sending the PADI. This PADO is received from a well known MAC. >From here on, everything just works using the 5.4 installation. On 5.5, until this point (PADI sent/PADO received) there is no difference. For 5.4, the PADR is sent to MAC which sent PADO. For 5.5, the PADR is sent to some other MAC I do not know. There was neither a PADO sent from that MAC nor is there any other occurence of that MAC in the tcpdump besides being target of PADR. In both installations, the PADO is sent from the same MAC. Service-Name, AC-Cookie and Host-Uniq do match. Interestingly, the PADR to the unknown MAC includes the same AC-Cookie. I do not know how to proceed from here on, so I would be glad to get some advice. Best Regards, Thorsten Bonck
Re: Problems with PPPoE, VLAN, 5.5 (amd64)
> maybe you could try to put pppoe0 on rl0, untag vlan10 on switch port
> where rl0 is connected and tag other vlans on the same port ...
Sadly, that is not possible for me.
rl0 is directly connected to a Ubiquiti NanoStation M, which is setup as
a blackbox transparent bridge by my ISP.
vlan id 10 is also required by my ISP for all PPP/PPPoE stuff.
I could put a tagging switch in between rl0 and the bridge as a last
resort, but for right now I am still looking for a reason why 5.4 works
and 5.5 seems to answer to a wrong MAC.
Looking through the changes from 5.4 I found "Stop defining
SIOC{S,G}ETVLAN as SIOC{S,G}IFGENERIC. Lets vlan(4) handle
pppoe(4)-specific ioctls and vice versa without smashing the stack of
the caller". Is it possible my problem is somehow connected to that?
Re: /var/mail permissions?
On Wed, 8 May 2002, J.D. Bronson wrote: > What is the correct permissions for /var/mail? $ sudo chown root:wheel /var/mail; sudo chmod 1777 /var/mail In this order. Not 1. And not g+s, that is only required on SYSV UNIXB.. > I am concerned about this as some POP3 daemons make the file rw-rw and > therefore any other user could read email! Then they suck ;) //mirabilos -- tarent Gesellschaft fC
greytrapping and sendmail
Hi all, imagine you have an address like "junk@" added to the list of traps. Now, someone sends mails to you (or, via spamlogd, vice versa) then spams the trap address. Sendmail will let him pass, and either reject junk@ or have it alia- sed to /dev/null. Not with me b: Diffs slightly hand-edited to only contain relevant material. == /etc/changelist diffs (-OLD +NEW) == --- /var/backups/etc_changelist.current Tue Aug 12 22:51:27 2008 +++ /etc/changelist Mon Jan 12 15:48:29 2009 @@ -56,14 +56,15 @@ /etc/mail.rc /etc/mail/access /etc/mail/aliases /etc/mail/genericstable /etc/mail/helpfile +/etc/mail/junk.sh /etc/mail/local-host-names /etc/mail/localhost.cf /etc/mail/mailertable /etc/mail/relay-domains /etc/mail/sendmail.cf /etc/mail/spamd.conf /etc/mail/submit.cf /etc/mail/trusted-users /etc/mail/virtusertable == /etc/mail/aliases diffs (-OLD +NEW) == --- /var/backups/etc_mail_aliases.current Fri Jan 9 14:54:44 2009 +++ /etc/mail/aliases Mon Jan 12 15:20:26 2009 @@ -81,3 +81,4 @@ # msgs: "|/usr/bin/msgs -s" root: [b&] +junk: "|/bin/mksh /etc/mail/junk.sh" == /etc/sudoers diffs (-OLD +NEW) == --- /var/backups/etc_sudoers.currentWed Jan 7 14:59:29 2009 +++ /etc/sudoersMon Jan 12 15:37:54 2009 @@ -42,3 +42,5 @@ # Samples # %users ALL=/sbin/mount /cdrom,/sbin/umount /cdrom # %users localhost=/sbin/shutdown -h now + +%daemonALL=(root) NOPASSWD: /usr/sbin/spamdb The sudoers line is needed to make c /etc/mail/junk.shc able to delete the whitelisting line and add a TRAPPED entry. It runs as daemon, as itbs listed in /etc/mail/aliases (do not forget to (cd /etc/mail; sudo make) after editingB9), although I considered adding a LOCAL_MAILER to openbsd-proto.mc espe- cially for this. Now, herebs junk.shB2: -BEGIN CUTTING HERE MAY DAMAGE YOUR SCREEN SURFACE- #!/bin/mksh export PATH=/bin:/sbin:/usr/bin:/usr/sbin:/usr/X11R6/bin:/usr/mpkg/bin:/usr/mpkg/sbi n:/usr/local/bin:/usr/local/sbin # yeah, we all love regular expressions and shell scripts like this one b: # you are not expected to understand this # b" change all "`" to "_" # b" change all newline to "`" # b" change all "` " (space or tab at beginning of line) to just a space # b" change back "`" to newline (above ones were joined) # b" take the first line beginning with "Received" # b" get the IPv4 or IPv6 address of the sender, without the "IPv6:" praefix # you are not expected to understand the above textual description # eitherB3, just look at the actual shell script line then :C> ip=$(tr '`\n' '_`' | sed 's/`[ ]/ /g' | tr '`' '\n' | \ grep '^Received' | head -1 | \ sed -n 's/^[^(]*: from [^(]* ([^]]*\[\(IPv6:\)*\([^]]*\)\]) by .*$/\2/p') if [[ -n $ip ]]; then # delete from whitelist sudo spamdb -d "$ip" # trap for 24 hours sudo spamdb -t -a "$ip" fi exit 0 -END CUTTING HERE MAY DAMAGE YOUR SCREEN SURFACE- If anyone has a better method for extracting the IPv4 or IPv6 address, Ibd be delighted to hear it. Using a custom mailer would actually make some macros available, but I donbt know if the sender socket address is among them. bye, //mirabilos b Kudos to OpenBSD for adding a Makefile there, by the way. Nice idea. b! While I did install http://mirbsd.de/mksh on the box, oksh (OpenBSD /bin/ksh) _might_ also work. Actually, should. I have an unofficial port for mksh waiting for anyone to commit itb& b" Like some of my colleagues b; -- tarent Gesellschaft fC
Re: xterm initial tty weirdness
Alexander Hall beard.se> writes: >Editing commands sometimes don't (seem to) work on the first shell >prompt in newly opened xterm window: > > $ bar^A^E^Afoo Interesting. I sometimes have that after exiting a curses application (e.g. lynx) or less.
Re: Problem with PPPoE
On Tue, Jul 22, 2014 at 05:28:46PM -0300, Felipe Mesquita de Oliveira wrote: > > # ifconfig pppoe0 > pppoe0: flags=28855 > mtu 1492 > priority: 0 > dev: bge0 state: PADI sent > sid: 0x0 PADI retries: 7 PADR retries: 0 > sppp: phase establish authproto pap authname "x...@xxx.com.br" > groups: pppoe > status: no carrier > inet 0.0.0.0 --> 0.0.0.1 netmask 0x > > > I've seen some messages discussing this, like this one: > http://comments.gmane.org/gmane.os.openbsd.misc/212649 The problem in the link was related to PPPoE over VLAN using a rl(4) card as physical interface. It seems to be caused by the card being unable to do VLAN tag insertion/stripping in hardware. After changing to a em(4) card, the problem was simply gone for me. Since your pppoe0 uses bge(4) and no VLAN, it has to be caused by something else. Like others suggested, capture the PPPoE negotiation using tcpdump and look for oddities like wrong MACs. Best Regards, Thorsten
Re: hp proliant dl 320e gen 8 for openbsd 5.5 64 bit ?
On Thu, Aug 07, 2014 at 02:08:02PM +0530, Indunil Jayasooriya wrote: > We got this server with 2 x 1 TB hard disks. Because we want a mirrored > Hard drive with 1 TB. But We got it WITHOUT getting it mirrored. > > We can tell the Hardware vendor to mirror this 2 hard disks. > > My question is after mirroring, Will OpenBSD work or again will have to > change BIOS settings ? Have a look at softraid(4), which explains how to set up and install OpenBSD on a RAID 1 like device. Best Regards, Thorsten
Re: sudo bad practice or inconsistency?
Alessandro DE LAURENZIS gmail.com> writes: (line-wrapped because of GMane) > #define SUDOCMD "-fn 7x14 -geometry 60x4 -e sudo su -c 'nohup \ > xfe >& /dev/null & sleep 1'" ^^ Note that this will not work on OpenBSD anyway; even mksh, which does implement this bashism, will not parse this as expected in POSIX mode. So, besides the changes the others already pointed out: #define SUDOCMD "-fn 7x14 -geometry 60x4 -e sudo su root -c 'nohup \ xfe >/dev/null 2>&1 & sleep 1'" bye, //mirabilos
Re: Problems with PPPoE, VLAN, 5.5 (amd64)
On Fri, May 02, 2014 at 08:14:40PM +, Peter J. Philipp wrote:
> On Fri, May 02, 2014 at 09:14:16PM +0200, thors...@bonck.net wrote:
> > > maybe you could try to put pppoe0 on rl0, untag vlan10 on switch port
> > > where rl0 is connected and tag other vlans on the same port ...
> >
> > Sadly, that is not possible for me.
> > rl0 is directly connected to a Ubiquiti NanoStation M, which is setup as
> > a blackbox transparent bridge by my ISP.
> > vlan id 10 is also required by my ISP for all PPP/PPPoE stuff.
> > I could put a tagging switch in between rl0 and the bridge as a last
> > resort, but for right now I am still looking for a reason why 5.4 works
> > and 5.5 seems to answer to a wrong MAC.
> >
> > Looking through the changes from 5.4 I found "Stop defining
> > SIOC{S,G}ETVLAN as SIOC{S,G}IFGENERIC. Lets vlan(4) handle
> > pppoe(4)-specific ioctls and vice versa without smashing the stack of
> > the caller". Is it possible my problem is somehow connected to that?
>
> Hello,
>
> I'd probably do a tcpdump -v -n -e -s 200 -X -i rl0 as well.. perhaps you
> can see somewhere where the aa:aa:aa:aa:aa:aa is coming from. Perhaps there
> is an offset logic error somewhere as the pppoe driver reads the PADO source
> address from the mbuf etherheader which was prepended. If you can't make
> out any 's in the frame then that's further suspicous.
After some staring at the dumped frames, I can describe the problem more
detailed.
Assume listed devices to have following macs:
rl011:22:33:44:55:66
PPPoE server aa:bb:cc:dd:ee:ff
rl0 broadcasts PADI to ff:ff:ff:ff:ff:ff.
PPPoE servers mac sends a PADO to rl0 mac.
rl0 sends a PADR to mac aa:bb:55:66:aa:bb.
timeout happens, since no device with aa:bb:55:66:aa:bb exists in the
network.
Now change rl0 mac to 11:22:33:44:77:88:
PADI,PADO phase unchanged.
rl0 sends a PADR to mac aa:bb:77:88:aa:bb.
I only have a very basic understanding of C and basically no
understanding of the OpenBSD kernel.
The PADR should be assembled around line 703 in src/sys/net/if_pppoe.c
but I could not find any clues where this mix up happens.
Best Regards,
Thorsten
Re: Problems with PPPoE, VLAN, 5.5 (amd64)
On Wed, May 07, 2014 at 02:11:09AM +, Stuart Henderson wrote:
> On 2014-05-02, Thorsten Bonck wrote:
> > On Fri, May 02, 2014 at 08:14:40PM +, Peter J. Philipp wrote:
> >> On Fri, May 02, 2014 at 09:14:16PM +0200, thors...@bonck.net wrote:
> >> > > maybe you could try to put pppoe0 on rl0, untag vlan10 on switch port
> >> > > where rl0 is connected and tag other vlans on the same port ...
> >> >
> >> > Sadly, that is not possible for me.
> >> > rl0 is directly connected to a Ubiquiti NanoStation M, which is setup as
> >> > a blackbox transparent bridge by my ISP.
> >> > vlan id 10 is also required by my ISP for all PPP/PPPoE stuff.
> >> > I could put a tagging switch in between rl0 and the bridge as a last
> >> > resort, but for right now I am still looking for a reason why 5.4 works
> >> > and 5.5 seems to answer to a wrong MAC.
> >> >
> >> > Looking through the changes from 5.4 I found "Stop defining
> >> > SIOC{S,G}ETVLAN as SIOC{S,G}IFGENERIC. Lets vlan(4) handle
> >> > pppoe(4)-specific ioctls and vice versa without smashing the stack of
> >> > the caller". Is it possible my problem is somehow connected to that?
> >>
> >> Hello,
> >>
> >> I'd probably do a tcpdump -v -n -e -s 200 -X -i rl0 as well.. perhaps you
> >> can see somewhere where the aa:aa:aa:aa:aa:aa is coming from. Perhaps
> >> there
> >> is an offset logic error somewhere as the pppoe driver reads the PADO
> >> source
> >> address from the mbuf etherheader which was prepended. If you can't make
> >> out any 's in the frame then that's further suspicous.
> >
> > After some staring at the dumped frames, I can describe the problem more
> > detailed.
> > Assume listed devices to have following macs:
> >
> > rl011:22:33:44:55:66
> > PPPoE server aa:bb:cc:dd:ee:ff
> >
> > rl0 broadcasts PADI to ff:ff:ff:ff:ff:ff.
> > PPPoE servers mac sends a PADO to rl0 mac.
> > rl0 sends a PADR to mac aa:bb:55:66:aa:bb.
> > timeout happens, since no device with aa:bb:55:66:aa:bb exists in the
> > network.
> >
> > Now change rl0 mac to 11:22:33:44:77:88:
> > PADI,PADO phase unchanged.
> > rl0 sends a PADR to mac aa:bb:77:88:aa:bb.
> >
> > I only have a very basic understanding of C and basically no
> > understanding of the OpenBSD kernel.
> > The PADR should be assembled around line 703 in src/sys/net/if_pppoe.c
> > but I could not find any clues where this mix up happens.
> >
> > Best Regards,
> > Thorsten
> >
> >
>
> I have pppoe-over-vlan working fine on 5.5 here. (More specifically:
> multiple pppoe interfaces with pppoedev on a vlan, vlandev on a
> trunk, failover trunk over a pair of bnx - trunk/bnx both show
> VLAN_HWTAGGING in "ifconfig hwfeatures").
>
> Considering what may be different between your setup and mine,
> I'm wondering if it could be either 1) your use of lladdr to adjust
> the MAC address or 2) use of rl(4) which doesn't do hw tagging..
>
Thanks to your hint to VLAN_HWTAGGING, I setup another installation on a
machine that provides a em(4) nic which has this feature.
Now everything, including the lladdr, just works.
Thanks to everyone who helped so far!
Best Regards,
Thorsten
Re: svnd is incredible slow... somebody else notice that?
Sebastian Rother jpberlin.de> writes: > vnconfig -cK 52527 -S saltfile /dev/sd0k /dev/svnd1c I think that does synchronous writes, even if you mount the svnd device async or softdep, which is why it is so slow. After losing a hard disc image _file_ to fsck on the filesy- stem containing the image file, I intended to use vnd(4) on a block device as well, but didnbt due to that very reason. I searched a little, but as Ibm not an experienced kernel developer did not find a solution to that problem. Talking to OpenBSD people I found out that itbs known. Canbt say moreb&
Re: OpenKSH?
Han Boetes dixit: >Thorsten Glaser already made a portable version of the OpenBSD >version of ksh, which compiles on lots of platforms and works >great. > > http://www.mirbsd.org/?mksh The canonical URI is http://mirbsd.de/mksh which redirects to the correct page, btw. See the website for information about mksh, the relationship to other korn shells (i.e. pdksh based), HTML and PDF versions of the manual page, etc. Tom McLaughlin dixit: >Sometime ago I created a FreeBSD port for OpenBSD's pdksh because of mksh is already in FreeBSDB. ports (and NetBSDB. pkgsrcB.), and I think they will be upgraded to mksh R29, actually tagged today, soon. I've got an OpenBSD port but somehow doubt someone will commit it. (If a committer thinks it's worth it please contact me privately.) bye, //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: cruxports for OpenBSD
Marc Balmer dixit: >what's next? The MirPorts Framework already runs on OpenBSD and Mac OSX, in addition to being the native package source for MirOS BSD. (It also works on In- terix aka MS Services for Unix 3.5, but that's currently not supported. It will come when time permits, though.) bye, //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Benny Siegert dixit: > A very pessimistic article but well worth a read: > > http://mail-index.netbsd.org/netbsd-users/2006/08/30/0016.html You could've just replied to it so that the References: header can be perused. I've changed this mail to reply to it for threading. Charles M. Hannum dixit: >Much of this early structure (CVS, web site, cabal, etc.) was copied >verbatim by other open source (this term not being in wide use yet) >projects -- even the form of the project name and the term "core". This >later became a kind of standard template for starting up an open source >project. > >Unfortunately, we made some mistakes here. As we've seen over the >years, one of the great successes of Linux was that it had a strong >leader, who set goals and directions, and was able to get people to do >what he wanted -- or find someone else to do it. On the other hand, the "bazaar" model of Linux leads to bad code and no well-defined APIs. While it's true that the "core-team" model _might_ benefit from a strong leadership, care should be taken to avoid Linux' "success" because it'll be its failure soon enough. (I mean, hey, 5 new kernels in 2 days, wtf?) Nick Guenther dixit: > Um. Wow. I think Theo wins. OpenBSD has had MicroBSD forked off twice, MirOS and ekkoBSD too. Travers Buda dixit: > As for Charles M. Hannum: fork! I don't think so, as long as he can improve the inner status of the NetBSD project. Forking is the solution if you're outside, want to improve and are ignored, or, if you're inside but don't see your interesting new ideas being accepted well or fitting within the project's overall policy (DragonFly). Andy Ball dixit: > suspend and resume work on my laptop. I know that work is being done > > on PowerNow! for AMD K6-2+, Athlon etc. Incidentally, Martin Vigiard's PowerNow work showed up in OpenBSD and FreeBSD. first, in NetBSD. last. @OpenBSD people: I did leave this mailing list, I'm just keeping the Cc: list. bye, //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Marc G. Fournier dixit: (Please don't keep individual persons in the Cc, only the lists, otherwise people will get the mails several times.) > Put together a *BSD "core" ... representative from each camp and try and steer > the *kernel* itself towards a more common BSD ... BSD is about an operating system, not about a kernel. //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Please don't Cc: people when you respond to mailing lists *sigh* Marc G. Fournier dixit: > for us, they need to write 1. Companies don't write drivers for BSD 2. Companies don't even release specs so that people can write drivers for BSD //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Marc G. Fournier dixit: > If the vendor is bought up, bankrupt, out of business, dead (like that person who ported g++ to Plan 9, whose window managers' copyright is now set in stone), etc... you're SOL9. //mirabile 9) wtf knows it -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Marc G. Fournier dixit: > I'm curious here, but why did the *kernel* diverge for each project? Because kernel, userland, ports and attitude come as a package, they cannot be separated, for together they are the operating system. //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
(When do you stop putting people into Cc instead of just the lists?) Marc G. Fournier dixit: > source code drivers provided by a > vendor, and supported by them Yeah, and what if the vendor goes out of business, is bought or simply bankrupt? You're pretty much SOL then. //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: The future of NetBSD
Marc G. Fournier dixit: > And what I'm learning with bsdstats.org is that there are more then just those > four ... GNU/kFreeBSD is reporting Now _that_'s funny ;) > are there any others? DragonFly DesktopBSD PC-BSD 4.3BSD-Quasijarus ekkoBSD (dead) MidnightBSD (nascent) MicroBSD (once dead, but apparently undead...) emBSD picoBSD nanoBSD The latter three are probably just stripped-down versions of the bigger ones. I think there are even more than these I listed above. The five big ones are DF/Free/Mir/Net/Open though. Daniel Seuffert (from AllBSD.de) today said he intends that your script to be installed AND ENABLED BY DEFAULT in every BSD release/snapshot, and he will definitively enable it in DesktopBSD 2.0. I will probably do the same for MirOS, as I currently don't have a way to even estimate the user base, other than bittorrent downloads, and it does not transmit any data qualifying as 'sensitive', uses almost no resources and is low-bandwidth. (We'll have to add some "jitter" to the execution time, though.) bye, //mirabile -- "Using Lynx is like wearing a really good pair of shades: cuts out the glare and harmful UV (ultra-vanity), and you feel so-o-o COOL." -- Henry Nelson, March 1999
Re: Passing random mac to lladdr in ifconfig
Nick Holmes dixit:
> found it in the Linux-based "Auditor" LiveCD) and that can generate a random
> MAC for the interface. Does anyone here have a script (Perl or otherwise)
> which
> can achieve the same in OpenBSD and they'd be willing to share?
Yuck, Perl, you are evil.
#!/bin/ksh
function a {
typeset -i16 "x= RANDOM % 256 + 256"
print -n ${x#16#1}
}
print "$(a):$(a):$(a):$(a):$(a):$(a)"
# ^-- gives you a random MAC address on stdout.
If you want to take care about the global/local bit and the
unicast/anycast bit, that's possible in ksh as well.
//mirabile
--
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only God invents algorithms, we merely copy them.
If you don't believe in God, just consider God as Nature if you won't deny
existence. -- Coywolf Qi Hunt
Re: mounting ext3fs via ext2fs
Frank Denis (Jedi/Sector One) dixit: > While running Linux, try ports/sysutils/e2fsprogs //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: mounting ext3fs via ext2fs
Frank Denis (Jedi/Sector One) dixit:
>On Mon, May 30, 2005 at 03:25:02PM +0000, Thorsten Glaser wrote:
>> ports/sysutils/e2fsprogs
>
> Sure, but to be fair, if he cares about his data, it's probably a bad idea
>to try a 3-years old version of e2fsprogs
I don't know which version is in OpenBSD ports at the moment,
but I've successfully fsck'd, resized and otherwise tortured
ext2fs/ext3fs with e2fsprogs 1.36 (on i386, admittedly).
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert).
;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus
dasr]
Re: perl -MCPAN checksum mismatch on anything
Wijnand Wiersma dixit: >Funny, I was trying to install Net::Jabber too. But all modules will >fail in the exact same way. I remember something about HASHFile() being broken some time ago; maybe that was the case? //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: Is /dev on mfs possible?
Jason Crawford dixit: >Yes, it is possible to have /dev on mfs, however that would mean you'd >have to run MAKEDEV on every boot after mounting the /dev memory file >system. Cannot -P do this? //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
speed of mac mini
Hello list, i think of buying a mac mini, but i don't know if a mac mini is fast enough. So i ask you: does anyone use an mac mini with gnome/kde or so? At the moment i have an dual-P3 and he's fast enough. Any coments, suggestions? Bye Thorsten
Re: speed of mac mini
Hello list, i will only do "normal" thinks:- some coding --> emacs/terminals/ddd - read www.openbsd.org --> firefox/dillo -read mails of misc@openbsd.org --> thunderbird - write some letters, do some calculations --> abiword/gnumeric - some statistik --> gnuplot - audio/video playing --> xmms/mplayer all with gnome or windowmaker. That's all. Bye Thorsten LiteStar numnums wrote: G'day, A friend of mine uses the mini for all of his foto processing with Photoshop and the like, whilst Illustrator and Safari are running. It seems fast enough. I've no idea what you want to really do with it (if it has a hard time with gnome/kde, that would be really bad, eh?), but for his needs it's fine. Cheers! On 6/16/05, Thorsten Johannvorderbrueggen <[EMAIL PROTECTED]> wrote: Hello list, i think of buying a mac mini, but i don't know if a mac mini is fast enough. So i ask you: does anyone use an mac mini with gnome/kde or so? At the moment i have an dual-P3 and he's fast enough. Any coments, suggestions? Bye Thorsten
[Fwd: Re: Theo gave an interview to Forbes Mag. about Linux]
Message-ID: <[EMAIL PROTECTED]> Date: Sat, 18 Jun 2005 11:50:27 +0200 From: Thorsten Johannvorderbrueggen <[EMAIL PROTECTED]> User-Agent: Mozilla Thunderbird 1.0.2 (X11/20050317) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Sparc Urani <[EMAIL PROTECTED]> Subject: Re: Theo gave an interview to Forbes Mag. about Linux References: <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> <[EMAIL PROTECTED]> In-Reply-To: <[EMAIL PROTECTED]> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Hello list, i have to comment this discussion. 1). The world is not black nor white. The world is colorful! 2). I'm a windows hater and a unix lover Sparc Urani wrote: >based on my experience, Linux is not a good start for the beginners. >for tough systems/network admins its bad. learning unix shall always >start with OpenBSD as they can see the difference when try using Linux >as their alternative OS. > > 3). Linux --> Linux is only the kernel and it's build tools. I think you mean a linux-distri. Ok, the most of them a really bad, but 4). have you ever used slackware? With slackware you have the total control of your system. It's cool (ok it's not so cool as OBSD)! >that was a very nice interview. Theo might encourage some other Linux >guys to start spending with OpenBSD for life ;-). > >Sparc >Computer Security Professional (CSP) has no reason to lie... to >cheat... to steal... nor tolerate among those who do. > >On 6/18/05, Richard Welty <[EMAIL PROTECTED]> wrote: > > >>On Fri, 17 Jun 2005 18:23:51 +0200 "J. Lievisse Adriaanse" >><[EMAIL PROTECTED]> wrote: >> >> >> >>>Not everybody there is happy about Theo's words...oh well, what gives ;-) >>> >>> 5). Theo says what he thinks, and thats really good! I like straightforward people! Theo says in the interview, what HE is thinking about linux and Co. In some case a think he's right. >>well, on the one hand, i largely agree with Theo, but on the other hand, >>Dan Lyons of Forbes has been on an anti-open source kick for some time. >> >>me, i'd be very wary of becoming a tool in someone else's agenda. >> >> 6). Your're right (in my opinion). The main problem are companies like MS and not Linux, OBSD or so! We shouln't start a religious war (that's one reason, why i don't "like" Stall)!! >>richard >>-- >>Richard Welty >>[EMAIL PROTECTED] >>Averill Park Networking >>Java, PHP, PostgreSQL, Unix, Linux, IP Network Engineering, Security >> "Well, if you're not going to expect unexpected flames, >> what's the point of going anywhere?" -- Truckle the Uncivil >> >> > > > > 7). If you want to learn kernel-developing, you have to write code. If you write code for the linux kernel, you have a good chance o see YOUR code in the kernel tree. If you have the time, you can get a maintainer! Is it that simple with OBSD? I don't think so! But to become a kernel-developer, you have to write "real" code! I in opinion it's easier to become a linux-kernel-developer than to become a OBSD-kernel-developer! --> That's the reason, why linux code is not as good a OBSD-code! 8). In some cases you have to use linux: I have a Benq-Joybook-8000 laptop. It's a laptop, and a laptop is mobile. So i need the battery state. On Slackware, no problem.OBSD only supports apm, no apci. Apm won't work on OBSD (apm won't work on linux too) with my laptop. (Why i use a Benq and not a IBM? Because i don't like the black boring .). To use my laptop as a laptop i need apci. At the moment i haven't got the knowlegde to write an apci implemtation for OBSD --> so i have to use slackware on it. 9). I think, that we (the people, who use OBSD) shouldn't flame linux user. Only the real developer (Theo, ..) could do so!! 10). OpenBSD is a really good OS!! So i trie to migrate every maschine from Linux to OBSD (like i did before, when i migrate from MS to Linux. Since 2001 i have not MS - Maschine). I hope you're not to angre with me! But i have to say this! Bye Thorsten
leap second support for (portable) OpenNTPD
Hello people,
I have adapted my leap second patch (which has been part of
OpenBSD's base system's rdate(8) for years) to (portable)
OpenNTPD and attached it below.
A patch against OpenBSD ntpd(8) has been sent to henning@
some time ago already.
The patch does not modify default behaviour - instead, like
with rdate, you have to specify the -c option if you are
running a "right" timezone (as in contrast to a "posix" one).
bye,
//mirabile
PS: Please direct replies to me too, I don't read misc@ any more.
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert). ;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus dasr]
= cutting here may damage your screen surface =
$MirBSD: Web/pub/openntpd-3.7p1-leapsecs.diff,v 1.1 2006/05/09 09:18:06 tg Exp $
diff -pruN openntpd-3.7p1.orig/LICENCE openntpd-3.7p1/LICENCE
--- openntpd-3.7p1.orig/LICENCE Sun Mar 13 13:02:56 2005
+++ openntpd-3.7p1/LICENCE Tue May 9 09:10:26 2006
@@ -111,4 +111,8 @@ daemon.c, sys-queue.h: The Regents of th
* SUCH DAMAGE.
*/
+The leap second support is an extension by Thorsten Glaser and
+is covered by the MirOS licence, but available under the ISC
+licence for use by the OpenBSD project in rdate(8) as well.
+
$Id: openntpd-3.7p1-leapsecs.diff,v 1.1 2006/05/09 09:18:06 tg Exp $
diff -pruN openntpd-3.7p1.orig/Makefile.in openntpd-3.7p1/Makefile.in
--- openntpd-3.7p1.orig/Makefile.in Wed Dec 15 01:59:53 2004
+++ openntpd-3.7p1/Makefile.in Tue May 9 09:09:19 2006
@@ -30,9 +30,9 @@ [EMAIL PROTECTED]@
PROG= ntpd
LIBCOMPAT=openbsd-compat/libopenbsd-compat.a
SRCS= ntpd.c buffer.c log.c imsg.c ntp.c ntp_msg.c config.c \
- server.c client.c util.c y.tab.c
+ server.c client.c util.c y.tab.c ntpleaps.c
OBJS= ntpd.o buffer.o log.o imsg.o ntp.o ntp_msg.o config.o \
- server.o client.o util.o y.tab.o
+ server.o client.o util.o y.tab.o ntpleaps.o
YFLAGS=
MAN= ntpd.8 ntpd.conf.5
@@ -45,7 +45,7 @@ $(LIBCOMPAT): config.h $(top_srcdir)/op
$(OBJS): config.h
ntpd: ntpd.o buffer.o log.o imsg.o ntp.o ntp_msg.o config.o \
- server.o client.o util.o y.tab.o $(LIBCOMPAT)
+ server.o client.o util.o y.tab.o ntpleaps.o $(LIBCOMPAT)
$(CC) $(CFLAGS) $(LDFLAGS) -o ntpd $(OBJS) $(LIBCOMPAT) $(LIBS)
.c.o:
diff -pruN openntpd-3.7p1.orig/contrib/redhat/openntpd.spec
openntpd-3.7p1/contrib/redhat/openntpd.spec
--- openntpd-3.7p1.orig/contrib/redhat/openntpd.specMon May 23 00:22:17 2005
+++ openntpd-3.7p1/contrib/redhat/openntpd.spec Tue May 9 09:15:21 2006
@@ -13,7 +13,8 @@ Requires: /sbin/chkconfig
#Patch1: openntpd-3.6p1-linux-adjtimex3.patch
%description
-NTP Time Synchronization Client - http://www.openntpd.org
+NTP Time Synchronisation Client - http://www.openntpd.org
+Enhanced by correct leap second support facility by MirOS
%prep
%setup -q -n %{name}-%{version}
diff -pruN openntpd-3.7p1.orig/ntpd.8 openntpd-3.7p1/ntpd.8
--- openntpd-3.7p1.orig/ntpd.8 Tue Dec 14 01:20:13 2004
+++ openntpd-3.7p1/ntpd.8 Tue May 9 09:08:29 2006
@@ -23,7 +23,7 @@
.Sh SYNOPSIS
.Nm ntpd
.Bk -words
-.Op Fl dSs
+.Op Fl cdSs
.Op Fl f Ar file
.Ek
.Sh DESCRIPTION
@@ -61,6 +61,12 @@ typically
.Pp
The options are as follows:
.Bl -tag -width "-f fileXXX"
+.It Fl c
+Correct leap seconds.
+Use this option if you are using one of the time zones from
+.Pa /usr/share/zoneinfo/right/
+(i\.e\. you are counting leap seconds in the kernel time,
+instead of following POSIX closely which denies their existence).
.It Fl d
Do not daemonize.
If this option is specified,
diff -pruN openntpd-3.7p1.orig/ntpd.c openntpd-3.7p1/ntpd.c
--- openntpd-3.7p1.orig/ntpd.c Sun Mar 13 12:41:24 2005
+++ openntpd-3.7p1/ntpd.c Tue May 9 09:08:29 2006
@@ -39,6 +39,7 @@ RCSID("$Release: OpenNTPD "OPENNTPD_VERS
#include
#include "ntpd.h"
+#include "ntpleaps.h"
void sighdlr(int);
__dead voidusage(void);
@@ -52,6 +53,7 @@ volatile sig_atomic_t quit = 0;
volatile sig_atomic_t reconfig = 0;
volatile sig_atomic_t sigchld = 0;
struct imsgbuf *ibuf;
+int corrleaps = 0;
void
sighdlr(int sig)
@@ -102,8 +104,11 @@ main(int argc, char *argv[])
log_init(1);/* log to stderr until daemonized */
res_init(); /* XXX */
- while ((ch = getopt(argc, argv, "df:sS")) != -1) {
+ while ((ch = getopt(argc, argv, "cdf:sS")) != -1) {
switch (ch) {
+ case 'c':
+ corrleaps = 1;
+ break;
case 'd':
conf.debug = 1;
break;
Re: FYI, 1and1 hosting fun (ip subnet zero)
(keeping cross-post of original poster because I
don't have a better idea)
Todd T. Fries dixit:
>The problem is that 1and1 hosting choses to have any root servers
>setup with `ip subnet zero'. That's a fancy way of stating that they
>expect systems to setup IP's with netmasks of 255.255.255.255 and then
>route to 10.255.255.1 `magically'.
Ah, they're doing it the same as Strato then.
Thanks to Matthias Kilian for discovering the following fix:
(hostname.fxp0)
inet 1.2.3.4 0x NONE
!route -n add -inet -host 5.6.7.8 -link -iface fxp0:
!route -n add -inet default 5.6.7.8
It does suck somehow, though. There are arp messages in dmesg.
bye,
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert). ;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus dasr]
Re: ntpd "dispatch_imsg in main: pipe closed"
Ray cyth.net> writes:
> Oct 11 09:29:24 sparky ntpd[30592]: dispatch_imsg in main: pipe closed
I've encountered this on several GNU/Linux boxen and tracked it down to
| listen on *
When I replace this by
| listen on ::
everything works fine.
On BSD systems, you have to use
| listen on ::
| listen on 0.0.0.0
instead because itojun sadly still doesn't believe in V4_MAPPED.
I also had a similar problem on BSD once, when setting up ntpd on a box
which had two interfaces connected to a bridge, and I was using the same
IP subnet (/64) on both physical interfaces (sis0, sis2 - Soekris net4801,
thanks Wim).
Probably this post will help people encountering the same problems in
the future (can you spell Redhat 5, Knoppix-HDINSTALL and OpenWRT?).
bye,
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine fC bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert). ;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmC$C ig ("Oooohhh"). [aus dasr]
Re: in-kernel pppoe
Can Erkin Acar dixit: >There should be no need for this in -current Thanks, will try. //mirabile -- > Hi, does anyone sell openbsd stickers by themselves and not packaged > with other products? No, the only way I've seen them sold is for $40 with a free OpenBSD CD. -- Haroon Khalid and Steve Shockley in gmane.os.openbsd.misc
Re: Hackathon 2005
[EMAIL PROTECTED] dixit: >Yes you're right but a "OpenHTTPD" with "OpenMOD_GZIP" and maybe IPv6 is >still missing. And maybe a OpenSQUID... :-))) >Just jokingoO(But httpD improvements would be realy cool. :) ) $ uname -a OpenBSD jiyu.gnook.org 3.7 GENERIC#97 i386 Doing a $ cvs -qz1 -d [EMAIL PROTECTED]:/cvs co -PA src/usr.sbin/httpd then removing "--enable-rule=EXPAT" from Makefile.bsd-wrapper at least gives a httpd mixture from OpenBSD and KAME which compiles (I didn't try if it works, but it should). It's got good IPv6 support, one of the modules in ports needs fixing (IIRC mod_perl), and vhosts are questionable (didn't work with wildcards, only per IP). Fup2p since it's going off-topic. //mirabile -- Hey, I just realized that OpenBSD CDs are $45. Any chance I could get you to update your sig? -- Steve Shockley after reading my previous signature
csh -> ksh script
Hi, any csh guru out here who wants to help rewriting src/usr.bin/vgrind/vgrind.sh in ksh? bye, //mirabile -- Hey, I just realized that OpenBSD CDs are $45. Any chance I could get you to update your sig? -- Steve Shockley after reading my previous signature
Re: IMAP servers
Gaby vanhegan dixit:
> What IMAP servers do people use for email access?
uw-imapd's imapd for imaps (port 993) access; sendmail with
uw-imapd's dmail/tmail instead of mail.local(8) for delivery
to MBX format mailboxes. Allows concurrent access.
uw-imaps allows reading arbitrary files on the server; I do
not consider this a problem since I can use chmod and chown.
The LP64 bugs in the code however are scary (but fixable).
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert).
;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus
dasr]
Re: IMAP servers
Gaby vanhegan dixit:
> we still have a lot of legacy mbox accounts.
uw-imapd can read (and write) a plethora of formats using
libc-client. Except Maildir, which they consider unsafe to
access and insane to implement.
In my ~/mail/ on the server, there are some mbox/unix
and some MBX folders in peaceful coexistence.
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert).
;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus
dasr]
Re: IMAP servers
Rod.. Whitworth dixit:
>You really believe those UW people really can consider something unsafe
It was considered so by the OpenBSD porter. UTSL.
>before they clean up their own exploit history? Insane? The sky is
>falling! I don't know about many IMAP servers but I know that UW-IMAP
>is considered less than favourably in many circles.
Prove an exploit.
>which shows you just what a nice guy Crispin is, eh? He really is the
>right guy to write RFCs, is he?
Now you're getting personal, eh? Sounds like FUD to me.
By the way, he DOES happen to have "invented" IMAP.
>(yeah, I know, DJB can be a bit of a Grumpy Old Man (to steal the title
>of a great TV series from GB) but I'd never take him for stupid. Mark,
>on the other hand lets his ego get in the way of reality and secure
>programming methods too, it seems to me.
You know, the pine suite (including imapd, mailutil etc.)
is not being written by one man.
Oh, and I've only replied because I think that it is
monoculture which sucks. I've never seen a tool handle
such a variety of both environments and mail formats
as mailutil (libc-client).
bye,
//mirabile
--
> emacs als auch vi zum Kotzen finde (joe rules) und pine f|r den einzig
> bedienbaren textmode-mailclient halte (und ich hab sie alle ausprobiert). ;)
Hallo, ich bin der Holger ("Hallo Holger!"), und ich bin ebenfalls
... pine-User, und das auch noch gewohnheitsmd_ig ("Oooohhh"). [aus dasr]
Re: IMAP servers
Rod.. Whitworth dixit: >I just wondered how somebody who has a string of alerts of his own can >regard anything else as unsafe. Eh, you are reading things I did not write, I think. >The Cheez Wizz 1998 alert. >The several buffer overflows detected in June 2001 I patched some of these; besides, don't we all have NXSTACK, NXHEAP, W^X and ProPolice now, these days? >The CRAM-MD5 problem of February THIS YEAR. Sure, but you would have to find someone to actually enable it. >But seeing you bring PINE into it: pine, uw-imapd and c-client all come in the same source file. >Is it true or false that PINE uses an undocumented extension (SCAN) tha I have no idea, but you might want to tcpdump. >UW-IMAP has and therefore fails with other IMAPDs that don't? It does not fail on other imapds. >And UW-IMAP cannot handle maildirs. Unless you use an "illegal" >non-Crispin addition. Eduardo Chappa, one of the head pine developers, maintains this patch and readily advocates it. For reasons to not integrate it see the other mail. And (except some proprietary databases) that's the only one missing which is in widespread use. > My signature is not overlong. It's exactly 4x78 (man McQuary). In contrast, *yours* is not correctly separated from the message body, *and* it is FIVE lines long, *and* YOU are replying to me personally even if I read the list. >Fortunately I think ALL IMAP sucks Why are you replying then, creating flamebait? >POP3 Oh yeah, no comment. >>From the land "down under": Australia. >Do we look from up over? > >Do NOT CC me - I am subscribed to the list. >Replies to the sender address will fail except from the list-server. > > > > > > > qed bye, //mirabile -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt
Re: ksh: bad number (with leading zeroes) should not work for 0 - 7
Abel Abraham Camarillo Ojeda verlet.org> writes: > Some of our shell scripts that work with dates and do something like: > > month=`date +%m` > something && month=$((month-1)) month=10#$(date +%m) bye, //mirabilos
Re: ksh: bad number (with leading zeroes) should not work for 0 - 7
Abel Abraham Camarillo Ojeda dixit: >I meant portable among other shells, or bourne shell... They donbt have $((b&)) either, sob& bye, //mirabilos -- Yay for having to rewrite other people's Bash scripts because bash suddenly stopped supporting the bash extensions they make use of -- Tonnerre Lombard in #nosec
Re: ksh: bad number (with leading zeroes) should not work for 0 - 7
Abel Abraham Camarillo Ojeda dixit: >On Thu, Sep 15, 2011 at 4:19 PM, Pascal Stumpf wrote: >>> month=10#$(date +%m) >> >> Is that a mkshism? No, that's proper Korn shell. >> The easiest solution here is: >> >> typeset -Z2 month That can break as soon as month is made an integer someplace down. Padding is meant for output, not input. >Yep, I did that in some scripts, just >don't know how portable it is... tg@stinky:~ $ /bin/ksh -c 'i=10#08; print $((i)) $(uname -a)' 8 SunOS stinky 5.8 Generic_117350-61 sun4u sparc SUNW,Sun-Fire-V210 Solaris So, portable across all Korn shells. bye, //mirabilos -- FWIW, I'm quite impressed with mksh interactively. I thought it was much *much* more bare bones. But it turns out it beats the living hell out of ksh93 in that respect. I'd even consider it for my daily use if I hadn't wasted half my life on my zsh setup. :-) -- Frank Terbeck in #!/bin/mksh
Re: linux default shell, how annoying
Followup-to: poster Marco Peereboom peereboom.us> writes: > All lies. Nothing to see there. Just someone who took some code and > pretends it's theirs. $ man mksh | fgrep -C3 recognises AUTHORS The MirBSD Korn Shell is developed by Thorsten Glaser and currently maintained as part of The MirOS Project. This shell is based upon the Public Domain Korn SHell. The developer of mksh recognises the efforts of the pdksh authors, who had dedicated their work into Public Domain, our users, and all contributors, such as the Debian and OpenBSD projects. See the documentation, CVS, and web site for details. No pretending there. Everything I wrote in the last nine or so years though is mine. Furthermore, the actual definition of Public Domain is that everyone can just b takeb it. I do admit I have started working on mksh a lot only after oksh cleanup and development had begun anew, but itbs gained a LOT of traction on its own. And happy users. (Android-x86 and Android 3, anyone?) > These so called changes are mostly build script > and other uninteresting updates. *laugh* Yes, the build script was a PITA (do you know how many broken shells are out there?), but itbs working and rather stable now b the rest of the shell is still being developed *very* actively. I just wish I had a part of the 1.7 Million US$ Ohloh says itbs worthb& ah yes, the ellipsis I so like to write; mksh handles UTF-8 just fine (in the Emacs editing mode and the scripting part, not the Vi editing mode, admittedly b patches welcome). > ksh works fine; it is maintained etc. Yeah, sure. x=$(case $foo in bar) echo baz ;; *) echo nope ;; esac) Just try this with oksh. Oh, and a gazillion other things. Ibm investing quite some work to align mksh with POSIX as well (mostly). Kevin Chadwick wrote: > I was going to replace the link from /bin/sh to dash with pdksh but I'm > guessing something somewhere will break!! Yes, it will. In Debian, neither pdksh nor ksh93 (!) are suitable for use as /bin/sh whereas mksh (although not necessarily the version in a stale release) is. This is off-topic for openbsd-misc, but considering that it had been mentioned already, this warning is probably justified. Still, my apologies for bringing in mentions of That Other OS. And as a parting note (yes, Ibll not even lurk here), Francois Pussault, youbre writing uninformed rubbish. (Although, NetBSDB. 1.5 and older cannot have their /bin/sh replaced by mksh, only their /bin/ksh, otherwise they wonbt boot, and repairing that on a DECstation/pmax took me a while. NetBSDB. 1.6 and up, and OpenBSD, however can. MidnightBSD uses mksh as its shell, even.)
Re: mounting ext3fs via ext2fs
Frank Denis (Jedi/Sector One) wrote: On Sun, May 29, 2005 at 11:00:34PM +0200, Rogier Krieger wrote: Feel free to correct me if I'm wrong, but as far as I know, ext3fs is not supported. ext3 is mostly ext2 with an extra inode to handle the journal. You can usually mount the partition as ext3 or ext2 without any special tweak. However on some distributions (at least Fedora it seems), directory hashing (htree) is enabled by default when partitions are formatted as ext3. And *BSD don't support htree yet. So maybe this is your showstopper. In order to this, my last formatted ext3-partition from a mandriva(cooker)-system, the OpenBSD 3.7-Machine could not handle with my external usb-hdd mounting it. Maybe something was wrong with the disk, after playing around with some mount_fooFS on the disk, there was no chance to read it under the usage of mandriva again either. Shure, no relevant data on it, just testing stuff, this was a big shame for me spending one day of my life :) Regards, Thorsten
Re: mounting ext3fs via ext2fs
Haluk Durmus wrote: [EMAIL PROTECTED]:~# /sbin/fsck.ext2 -y /dev/sda1 e2fsck 1.35 (28-Feb-2004) /dev/sda1: clean, 379761/9781248 files, 18008327/19537040 blocks No errors war found ! after that I did a fsck on OpenBSD [EMAIL PROTECTED]:~# fsck_ext2fs /dev/sd0i ** /dev/rsd0i ** File system is clean; not checking It was than posible to do: mount_ext2fs /dev/sd0i /mnt/usb/ Did some notes, thanks. It4s nice to know that for the next experiments with linux :)
Re: openbsd list fckery
JR Dalrymple wrote: The installer is awesome, it fits right beside the base kernel on ONE 1.44 floppy (speaking for i386). Yes, really thin. I just like that un-bloated one, no extras, just rude basement. That is enough for setting up a raw machine. More "tuning" could be done in later time in my opinion. Regards, Thorsten
Re: openbsd list fckery
Will H. Backman wrote: About a week ago, I was trying to upgrade my dual boot laptop to 3.7. I had to run the installer about 20 times to figure out my problem and correct it. In the process, I learned more about fdisk and disklabel than I had ever needed to before, and I count that as a good thing. It took no more than about 5 minutes each time to run the installer from scratch to completion in each case. Typing Ctrl-C and then "install" when you make a mistake isn't that difficult. I think the installer should be the last thing to go "user friendly". OpenBSD is not point and click. If you can figure out the installer, it means you actually read instructions. If you could install OpenBSD by just clicking "Next", you would be in for a rough ride after. Uh, I do not think so. The OpenBSD-installer is as easy as some gui-stuff, maybe much easier. Now blinking "Touch me"-Buttons, just straight work. I have done my first installation just without any documentation, it worked, but do not ask how the layout was :) In my opinion not the user-friendly task is important, but the easy and fast setup-possibility. Sometimes I neeed just a raw installation with an anonymous ftp for backup up some machine in trouble or for fileserving some data for a temporary issue, for that cases I love OpenBSD (as for some other reasons). Regards, Thorsten
OpenBSD 3.7 in a virtualpc-machine
Hi. Maybe the wrong side of the coin, but I think this list is right for that than [EMAIL PROTECTED] I tried to install OpenBSD on my iBooks "VirtualPC 6" for several times now, anytime just a "processor failure" when installing the base-packages. Maybe this will be fixed in version 7, a friend of mine will test this soon. Googeling around brought no enligthenment to me. So any suggestions by your person? Thanks and some special weekend regards, Thorsten
Re: OpenBSD 3.7 in a virtualpc-machine
Sevan / Venture37 wrote: brief info on issues with M$ VPC are mentioned here http://slagheap.net/openbsd/ best option is to use qemu or boches there are some qemu images here: http://www.freeoszoo.org/ Thanks for that!
Re: OpenBSD 3.7 in a virtualpc-machine
Hi Steven. steven n fettig wrote: I've been trying to get 3.7 to run in VirtualPC 7.0.1 on my PowerBook without any success. Right now, I'm actually working on getting 3.6 working so I can upgrade it to 3.7. The problems with 3.7 occur during formatting. I usually have 5 slices set up, / , swap, /tmp, /usr, /home and have tried various sizes. The biggest problem is trying to get a trouble report out of the install because when the machine crashes, I can't see exactly what is going on because it goes by so quickly. I'll let you know if I can get 3.6 to work... at least that would be a start. (Oh, I've tried it on some G5's I have, too, with no luck.) I4ll try that tomorow with my cd-sets of 3.4-3.6, maybe with good luck and will report the "sucess" here, if there would be one :) Regards, Thorsten
Re: OpenBSD 3.7 in a virtualpc-machine
Sevan / Venture37 wrote: brief info on issues with M$ VPC are mentioned here http://slagheap.net/openbsd/ best option is to use qemu or boches there are some qemu images here: http://www.freeoszoo.org/ Took that one and it runs famous for me. Thank you :)
greek website out of release-number
Hello, openbsd.org is just updated, the other language-sites are on the run, I think, but the el(greek)-site is just at 3.5. Maybe someone of the hellenic geeks is reading this one. Regards, Thorsten
Re: OpenBSD 3.7 Torrents are now available
andrew fresh wrote: You can get OpenBSD 3.7 from the torrent site here: i386: http://openbsd.somedomain.net/torrents/OpenBSD_3_7_i386-2005-05-19-2115.torrent Great speed, great service, thank you Andrew! Regards, Thorsten
