Re: Sun Fire T2000 available

[sarlok]

On 05.04.2016 06:19, Lukas Kaminski wrote:

Hello List(s),

If someone needs a T2000, it's available for free in Duisburg, 
Germany.


I don't have time for checking the hardware, i think there are minor
issues with the fans, but it should be functional.

No HDDs included. We have some Fire X4200 and Fire X4100, too. I
think they share some spare parts.


I also have a T2000 in BC, Canada, free to a good home.
8-core 1.2ghz, 32gb ram - last I checked.   I *think* it still has 
drives, but it's not within my immediate line of sight at this time.

Re: Sun Fire T2000 available

[Gareth Nelson]

Would you be willing to ship to the UK?


On Tue, Apr 5, 2016 at 2:19 PM, Lukas Kaminski 
wrote:

> Hello List(s),
>
> If someone needs a T2000, it's available for free in Duisburg, Germany.
>
> I don't have time for checking the hardware, i think there are minor
> issues with the fans, but it should be functional.
>
> No HDDs included. We have some Fire X4200 and Fire X4100, too. I think
> they share some spare parts.
>
>
>
> --
> Lukas Kaminski | +49/203/30597-953 | lukas.kamin...@krankikom.de
> ___
>
> Krankikom Alexander Kranki Kommunikation GmbH
> Schifferstraße 200  |  D-47059 Duisburg  |  Germany
> AG Duisburg HRB 6669  |  Ust.-Ident.-Nr. DE811853024
> Geschäftsführer: Alexander Kranki, Arne Rudolph, Holger Ruhfus
>
> www.krankikom.de

Re: sendmail mx question

[Claus Assmann]

> so the real smtp has the lower number but higher priority but like I said my
> sendmail always ends up with shit.example.not.nz.

What does "sendmail always ends up with shit.example.not.nz." mean?

Of course sendmail tries the secondary MX after trying the main MX.

Still no real data/logs/output of a verbose queue run for the domain/...

-- 
Note: I will most likely not reply to mails that
- use HTML
- top post
- quote more than necessary

Re: sendmail mx question

[Craig Skinner]

Hi Markus,

On 2016-04-05 Tue 16:39 PM |, Markus Rosjat wrote:
> 
> no the real setup is the other way arround
> 

 5 smtp.example.not.nz. # <<--- real server
10 shit.example.not.nz. # <<--- always defering server

Their setup is fine then, a classic highlisting arrangement.

> 
> so the real smtp has the lower number but higher priority but like I said my
> sendmail always ends up with shit.example.not.nz.
> 

Can you ping & connect to their primary MX?

$ nc smtp.example.not.nz 25

It could be a routing or packet filtering fault somewhere.

Cheers.
-- 
Arithmetic is being able to count up to 23
without taking off your shoes or pants.

Sun Fire T2000 available

[Lukas Kaminski]

Hello List(s),

If someone needs a T2000, it's available for free in Duisburg, Germany.

I don't have time for checking the hardware, i think there are minor 
issues with the fans, but it should be functional.


No HDDs included. We have some Fire X4200 and Fire X4100, too. I think 
they share some spare parts.




--
Lukas Kaminski | +49/203/30597-953 | lukas.kamin...@krankikom.de
___

Krankikom Alexander Kranki Kommunikation GmbH
Schifferstraße 200  |  D-47059 Duisburg  |  Germany
AG Duisburg HRB 6669  |  Ust.-Ident.-Nr. DE811853024
Geschäftsführer: Alexander Kranki, Arne Rudolph, Holger Ruhfus

www.krankikom.de

Re: sendmail mx question

[Markus Rosjat]

hi there,

no the real setup is the other way arround

1 shit.example.not.nz.  10 # <<--- always defering server
2 smtp.example.not.nz.  5 # <<--- real server

so the real smtp has the lower number but higher priority but like I 
said my sendmail always ends up with shit.example.not.nz.


sendmail 8.14.1  ... yeah it's old, yeah need to update and yes no time 
for now :(


Am 05.04.2016 um 16:01 schrieb Claus Assmann:

On Tue, Apr 05, 2016, Craig Skinner wrote:


1 shit.example.not.nz. # <<--- always defering server
2 smtp.example.not.nz. # <<--- real server



Your server connects to 'shit.example.not.nz', which defers the mail,
telling your server to try again later. So,. your server tries again
later!!! It has no need to try the backup MX machine, it got told to try


Really?

Which MTA does that?
sendmail 8.x?


Well, it would be nice if the OP provides some real info, but since
he didn't do that, I didn't reply...



--
Markus Rosjatfon: +49 351 8107223mail: ros...@ghweb.de

G+H Webservice GbR Gorzolla, Herrmann
Königsbrücker Str. 70, 01099 Dresden

http://www.ghweb.de
fon: +49 351 8107220   fax: +49 351 8107227

Bitte prüfen Sie, ob diese Mail wirklich ausgedruckt werden muss! Before 
you print it, think about your responsibility and commitment to the 
ENVIRONMENT

Re: sendmail mx question

[Claus Assmann]

On Tue, Apr 05, 2016, Craig Skinner wrote:

> 1 shit.example.not.nz. # <<--- always defering server
> 2 smtp.example.not.nz. # <<--- real server

> Your server connects to 'shit.example.not.nz', which defers the mail,
> telling your server to try again later. So,. your server tries again
> later!!! It has no need to try the backup MX machine, it got told to try

Really?

Which MTA does that?
sendmail 8.x?


Well, it would be nice if the OP provides some real info, but since
he didn't do that, I didn't reply...

Re: sendmail mx question

[Craig Skinner]

Hi Markus,

On 2016-04-05 Tue 14:22 PM |, Markus Rosjat wrote:
> 
> yeah my server does retries but always ends up on the mailserver with the
> lower priority :(
> 

That is the correct behaviour.

Without the domain name, I'm guessing with English words what you mean;-

Pretending their broken setup is like this:

$ dig example.not.nz MX +short
1 shit.example.not.nz. # <<--- always defering server
2 smtp.example.not.nz. # <<--- real server

Your server looks up the DNS records and finds the primary MX machine is
'shit.example.not.nz'.

Your server connects to 'shit.example.not.nz', which defers the mail,
telling your server to try again later. So,. your server tries again
later!!! It has no need to try the backup MX machine, it got told to try
again later.

If I read your words right, their setup is fucked.

They've confused nolisting with highlisting, and got them arse about.
-- 
I think there's a world market for about five computers.
-- attr. Thomas J. Watson (Chairman of the Board, IBM), 1943

Re: Crash after updating to today's snapshot

[Luke Tidd]

Working again as expected with todays snap. For posterity the logs
provided were from a t430s not an x230.

On 4/2/16, Luke Tidd  wrote:
> Thanks for the info everyone. I was going to try one more snap before
> filing a bug but I can do that next time if it's appropriate.
> On Apr 2, 2016 1:37 AM, "Theo Buehler"  wrote:
>
>> On Sat, Apr 02, 2016 at 12:24:13AM -0400, Luke Tidd wrote:
>> > Machine is a Thinkpad x230. First crash after an update.
>>
>> Thanks for the report. Pleasae report bugs to the bugs@ mailing list
>> because not everybody reads misc@. The change that led to this has been
>> reverted:
>>
>> https://marc.info/?l=openbsd-bugs=145954247112249=2
>>
>>
>


-- 
Luke Tidd
Google
7055 Pleasant Dr
Austell, GA 30168
404-939-0306

Re: sendmail mx question

[Markus Rosjat]

Hi peter,

yeah my server does retries but always ends up on the mailserver with 
the lower priority :(


Am 05.04.2016 um 12:44 schrieb Peter N. M. Hansteen:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 04/05/16 11:55, Markus Rosjat wrote:


I have a mail to deliver to a domain that has two mx record but the
2nd record isn't really a mx (so I got told but the need to keep
that for some reasons).


I would question their competence right there. If it's an MX, it needs
to actually handle mail. (Ok, there is the slightly perverse case
where the only thing actually listening on port 25 is spamd(8), but stil
l)


So far so good the priority on the 2nd mx is also lower so my
sendmail daemon should figure to send to the server with the
highest priority but it does not.

So here is what I have done to get my server to try to deliver the
mail to the right server:

- restarted sendmail - restarted named

Is there something I can do still to get my sendmail to deliver to
the mailsserver with the higher priority?


As long as your side keeps retrying, it will eventually manage to
deliver to the one that actually accepts mail. But please tell the
other side to just fix their setup.


I talked to the guy on the other "Mailserver" side and he says we
are the only one who have a problem since they deployed that setup
in 2014.


The only one seeing the problem? No. The only one looking into the
problem and telling them about it? Quite possible. As in, most people
wouldn't know what to look for, and in most cases mail would
eventually be delivered anyway, but delivery would not happen immediatel
y.

The only advice I can offer is to check that your side has a
reasonable retry period (IIRC default setups for all the MTAs on
OpenBSD come with reasonable settings, but do check), and tell the
other side that for their own sake they need to fix their setup.

- --
Peter N. M. Hansteen, member of the first RFC 1149 implementation team
http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/
"Remember to set the evil bit on all malicious network traffic"
delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.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=P8Om
-END PGP SIGNATURE-



--
Markus Rosjatfon: +49 351 8107223mail: ros...@ghweb.de

G+H Webservice GbR Gorzolla, Herrmann
Königsbrücker Str. 70, 01099 Dresden

http://www.ghweb.de
fon: +49 351 8107220   fax: +49 351 8107227

Bitte prüfen Sie, ob diese Mail wirklich ausgedruckt werden muss! Before 
you print it, think about your responsibility and commitment to the 
ENVIRONMENT

Re: sendmail mx question

[Peter N. M. Hansteen]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256

On 04/05/16 11:55, Markus Rosjat wrote:

> I have a mail to deliver to a domain that has two mx record but the
> 2nd record isn't really a mx (so I got told but the need to keep
> that for some reasons).

I would question their competence right there. If it's an MX, it needs
to actually handle mail. (Ok, there is the slightly perverse case
where the only thing actually listening on port 25 is spamd(8), but stil
l)

> So far so good the priority on the 2nd mx is also lower so my
> sendmail daemon should figure to send to the server with the 
> highest priority but it does not.
> 
> So here is what I have done to get my server to try to deliver the
> mail to the right server:
> 
> - restarted sendmail - restarted named
> 
> Is there something I can do still to get my sendmail to deliver to
> the mailsserver with the higher priority?

As long as your side keeps retrying, it will eventually manage to
deliver to the one that actually accepts mail. But please tell the
other side to just fix their setup.

> I talked to the guy on the other "Mailserver" side and he says we
> are the only one who have a problem since they deployed that setup
> in 2014.

The only one seeing the problem? No. The only one looking into the
problem and telling them about it? Quite possible. As in, most people
wouldn't know what to look for, and in most cases mail would
eventually be delivered anyway, but delivery would not happen immediatel
y.

The only advice I can offer is to check that your side has a
reasonable retry period (IIRC default setups for all the MTAs on
OpenBSD come with reasonable settings, but do check), and tell the
other side that for their own sake they need to fix their setup.

- -- 
Peter N. M. Hansteen, member of the first RFC 1149 implementation team
http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/
"Remember to set the evil bit on all malicious network traffic"
delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.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=P8Om
-END PGP SIGNATURE-

Re: doas.conf cmd with argument(s)

[Tor Houghton]

On Mon, Apr 04, 2016 at 08:08:19AM +0100, Jason McIntyre wrote:
> 
> it is a bit inconsistent, yes.
> 
> it is very much less readable with a line break. you could remove the
> offset, but that doesn;t look great either. you could specify a smaller
> offset and juggle the actual text a bit.
> 
> the text is clear enough. i don;t really have a problem with what's
> there. at least, i don;t see an easy way to both make the change you're
> requesting and have it still read nicely.
> 

I made a "patch comment" in my reply to Guenther; perhaps all you need is to
add another example with args, e.g. under the line with the "tedu" user?

Tor

sendmail mx question

[Markus Rosjat]

Hi there,

this more a mail about confirming the problem isn't on my site here.

So that's the case:

I have a mail to deliver to a domain that has two mx record but the 2nd 
record isn't really a mx (so I got told but the need to keep that for 
some reasons). So far so good the priority on the 2nd mx is also lower 
so my sendmail daemon should figure to send to the server with the 
highest priority but it does not.


So here is what I have done to get my server to try to deliver the mail 
to the right server:


 - restarted sendmail
 - restarted named

Is there something I can do still to get my sendmail to deliver to the 
mailsserver with the higher priority?


I talked to the guy on the other "Mailserver" side and he says we are 
the only one who have a problem since they deployed that setup in 2014.
But on my side it's the same, this is the only domain where this is 
happening at all.


I'm greatful for any advice

regards

--
Markus Rosjatfon: +49 351 8107223mail: ros...@ghweb.de

G+H Webservice GbR Gorzolla, Herrmann
Königsbrücker Str. 70, 01099 Dresden

http://www.ghweb.de
fon: +49 351 8107220   fax: +49 351 8107227

Bitte prüfen Sie, ob diese Mail wirklich ausgedruckt werden muss! Before 
you print it, think about your responsibility and commitment to the 
ENVIRONMENT

Re: OT: True hardware UNIX terminal

[John Long]

On Mon, Apr 04, 2016 at 04:40:20PM -0600, Nick Bender wrote:

> I wonder if any FORTRAN programmers out there remember the trick of putting
> line numbers after column 72 so the card sort could sort your program back
> into order when you dropped your card deck?

This was not limited to FORTRAN. We always used sequence numbers in 73-80
for exactly this reason. To this day the MVS (z/OS) editor will place them
for you in those colums automatically when you say "num on" or "renum". This
works for assembler, COBOL, and PL/I too.

And yeah you won't understand unless you ever dropped a box of cards or saw
the look of horror on somebody else's face when he did.

> Finally I'll never get back the three days I spent finding the zero I had
> mistakenly put in place of the letter O in my JCL at the front of the card
> deck. Good times...

We're still keeping the faith!

/jl

-- 
ASCII ribbon campaign ( ) Powered by Lemote Fuloong
 against HTML e-mail   X  Loongson MIPS and OpenBSD
   and proprietary/ \http://www.mutt.org
 attachments /   \  Code Blue or Go Home!
 Encrypted email preferred  PGP Key 2048R/DA65BC04 

Running OpenBSD 5.9 from libreboot without proprietary blobs is possible. Detailed instructions inside.

[silent_wanderer]

Libreboot is a free software BIOS replacement, see https://libreboot.org
for details. It is a distribution of Coreboot without
proprietary blobs, including CPU microcode.
All tests were performed with Thinkpad X200, but it should work
for most or all libreboot and autoboot machines.

Since 5.9, OpenBSD supports EFI boot mode, which means that it also
have had to support framebuffer out of the box, so lack of
proprietary VGA BIOS blob is no longer a problem and you can boot it
with unmodified Libreboot binary release 20150518. Earlier releases
won't probably work because of lack of some GRUB modules.

In order to install OpenBSD on such a machine you will need some
additional preparations, since regular install59.fs won't work
because bsd.rd doesn't have a framebuffer console.
I can't tell what happens exactly, but the device reboots by itself
if I try to load it.

However, bsd and bsd.mp kernels work just fine (I do not mean
/bsd, it's some gzip archive, I mean /5.9/$arch/bsd)

So if you want to install OpenBSD, you have to either
a) build bsd.rd kernel with framebuffer support, or
b) use regular kernel in conjunction with bsd.rd content, or
c) put the hard drive into another machine and install from there.

I describe the option b:

1) Find another computer with regular BIOS or UEFI, which can boot from
OpenBSD installation media.
QEMU or other virtual machine will be fine too.

2) Boot it from OpenBSD 5.9 install59.iso or install59.fs or miniroot.fs
You can use anything with a valid bsd.rd kernel.

3) Make a new partition for OpenBSD installation rootfs on a removable
drive, or use an existing one. It should be at least 100 megabytes
if you plan to use cp or something around 10m if you use something else,
and preserve hardlinks.
You can try using existing partition from install59.fs, removing files
outside of 5.9 directory if necessary, or make a new partition on same
removable media, or even prepare another one.
You might need to issue something like "cd /dev; sh MAKEDEV sd1"
to access your removable drive from the shell and maybe
you'd have to run fdisk and/or disklabel to prepare the partition.

4) Make new OpenBSD UFS on this partition (refer to newfs docs).
Then mount it under /mnt2 or make a new directory for it.

5) Copy everything from / to newly mounted /mnt2 partition.
It's a tricky part: according to OpenBSD man cp, it cannot preserve 
hardlinks.

It also doesn't seem to have guards against copying a directory into
subdirectory of itself, unlike GNU cp, thus use grep -v to prevent it.
Since all executables on the ram disk are hardlinked, it will take
around 89MiB of disk space instead of just 2-3 like it is supposed to:

cp -Rpi `ls -A / | grep -v mnt2` /mnt2

You can also use tar (or pax or cpio or...), and it should preserve
hardlinks. Untested, correct the command if it doesn't work for you:

tar cXpvf - / | tar -f - -C /mnt2

6) Make sure that you have copied everything, including /.profile
and make an empty directory for mnt2:

ls -la /mnt2
mkdir /mnt2/mnt2

7) If you are using a separate media, you might want to copy entire
5.9/$ARCH/ directory onto your partition too. If you are using the same
media, you can leave it on their own partition. Make sure you have bsd
or bsd.mp in a known location, accessible to libreboot.

8) umount /mnt2. Now you are ready to install it on a libreboot laptop.
Don't forget the name of partition where it was (something like sd1k),
you will need it later. The digit could change between different
machines, but the only flash drive is sd1 on x200

9) There is a kopenbsd command in GRUB2, which can load OpenBSD kernels
 directly. It might some disadvantages compared to native OpenBSD
bootloader, but they do not affect user experience, as far as I can
see. Go to GRUB2 console and issue commands like this:

kopenbsd -r sd1a (usb0,openbsd1)/5.9/amd64/bsd.mp
boot

-r option gives the name of partition which you prepared earlier,
the second option has full path to the kernel you want to load,
including partition name.

The system should load. If you have set -r option correctly,
you will get to regular OpenBSD installer prompt.

If you need help with OpenBSD command line, refer to 
http://man.openbsd.org/

or other OpenBSD documentation. Installation media has no man pages
and some options are different from GNU. For example:
cp needs -R, not -r and ls dir -l won't work, but ls -l dir will

Perhaps, there is a way to extract disk image from bsd.rd and put it
onto a partition directly, but I do not know whenever it has valid ufs
format and what is its offset.

If you want, you can boot NetBSD from libreboot GRUB prompt too.
In this case, regular NetBSD install media works fine without changes.

I do not mind if you integrate this text in modified or unmodified
form into a wiki or other documentation source, with or without
mentioning the source and use and share it however you like.

Re: muting keyboard bell broken on amd64 -current?

[lists]

On Mon, Apr 04, 2016 at 12:31:26AM +, li...@ggp2.com wrote:
> > Can you reproduce it with an recent original snapshot kernel?
> 
> My kernel is newer than the snapshot on the mirror I checked an hour or
> so ago, so I'll wait until a new one hits and test.

Using the latest snapshot the problem is fixed.  I'm not sure if it was
the reboot (could have been the Thinkpad firmware?) or the snapshot, but
everything is peachy now :)