Re: switching to DUIDs (and back)

[Alan Corey]

>On Sun, Oct 29, 2017 at 11:30:51AM -0400, Josh Grosse wrote:
>> Basically, replace "wd0" with the drive's DUID.

>I'm wrong, of course.  Replace "/dev/wd0" with the drive's DUID,
>then append "." followed by the partition.

Thank you, I was thinking every partition has a UUID and I needed to
find and use those.  But even in Linux it's apparently only devices
that have UUIDs.  They're almost like DOS/Windows drive serial
numbers, but those are generated when you format a partition and only
apply to the partition.  Yes, I was looking at man pages but what I
was looking for doesn't exist.

switching to DUIDs (and back)

[Alan Corey]

I thought this was documented somewhere but I'm not finding it in Googling.

I don't really like DUIDs, but I want to stick in a second drive
controller temporarily to recover data off some old hard drives.
Which means /dev/wd0a etc is going to change since the added
controller (in a PCI slot) becomes primary.  I'm perfectly comfortable
manually mounting and unmounting the old drives, I don't want to add
them to an fstab.

I want to replace my fstab with one that accesses my current
partitions using DUIDs.  Disklabel shows me a DUID for the drive, how
do I set up individual partitions?  Or is there already a DUID (or
UUID) for each partition that I need to find and use?

The main thing right now is to change fstab so it boots back up
smoothly with DUIDs.  I'll comment out my current entries and put the
DUID ones below.  And yes, I already made a backup copy.  All IDE,
can't afford SCSI.

Re: Deleted everything in /

[Alan Corey]

Hah!  I got an err M again 3 years later.  I was running multiboot
whereby you do a dd from wd0 into a file.  So I booted from an install
disk, shelled out, did fsck on the hard drive partitions then used dd
to copy back the bootsector file.  It worked.

I just did:
dd of=/dev/wd0 if=obsd.bin
and rebooted, everything came back.



On 3/6/14, Alan Corey  wrote:
> Got it.  Thanks.  I burned a 5.2 install and used the ramdisk
> /usr/mdec/installboot from that. I don't have 5.5 and it would take
> weeks by modem to get it.
>
> On 3/6/14, Chris Cappuccio  wrote:
>> Alan Corey [alan01...@gmail.com] wrote:
>>> I'm at 5.2. Booting from a 5.4 install image I mounted my / as /mnt
>>> then my /usr as /mnt2. Then I did:
>>>
>>> /mnt2/mdec/installboot -n -v /mnt/boot /mnt2/mdec/biosboot /dev/wd0c
>>> and get: Bad system call
>>>
>>> There's a /mnt/boot in place copied from /mnt2/mdec
>>>
>>
>> You need to run installboot from /usr/mdec (or /usr/sbin on 5.5)
>> on the install image ramdisk, not the 5.2 host.
>>
>> And you really need to use the installer and let it do all
>> this for you, or else you should read the install/upgrade scripts
>> and figure out the stuffs.
>>
>
>
> --
> Credit is the root of all evil.  - AB1JX
>


-- 
-
No, I won't  call it "climate change", do you have a "reality problem"? - AB1JX
Impeach  Impeach  Impeach  Impeach  Impeach  Impeach  Impeach  Impeach

Re: Content filtering through pf?

[Alan Corey]

I'm looking at privoxy although I'm not sure it's more appropriate
than squid.  I'm hoping to run this on a Raspberry Pi or Zero so it'll
most likely be under Raspbian.

Right now I use the standard Android "Portable Wi-Fi hotspot" in the
phone.  I run it open (no password) because I'm in a very rural area
and don't need them.  I want to tether by USB to some box I have
better control over.  Then set up an AP on that which effectively
replaces the one in the phone (with a gain antenna to boot).

One thing that just occurred to me is that I can set up in the AP's
dhcpd.conf the MAC addresses of my trusted machines so they will
bypass the proxy entirely, or maybe use a different one just for ad
blocking.  And hopefully prioritize bandwidth usage, setting my
trusted MAC addresses with the highest priority.  Everything else by
default will get fed through the proxy.

I'd rather not rely on mime types because I don't know that mime is
even used by proprietary things like the Washington Post, BBC, NPR,
etc. Android news clients.  They could be specialized web browsers, or
they could work with pure binary data.  There's no reason for them to
be compatible with the rest of the world since they run the servers
and write the clients.  I suspect they were lazier than that though.

-- 
-
No, I won't  call it "climate change", do you have a "reality problem"? - AB1JX
Impeach  Impeach  Impeach  Impeach  Impeach  Impeach  Impeach  Impeach

Content filtering through pf?

[Alan Corey]

I'm wondering if it's possible to do content filtering in a firewall.
Maybe with something that cooperates with pf.  I'm on a very limited
(5 GB/month) metered internet connection through a cell phone and I'm
not the only user when I have it shared over wifi.  I'd like to block
video because it's an incredible waste.  Problematic clients are
Android/Kindle.  User competence in not clicking where they shouldn't
is sometimes an issue.

I can see this happening if there's a file size available during
transfers, if the size is under a certain threshold value it just
passes without interference, over a certain size the first few bytes
of the file get checked.  If it fails the check that exact URL to the
file would get blacklisted for maybe 24 hours.  I've noticed watching
random transfers with wget that in some cases it knows the file size
from somewhere and sometimes not.  Presumably there's no size
available on streaming video so just block it.

There seems to be an abundance of video in advertising in apps but
also in news apps there's a mix of video and text stories.  Most of
the world assumes bandwidth is free and fast.  Some videos are bigger
than entire operating systems, and most are fairly pointless.  If the
transfer is happening over an ssl connection maybe not much can be
done since from the firewall's perspective it's just encrypted data,
essentially inside a tunnel.

Re: Browser is getting slower?

[Alan Corey]

I don't use Chromium (don't tell me I don't need a menu) but in
Firefox the user profile gets clogged up with cruft about once a year.
A quick test is to just make a new profile and see if it's faster.
Then copy over your bookmarks and gradually reestablish your cookies
which keep you logged into sites between sessions by manually logging
into each one.

You can get to Firefox's profile manager with
firefox -ProfileManager
at a command line.  It's probably in the GUI somewhere.  I use
Chromium about once a year, don't know much about it.

-- 
Credit is the root of all evil.  - AB1JX

Re: Why on earth would online voting be insecure?

[Alan Corey]

OK, it's relevant to OpenBSD because I wouldn't consider anything else
safe enough to run on the servers.  Not that I'm in a position to do
any of it.  The servers could even be run from custom official live
CDs so they were harder to tamper with, with maybe a RAM drive for
speed.

There seems to be a conflict between having anonymous votes and having
something similar to paper ballots that can be recounted.  So let
authentication, identification, etc. be handled by one machine and
stored in one database then the transaction is handed over to another
machine which stores the votes.  That could be something simple like a
tab-delimited file which could be counted by hand, one line per voter.
The file could be only writeable by the owner. The same person can't
vote twice because the first machine wouldn't allow them in a second
time.

I'm assuming there's physical security over the server room, if that
was compromised all bets are off.  When I last voted I verbally
identified myself to one person who handed me my ballot, which I
checked off in pencil, then identified myself to another worker who
cranked my ballot into a simple counting machine about 40 years old.
Yes, if one person got access to the files in seclusion they could
alter something assuming they were root, that would have to be as
impossible as erasing the pencil marks on the ballots and changing
them.  I assume there are always multiple scrupulous workers present.

It doesn't have to be an SSN, a driver's license number would work as
well.  Some long number known mostly only to the voter and to the
government which doesn't arrive by the same mailing as the key the
town sends  Somewhat analogous to a public key, with the private key
being the number the town mails each voter for each election.

Laziness isn't the only reason to do this, I would hope to expand it
to maybe a weekly vote on things that are put to the House and Senate
so there's direct input from voters instead of only electing people
who do their voting.  There probably wouldn't be a lot of interest but
being able to provide feedback to elected representatives could be
useful, conversely there would be statistics on what percentage of the
time they voted as the public wanted.

Instead of voting with a web browser, there might be some security to
be gained by using a dedicated client.  Or voting from something like
an Android phone (I have no experience with IOS).  Android security
seems almost excessive.  Incorporating the phone numbers on each end
could be useful although not to be trusted as identification by
itself.  An app could connect to a phone number and load a ballot,
fill it out offline, then dial another number to submit it in
milliseconds which lessens the load on the server.  For that matter
you could produce live CDs to be booted and used only for voting, any
operating system you want.

I think bouncing ideas off a community of knowledgeable computer
hobbyists and professionals is a useful thing to do.  I became an
OpenBSD user about 2001 because I inherited a Linux box at my job that
had been root kitted and I needed something more secure, it's still my
first choice.  I later firewalled the entire office through another
OpenBSD box, it worked very well.  So yes, security in academia where
student records were concerned, we had thousands of transcripts.
-- 
Credit is the root of all evil.  - AB1JX

Why on earth would online voting be insecure?

[Alan Corey]

This sounds like heel-dragging to me, or they're trying to do it under
Windows or something:
https://www.washingtonpost.com/news/post-nation/wp/2016/05/17/more-than-30-states-offer-online-voting-but-experts-warn-it-isnt-secure/

It seems simple to me, you use firewalls and only make the results
writeable by the process that should be writing to it, probably
nothing needs to have read access in the short term.  As far as
security after the election, mount the servers in a Brinks truck or
something, it just sounds like a ludicrous excuse.

Something like: for each election the town government mails you a
random number that's your key to vote that election. You go to a
website and put in your town, name, SSN, and the key. If somebody
steals the mail they won't have your SSN. If Russian hackers or
whoever tries to impersonate you online they won't have the key. It's
bringing those 2 pieces of information plus your name and town
together that makes it secure. Just guessing. Did I overlook anything?

-- 
Credit is the root of all evil.  - AB1JX

Re: 2 files, same name, same dir

[Alan Corey]

I can probably blame it on the Joe editor, which leaves trailing
spaces on lines in files too.  I wrote a program once to strip them
off, probably doesn't bother anybody else.  When you save a file it
prompts with a (previous) filename with the cursor at the end, bumping
the spacebar then would do it, I've hit other characters by accident
and had those end up in filenames.  I've been using Joe for ~20 years.

But a trailing space on a filename, after the extension, is pointless
to keep.  Probably featureitis to build in looking for them though.

On 10/22/16, Alan Corey  wrote:
> Bingo, a trailing space.  I would have thought that would get trimmed
> off somehow.  I don't normally put spaces in filenames, I must have
> fat-fingered something.  But I see no reason to keep trailing spaces.
>
> -rw-r--r--  1 alan  wheel25549 Oct 22 10:55 pi.c\$
> -rw-r--r--  1 root  wheel25152 Oct 22 00:07 pi.c \$

Re: 2 files, same name, same dir

[Alan Corey]

Bingo, a trailing space.  I would have thought that would get trimmed
off somehow.  I don't normally put spaces in filenames, I must have
fat-fingered something.  But I see no reason to keep trailing spaces.

-rw-r--r--  1 alan  wheel25549 Oct 22 10:55 pi.c\$
-rw-r--r--  1 root  wheel25152 Oct 22 00:07 pi.c \$

> Likely one of them has  space or so in it's name. Use
>
> $ ls -l | vis -l
>
> or some variation.
>
>   -Otto
>


-- 
Credit is the root of all evil.  - AB1JX

2 files, same name, same dir

[Alan Corey]

This is in 5.7 but I don't understand how it could happen (and I never
noticed it before).  Two files in the same directory with the same
name, different owners.  Last night I was noticing that my edits to my
program sometimes weren't taking effect, this seems to be why.  Ideas?
 I can delete the wrong one from mc probably, I just wanted to mention
it.

-rw-r--r--1 alan  wheel25549 Oct 22 10:55 pi.c
-rw-r--r--1 root  wheel25152 Oct 22 00:07 pi.c
freebie# whoami
root


-- 
Credit is the root of all evil.  - AB1JX

Re: Weird cursor problem

[Alan Corey]

Just to clarify, I see the left-pointing hand, but that's just a
symptom, I can't do anything with the mouse like use the pager or any
other application.  Cursor movement isn't restricted to one window.  I
get out of it by using the keyboard shortcuts of Ctrl-arrow which
normally jumps to another pane in the pager.  So ctrl-right, ctrl-left
and I'm usually back where I started with the problem cleared.  I use
fvwm which is the default window manager.  I don't know what the
version history of that is but it could be the problem.

> It would be helpful if I could get the synaptics driver running first. ;)
> The problem is I am unsure what kind of driver my touchpad uses, and
> xinput reveals this:

Take a look at the mouse stuff in your /var/log/Xorg.0.log, it should
show you what mouse driver you're using.  I know what you mean, I've
gone to great lengths to turn off "tap to click" in a few laptops.
With my current Dell Latitude D530 it defaults to off, at least I
couldn't find where I'm turning it off and I don't remember doing it.
I don' t have an Xorg.conf, everything works (almost) perfectly
without one.  I am using the synaptics driver and no tap to click.  It
was just the default.
-- 
Credit is the root of all evil.  - AB1JX

Re: Weird cursor problem

[Alan Corey]

Anybody else see this?  It's happening at least 6 times a day, it's a
little annoying.  It's happened a few times on my laptop (same 5.7
i386).  It does happen without Firefox open but most of the time
that's open anyway so I've only caught the cursor problem without
Firefox a few times.  Ctrl and any arrow gets out of it.

On 12/3/15, Alan Corey  wrote:
> I don't see anything in the archives about this.  On 5.7 i386 with
> fvwm several times a day my cursor changes to the left-pointing finger
> arrow that Firefox uses to point to links and clicking on things has
> no effect.  I can't change the focus, if an rxvt window has the focus
> I can type in it. Today it happened when Firefox hadn't been running
> in at least a couple hours.  If I can manage to kill Firefox that
> usually stops it.  Today I discovered that ctrl-arrow to change panes
> of the desktop stops it, 100% of the time so far.  Repaint window,
> change cursor I guess.  Not a huge problem, just thought I'd mention
> it.  Just a curiosity.
>
> --
> Credit is the root of all evil.  - AB1JX
>


-- 
Credit is the root of all evil.  - AB1JX

Re: FOSS tools for flashing motherboard BIOS?

[Alan Corey]

> Are you sure there is no iso image with self boot and update, ready to
> be burned on a cdrom? I don't know for sure, but I think last time I
> was using that. I don't own HP gear, but IBM/Lenovo has this kind of
> iso image and I've used it a lot.

Not as far as I can tell, and the guy at HP support I've been chatting
with maintains that Vista is the only way to do it.  I remember Vista
beta came on DVD it was so big.  I used Rufus (Windows freeware) to
load my XP ISO onto an SD card and boot from it to install XP on a
hard drive, then did the BIOS update from that.  Dell would have a
more practical solution too.  I think they had self-extracting files
that wrote an image onto a boot floppy last time I needed to use one
about 2009.  Being tied to Vista?  Yuck.

-- 
Credit is the root of all evil.  - AB1JX

Re: FOSS tools for flashing motherboard BIOS?

[Alan Corey]

The hard drive I stick in when I want some operating systems on one
has XP Pro and OpenBSD 5.2.  OpenBSD 5.2 doesn't have libpci even in
ports so that rules out flashrom unless I want to try to build libpci
from sources (or a distfile).

I tried in XP safe mode running HP's sp40750.exe (which is a SoftPaq
like Compaq uses) and got:
  Can not open driver Flash1.sys.
  The driver is not existed or the OS is not in administrator privilege.

Flash1.sys is contained in sp40750.exe so it must be safe mode that's
killing me.

This is a Phoenix BIOS, not Award, hence the winphlash.  I think I'm
going to try the Vista drivers on HP's page, they might work with XP
and let me boot not into Safe Mode.

On 6/6/16, li...@wrant.com  wrote:
> Mon, 6 Jun 2016 23:36:46 +0300 li...@wrant.com
>> Mon, 6 Jun 2016 16:09:46 -0400 Alan Corey 
>> > > Upgrading sqlports-compact keeps your pkg_mgr aware of current ports.
>> > >
>> > >> So, I guess I should look at BiosDisk too.
>> > >
>> > > If one utility warns you could potentially have unpredictable
>> > > flashing
>> > > results, how is the other utility that does not warn you so
>> > > different?
>> > >
>> > > Tech support may not be engineering support, which you could then
>> > > need.
>> > > Simply just use the manufacturer tools and not blame third party
>> > > tools.
>> >
>> > Because of the dozens of different machines on which this runs I'm not
>> > sure that mine is one that has the problem.  I'd like to actually see
>> > the warning before I decide not to use it.  Biosdisk is about 6 years
>> > old so I decided to not go that route.  Of course the machine is about
>> > 10 years old and so is the BIOS image.
>> >
>> > The manufacturer tools rely on Windows Vista which I don't have or
>> > want.  If they had a more typical floppy image file I'd use it.  I
>> > suspect it might run  under XP but I don't have a way to put XP on the
>> > machine, since it's booting from CD/DVD I'm trying to fix.  Although I
>> > do have a hard drive with XP on it from another machine which blue
>> > screens on this one, maybe I can bring it up in safe mode or
>> > something.  Especially if it doesn't need GUI.
>>
>> I think you're quite competent to continue but have in mind the gotchas.
>> From past experience I recall the command prompt stopped being truly DOS
>> compatible around win98, yet XP could be used to create a DOS compatible
>> boot disk, that would also be able to run award flashing tools.  The HP
>> proprietary stuff however may be another very different piece of... you
>> know what, so the risks that flashboot warns about may actually be Real.
>
> I meant flashrom, not flashboot (different topic).  Back in the days there
> were also uniflash (abandoned) and some other (universal) flash utilities.
>
> From quick searches online HP seem to have some support resources, which is
> your best bet, including the live (preboot environment) for WinXP
> (offlist).
>
>


-- 
Credit is the root of all evil.  - AB1JX

Re: FOSS tools for flashing motherboard BIOS?

[Alan Corey]

> Upgrading sqlports-compact keeps your pkg_mgr aware of current ports.
>
>> So, I guess I should look at BiosDisk too.
>
> If one utility warns you could potentially have unpredictable flashing
> results, how is the other utility that does not warn you so different?
>
> Tech support may not be engineering support, which you could then need.
> Simply just use the manufacturer tools and not blame third party tools.
>

Because of the dozens of different machines on which this runs I'm not
sure that mine is one that has the problem.  I'd like to actually see
the warning before I decide not to use it.  Biosdisk is about 6 years
old so I decided to not go that route.  Of course the machine is about
10 years old and so is the BIOS image.

The manufacturer tools rely on Windows Vista which I don't have or
want.  If they had a more typical floppy image file I'd use it.  I
suspect it might run  under XP but I don't have a way to put XP on the
machine, since it's booting from CD/DVD I'm trying to fix.  Although I
do have a hard drive with XP on it from another machine which blue
screens on this one, maybe I can bring it up in safe mode or
something.  Especially if it doesn't need GUI.

Inside HP's upgrade file is:

total 7600
drwxr-xr-x  1 root  wheel16384 Jun  5 15:14 INSIDE

total 96
-rwxr-xr-x  1 root  wheel  36864 Mar 15  2007 FLASHER.EXE
-rwxr-xr-x  1 root  wheel175 Sep 12  2006 FLASHER.INI
drwxr-xr-x  1 root  wheel  16384 Jun  5 15:14 winphlash
drwxr-xr-x  1 root  wheel  16384 Jun  5 15:14 winphlash64

./INSIDE/winphlash:
total 4096
-rwxr-xr-x  1 root  wheel  1083214 Aug 26  2008 30CDF2D.WPH
-rwxr-xr-x  1 root  wheel32768 Aug 14  2007 FLASH.DLL
-rwxr-xr-x  1 root  wheel  567 Aug 26  2008 FLASH1.INI
-rwxr-xr-x  1 root  wheel 2816 Jun 20  2007 FLASH1.SYS
-r-xr-xr-x  1 root  wheel   995383 Jun  6  2002 MFC42.DLL
-r-xr-xr-x  1 root  wheel   401462 Jun  6  2002 MSVCP60.DLL
-r-xr-xr-x  1 root  wheel   266293 Jun  6  2002 MSVCRT.DLL
-rwxr-xr-x  1 root  wheel  298 Oct  5  2007 PHLASH.INI
-rwxr-xr-x  1 root  wheel26883 Mar  3  2006 Phlash9X.vxd
-rwxr-xr-x  1 root  wheel   200704 Mar  1  2006 PhlashLc.dll
-rwxr-xr-x  1 root  wheel38592 Oct 19  2006 PhlashNT.sys
-rwxr-xr-x  1 root  wheel45056 Aug 14  2007 WFLASH.EXE
-rwxr-xr-x  1 root  wheel   283648 Aug  3  2004 WINHLP32.EXE
-rwxr-xr-x  1 root  wheel   380454 Jan  8  2003 WINPHLASH.HLP
-rwxr-xr-x  1 root  wheel   290816 Oct 19  2006 WinPhlash.exe

./INSIDE/winphlash64:
total 5824
-rwxr-xr-x  1 root  wheel  1083214 Aug 26  2008 30CDF2D.WPH
-rwxr-xr-x  1 root  wheel32768 May 11  2007 FLASH.DLL
-rwxr-xr-x  1 root  wheel  562 Aug 26  2008 FLASH2.INI
-rwxr-xr-x  1 root  wheel36864 May 11  2007 FLASH64.EXE
-rwxr-xr-x  1 root  wheel   635392 Mar 15  2007 FLASH64H.EXE
-r-xr-xr-x  1 root  wheel   995383 Jun  6  2002 MFC42.DLL
-r-xr-xr-x  1 root  wheel   401462 Jun  6  2002 MSVCP60.DLL
-r-xr-xr-x  1 root  wheel   266293 Jun  6  2002 MSVCRT.DLL
-rwxr-xr-x  1 root  wheel  298 Oct  5  2007 PHLASH.INI
-rwxr-xr-x  1 root  wheel   200704 Sep 28  2005 PhlashLc.dll
-rwxr-xr-x  1 root  wheel44400 Mar 28  2007 PhlashNT.sys
-rwxr-xr-x  1 root  wheel   283648 Aug  3  2004 WINHLP32.EXE
-rwxr-xr-x  1 root  wheel   356352 Apr 10  2007 WinPhlash64.exe
-rwxr-xr-x  1 root  wheel  1498301 Dec 22  2006 Winphlash.HLP


-- 
Credit is the root of all evil.  - AB1JX

Re: FOSS tools for flashing motherboard BIOS?

[Alan Corey]

OK, so that's why it wasn't in pbrowser yet.  Running in single user
mode sounds like a pain.  Also the install puts the man page in
/usr/local/share/man/man8 .  And I haven't gotten to why it says not
to run on laptops yet.

On 6/6/16, Daniel Bolgheroni  wrote:

> See this:
>
> http://marc.info/?l=openbsd-ports&m=146464871825362&w=2
>
> --
> db
>


-- 
Credit is the root of all evil.  - AB1JX

Re: FOSS tools for flashing motherboard BIOS?

[Alan Corey]

>From flashrom's man page:

   Laptops

  Using flashrom on laptops is dangerous and may easily make your
  hardware unusable (see also the BUGS section). The embedded
  controller (EC) in these machines often interacts badly with
  flashing.  More information is in the wiki
  <https://flashrom.org/Laptops>.  For example the EC firmware
  sometimes resides on the same flash chip as the host firmware.
  While flashrom tries to change the contents of that memory the
  EC might need to fetch new instructions or data from it and
  could stop working correctly. Probing for and reading from the
  chip may also irritate your EC and cause fan failure, backlight
  failure, sudden poweroff, and other nasty effects. flashrom will
  attempt to detect if it is running on a laptop and abort
  immediately for safety reasons if it clearly identifies the host
  computer as one. If you want to proceed anyway at your own risk,
  use

flashrom -p internal:laptop=force_I_want_a_brick

  We will not help you if you force flashing on a laptop because
  this is a really dumb idea.

So, I guess I should look at BiosDisk too.


On 6/6/16, Alan Corey  wrote:
> OK, so that's why it wasn't in pbrowser yet.  Running in single user
> mode sounds like a pain.  Also the install puts the man page in
> /usr/local/share/man/man8 .  And I haven't gotten to why it says not
> to run on laptops yet.
>
> On 6/6/16, Daniel Bolgheroni  wrote:
>
>> See this:
>>
>> http://marc.info/?l=openbsd-ports&m=146464871825362&w=2
>>
>> --
>> db
>>
>
>
> --
> Credit is the root of all evil.  - AB1JX
>


-- 
Credit is the root of all evil.  - AB1JX

FOSS tools for flashing motherboard BIOS?

[Alan Corey]

I have an HP Pavilion DV2700 laptop with an old BIOS version I'd like
to update.  HP's official solution is to run something under the
dinosaur Windows Vista, which I was glad to wash my hands of about 7
years ago.

I opened up HP's exe file (rename to a zip and unzip), inside is a 1
meg file 30cdf2d.wph which I think is the payload.  It's a Phoenix
BIOS so it's set up to use Phoenix's WinPhlash utility.  I have no
operating system at all on the machine at present, it has no floppy
drive, doesn't boot from the CD/DVD.  I can boot from USB or the hard
drive only.  I can put in an old hard drive from another laptop and
boot it into OpenBSD.  Not booting from the CD is what I'm trying to
fix, a guy at HP support seems to think flashing the BIOS may help.
It's possible there are remnants of Windows "Secure Boot" around from
somebody trying to load Windows 7+ although since I can boot from USB
(a gparted ISO written to an SD card plugged into a USB reader) I
doubt it.

On the Arch Linux page at
https://wiki.archlinux.org/index.php/Flashing_BIOS_from_Linux there's
mention of a couple programs that might work: BiosDisk and Flashrom.
Anybody use either of those under OpenBSD?  I haven't tried chasing
down the source and trying to build them, I was a little surprised
they aren't in sysutils.  I don't see anything in there for flashing,
just looking in pbrowser.

-- 
Credit is the root of all evil.  - AB1JX

Re: the balance between OpenBSD and life

[Alan Corey]

For me, I went from Linux -> FreeBSD -> OpenBSD about 15 years ago and
stayed here.  I dabble in other things like Linux du jour but OpenBSD
is the reliable backbone, reads partitions of other operating systems
on multiboot machines (ok, not ext4).  I've never screwed it up so bad
I couldn't fix it.

Every operating system wants you to devote significant chunks of your
life to learning it.  Debian's really friendly until you break
something then it's so horribly complex (that's putting it nicely)
that no mere human can fix it.  Searchable documentation seems to be
the key, or Googling things.  Searchable mailing list archives.
Unless you intend to actually read every bit of documentation for each
operating system you use.  And by then it will have changed enough so
you need to start over.

-- 
Credit is the root of all evil.  - AB1JX

[no subject]

[Alan Corey]

Re: Performance of Firefox and Chromium

> I'm not  sure what  hardware you guys  run OpenBSD on,  but on  my (old,
> crusty,  crummy, shitty)  laptop,  it  and a  lot  of Gui-requiring  and

My laptop was made in 2008, my desktop in 2002.

>  javascript heavy sites send
> javascript from many domains which is slow and insecure and probably
> increases threading a lot by it's distributed nature.

I blame a lot on Javascipt libraries like Ajax where lazy webmasters
include them to use 1% of what they stick in the page.  But yes, every
page requires 50 or so DNS lookups.  And web pages have gotten to be 3
megabytes and over.  And web servers have timeouts so lots of times I
see them time out before the pages load, I have to hit reload 5-10
times fairly often.  The OpenBSD site is fast, stupid sites like
Facebook I avoid.

-- 
Credit is the root of all evil.  - AB1JX

Re: Performance of Firefox and Chromium

[Alan Corey]

>> A language has nothing to do with speed of execution!

It seems like Javascript's gotten faster in the last 10 years or so.

I used to write little benchmarks to compare Turbo C and Turbo Pascal,
Pascal always won.

> (Point being once things reach a certain level of complexity, issues
> like available developer time and architectural decisions and so on
> can become rather significant.

For a one-time use program sure, but things like Python shouldn't be
unleashed on an unsuspecting public.  Gimp 2.8 is noticeably slower
than 2.6 I think it was in OpenBSD 5.2.  Move the cursor over the
image and it's like it's in la-la land.  Try to sign your name with
the mouse.  Of Inkscape and Libre Office Draw, surprisingly Libre
Office is the faster and works better for an SVG signature.  But not
as fast as this amazing little page:
http://mcc.id.au/2010/signature.html


-- 
Credit is the root of all evil.  - AB1JX

Re: Performance of Firefox and Chromium

[Alan Corey]

Re: Performance of Firefox and Chromium

Several seconds?  Oh my.  Try 20 minutes or more on some of the most
bloated sites, with lots of reloads and watching iftop to see when
they're stuck like on my connection.  But thanks for the tip on
Noscript, I'm trying it out.
-- 
Credit is the root of all evil.  - AB1JX

Re: Trying to move my httpd chroot

[Alan Corey]

Bingo.  /usr does it.  One clue I guess was that it was logging into
/usr/logs.  With Apache at least the chroot dir wasn't the same as the
document root.  And you don't want the logs dir readable through the
httpd.  So essentially there's htdocs and logs inside of what you
specify as a chroot dir.

On 3/16/16, Rick Hanson  wrote:
> On Wed, Mar 16, 2016 at 8:58 PM, Alan Corey  wrote:
>> I don't have enough room in / to have my htdocs there so I want to
>> move it to /usr/htdocs. This is in 5.7.   No problem I thought, I've
>> had to do it before.  So my /etc/httpd.conf looks like this:
>>
>> chroot "/usr/htdocs"
>
> It's probably supposed to be
>
> chroot "/usr"
>
> Check out `man httpd.conf`.  Look at the descriptions for the `chroot`
> and `root` settings.  It appears that both of these settings combine
> to get you what you're looking for in this case.
>
>> server "d530.my.domain" {
>>   listen on * port 80
>> }
>>
>> And I get logging into /usr/htdocs/logs but httpd doesn''t seem to
>> find files in /usr/htdocs.  I get a 404 error that says OpenBSD httpd
>> in it but it can't find even index.html which does exist.  I've played
>> with htdocs vs htdocs/.  If I comment out the chroot line it finds
>> files in /var/www/htdocs.  My /usr is in a different MBR partition
>> (actually an exended one) with 129 gigs free.
>>
>> Anybody tried to move their htdocs?  I didn't find anything by
>> searching.  I wouldn't want to write something and put it out there
>> for everybody to beat on.  I did read the PDF and man pages.
>>
>> Also I found that if I set httpd_flags to "-d -v" in
>> /etc/rc.conf.local then booting  the machine seems to hang there.
>> Permissions on the file look like:
>> -rwxr--r--  1 www  daemon  4022 Jan 19  2015 index.html
>>
>> --
>> Credit is the root of all evil.  - AB1JX
>


-- 
Credit is the root of all evil.  - AB1JX

Trying to move my httpd chroot

[Alan Corey]

I don't have enough room in / to have my htdocs there so I want to
move it to /usr/htdocs. This is in 5.7.   No problem I thought, I've
had to do it before.  So my /etc/httpd.conf looks like this:

chroot "/usr/htdocs"

server "d530.my.domain" {
  listen on * port 80
}

And I get logging into /usr/htdocs/logs but httpd doesn''t seem to
find files in /usr/htdocs.  I get a 404 error that says OpenBSD httpd
in it but it can't find even index.html which does exist.  I've played
with htdocs vs htdocs/.  If I comment out the chroot line it finds
files in /var/www/htdocs.  My /usr is in a different MBR partition
(actually an exended one) with 129 gigs free.

Anybody tried to move their htdocs?  I didn't find anything by
searching.  I wouldn't want to write something and put it out there
for everybody to beat on.  I did read the PDF and man pages.

Also I found that if I set httpd_flags to "-d -v" in
/etc/rc.conf.local then booting  the machine seems to hang there.
Permissions on the file look like:
-rwxr--r--  1 www  daemon  4022 Jan 19  2015 index.html

-- 
Credit is the root of all evil.  - AB1JX

Re: uvm fault booting

[Alan Corey]

It happened again.  I'm using a laptop with a short power cord and
sometimes I recline my recliner and don't notice it fell out.  I'm
usually only up here in the winter.

Anyway I did the drive diagnostics again, saw the end this time and it
didn't find anything wrong.  Rebooted and it fscked the big partition
OK.

So I'm wondering if maybe when fsck runs on multiple partitions it
could be not freeing some memory or something between?  Or is it doing
them simultaneously and that fails?.  The messages look like it's
doing one at a time, I see the first 2 finish then I get the error.

Next time I see that I'll try just rebooting without doing anything else.

But it happened this time without doing anything with USB at all, I
just ran the battery down.

On 2/29/16, Alan Corey  wrote:
> SMART is on but it's never been tripped.
>
> Windows is scanning the FAT32 partition of my SD card  but it won't do the
> ext3.  I had Linux on a phone but I was having selinux problems anyway.  So
> I was going to put NetBSD on it for my Raspberry Pi.
>
> Sent from my Motorola XT1505
> On Feb 29, 2016 10:23 PM, "Josh Grosse"  wrote:
>
>> On Mon, Feb 29, 2016 at 10:19:23PM -0500, Alan Corey wrote:
>> > ...So maybe the Seagate stuff remapped some
>> > sectors, maybe not.
>>
>> It's hard to say, Alan.  If the drive reports S.M.A.R.T. status
>> it's possible the drive's electronics will admit remapped
>> sectors to smartmontools.  But if the drive doesn't report,
>> it's not proof.
>>
>> I'm glad things are working for you.  :)
>>
>


-- 
Credit is the root of all evil.  - AB1JX

Open source SD card utilities?

[Alan Corey]

They aren't hard drives, it's a whole different process.  They work
superficially the same because that's a layer designed on.  If you go
to sdcard.org there are technical specifications and formatters for
Windows and Mac, like that's the whole world.

I've seen formatting with hard disk tools work, I've also had better
luck formatting in cameras or phones.  And many times scanning for bad
 blocks causes a hard crash of the whole computer.  I've got about 3
USB memory sticks here about 10 years old that I don't dare to scan
again.  But that in itself is odd, that for 10 years we haven't come
up with a better approach.  There are hundreds of pages of PDFs on how
they work at http://www.sdcard.org but there seem to be limitations on
how you can use that information.  Like that .org maybe should have
been a .com or .biz  Despicable.

But as we enter an era of solid state drives having proper utilities
becomes more important.  And many of us experiment with rooted phones
with Linux on SD cards, and small single board computers like the
Raspberry Pi, Beagleboard, etc. using SD cards instead of hard drives.

You have to accept this disclaimer before you can download their PDFs:
The information contained in the Simplified Specifications are
presented only as a standard specification for SD cards and SD
host/ancillary products and is provided "AS-IS" without any
representations or warranties of any kind. No responsibility is
assumed by the SD Group, SD-3C, LLC or the SD Card Association for any
damages, any infringements of patents or other right of the SD Group,
SD-3C, LLC, the SD Card Association or any third parties, which may
result from there use or any portion there of. No license is granted
by implication, estoppel or otherwise under any patent or other rights
of the SD Group, SD-3C, LLC, the SD Card Association or any third
party. Nothing herein shall be construed as an obligation by the SD
Group, the SD-3C, LLC or the SD Card Association to disclose or
distribute any technical information, know-how or other confidential
information to any third party.

Ooops, I probably broke some law by posting that. :)  I was never
interested in law school.

-- 
Credit is the root of all evil.  - AB1JX

Re: uvm fault booting

[Alan Corey]

It's running again.  I ran Seagate Diagnostics, the long version, like
2 hours.  When it was about 97% done it hadn't given any messages, and
I went to dinner.  When I came back I found I'd knocked the power plug
out, the battery had run down, so I booted it up cold.  Took longer
than usual but it came up.  So maybe the Seagate stuff remapped some
sectors, maybe not.

So do I dare scan my SD card or not?  I think I'll try it from
Windows, that's more expendable.  Over half the drive is OpenBSD, the
rest is Windows and Linux.

On 2/29/16, Alan Corey  wrote:
> I forgot, wd0a and e aren't the only OpenBSD partitions, the biggie is
> wd0m, about 500 gigs.  There's 4 gigs of RAM and I never had a problem
> with it before, but I can reproduce the uvm fault by trying to fsck it
> in single user mode.  I don't remember whether swap is on at that
> point or not.  Maybe it's time to download Seagate's drive diagnostics
> and run them, but the drive's less than 1 year old.  Maybe it needs to
> remap a sector.
>
> The acpi and USB messages appear every time I open the lid on the
> laptop, I don't pay much attention to them.  It was running badblocks
> on an SD card plugged into a USB reader that started all this, but I
> don't see how a USB problem would persist through reboots.  I've seen
> some nasty crashes dealing with SD and USB memory sticks, it's almost
> like they need their own utilities since they aren't really hard
> drives.
>
> OpenBSD 5.7 (GENERIC.MP) #767: Sun Mar  8 11:04:48 MDT 2015
> dera...@i386.openbsd.org:/usr/src/sys/arch/i386/compile/GENERIC.MP
> cpu0: Intel(R) Core(TM)2 Duo CPU T7250 @ 2.00GHz ("GenuineIntel"
> 686-class) 2 GHz
> cpu0:
> FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,NXE,LONG,SSE3,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,LAHF,PERF
> real mem  = 3747885056 (3574MB)
> avail mem = 3674271744 (3504MB)
> mpath0 at root
> scsibus0 at mpath0: 256 targets
> mainbus0 at root
> bios0 at mainbus0: date 07/16/13, BIOS32 rev. 0 @ 0xffa10, SMBIOS rev.
> 2.4 @ 0xf6bd0 (62 entries)
> bios0: vendor Dell Inc. version "A12" date 07/16/2013
> bios0: Dell Inc. Latitude D530
> acpi0 at bios0: rev 2
> acpi0: sleep states S0 S3 S4 S5
> acpi0: tables DSDT FACP HPET APIC ASF! MCFG TCPA SLIC SSDT
> acpi0: wakeup devices PCI0(S5) PCIE(S4) USB1(S0) USB2(S0) USB3(S0)
> USB4(S0) USB5(S0) EHC2(S0) EHCI(S0) AZAL(S3) RP01(S3) RP02(S4)
> RP03(S3) RP04(S3) RP05(S3) RP06(S5) [...]
> acpitimer0 at acpi0: 3579545 Hz, 24 bits
> acpihpet0 at acpi0: 14318179 Hz
> acpimadt0 at acpi0 addr 0xfee0: PC-AT compat
> cpu0 at mainbus0: apid 0 (boot processor)
> mtrr: Pentium Pro MTRR support, 8 var ranges, 88 fixed ranges
> cpu0: apic clock running at 199MHz
> cpu0: mwait min=64, max=64, C-substates=0.2.2.2.2, IBE
> cpu1 at mainbus0: apid 1 (application processor)
> cpu1: Intel(R) Core(TM)2 Duo CPU T7250 @ 2.00GHz ("GenuineIntel"
> 686-class) 2 GHz
> cpu1:
> FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,NXE,LONG,SSE3,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,LAHF,PERF
> ioapic0 at mainbus0: apid 2 pa 0xfec0, version 20, 24 pins
> ioapic0: misconfigured as apic 0, remapped to apid 2
> acpimcfg0 at acpi0 addr 0xf800, bus 0-63
> acpiprt0 at acpi0: bus 3 (PCIE)
> acpiprt1 at acpi0: bus -1 (AGP_)
> acpiprt2 at acpi0: bus 11 (RP01)
> acpiprt3 at acpi0: bus 12 (RP02)
> acpiprt4 at acpi0: bus -1 (RP03)
> acpiprt5 at acpi0: bus -1 (RP04)
> acpiprt6 at acpi0: bus -1 (RP05)
> acpiprt7 at acpi0: bus 9 (RP06)
> acpiprt8 at acpi0: bus 0 (PCI0)
> acpicpu0 at acpi0: C3, C2, C1, PSS
> acpicpu1 at acpi0: C3, C2, C1, PSS
> acpitz0 at acpi0: critical temperature is 99 degC
> acpibtn0 at acpi0: LID_
> acpibtn1 at acpi0: PBTN
> acpibtn2 at acpi0: SBTN
> acpiac0 at acpi0: AC unit online
> acpibat0 at acpi0: BAT0 model "DELL" serial 1 type LION oem "Panasonic"
> acpibat1 at acpi0: BAT1 not present
> acpidock0 at acpi0: GDCK not docked (0)
> acpivideo0 at acpi0: VID_
> acpivideo1 at acpi0: VID_
> acpivout0 at acpivideo1: LCD_
> acpivideo2 at acpi0: VID2
> bios0: ROM list: 0xc/0xf000! 0xcf000/0x1000
> cpu0: Enhanced SpeedStep 1995 MHz: speeds: 2001, 2000, 1600, 1200, 800 MHz
> pci0 at mainbus0 bus 0: configuration mode 1 (bios)
> pchb0 at pci0 dev 0 function 0 "Intel GM965 Host" rev 0x0c
> vga1 at pci0 dev 2 function 0 "Intel GM965 Video" rev 0x0c
> intagp0 at vga1
> agp0 at intagp0: aperture at 0xe000, size 0x1000
> inteldrm0 at vga1
> drm0 at inteldrm0
> composi

Re: uvm fault booting

[Alan Corey]

at pci2 dev 0 function 0 "Atheros AR5418" rev 0x01: apic 2 int 17
athn0: MAC AR5418 rev 2, RF AR5133 (2T3R), ROM rev 8, address 00:21:63:30:0e:4c
ppb2 at pci0 dev 28 function 5 "Intel 82801H PCIE" rev 0x02: apic 2 int 17
pci3 at ppb2 bus 9
bge0 at pci3 dev 0 function 0 "Broadcom BCM5755M" rev 0x02, BCM5755 A2
(0xa002): msi, address 00:1d:09:d5:dd:b6
brgphy0 at bge0 phy 1: BCM5755 10/100/1000baseT PHY, rev. 0
uhci2 at pci0 dev 29 function 0 "Intel 82801H USB" rev 0x02: apic 2 int 20
uhci3 at pci0 dev 29 function 1 "Intel 82801H USB" rev 0x02: apic 2 int 21
uhci4 at pci0 dev 29 function 2 "Intel 82801H USB" rev 0x02: apic 2 int 22
ehci1 at pci0 dev 29 function 7 "Intel 82801H USB" rev 0x02: apic 2 int 20
usb1 at ehci1: USB revision 2.0
uhub1 at usb1 "Intel EHCI root hub" rev 2.00/1.00 addr 1
ppb3 at pci0 dev 30 function 0 "Intel 82801BAM Hub-to-PCI" rev 0xf2
pci4 at ppb3 bus 3
cbb0 at pci4 dev 1 function 0 "O2 Micro OZ711EZ1 CardBus" rev 0x21:
apic 2 int 19
"O2 Micro Firewire" rev 0x02 at pci4 dev 1 function 4 not configured
cardslot0 at cbb0 slot 0 flags 0
cardbus0 at cardslot0: bus 4 device 0 cacheline 0x0, lattimer 0x20
pcmcia0 at cardslot0
ichpcib0 at pci0 dev 31 function 0 "Intel 82801HBM LPC" rev 0x02: PM disabled
pciide0 at pci0 dev 31 function 1 "Intel 82801HBM IDE" rev 0x02: DMA,
channel 0 configured to compatibility, channel 1 configured to
compatibility
atapiscsi0 at pciide0 channel 0 drive 0
scsibus1 at atapiscsi0: 2 targets
cd0 at scsibus1 targ 0 lun 0:  ATAPI
5/cdrom removable
cd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 2
pciide0: channel 1 ignored (disabled)
pciide1 at pci0 dev 31 function 2 "Intel 82801HBM SATA" rev 0x02: DMA,
channel 0 configured to native-PCI, channel 1 configured to native-PCI
pciide1: using apic 2 int 18 for native-PCI interrupt
wd0 at pciide1 channel 0 drive 0: 
wd0: 16-sector PIO, LBA48, 953869MB, 1953525168 sectors
wd0(pciide1:0:0): using PIO mode 4, Ultra-DMA mode 6
ichiic0 at pci0 dev 31 function 3 "Intel 82801H SMBus" rev 0x02: SMI
iic0 at ichiic0
spdmem0 at iic0 addr 0x50: 2GB DDR2 SDRAM non-parity PC2-5300CL6 SO-DIMM
spdmem1 at iic0 addr 0x52: 2GB DDR2 SDRAM non-parity PC2-5300CL6 SO-DIMM
usb2 at uhci0: USB revision 1.0
uhub2 at usb2 "Intel UHCI root hub" rev 1.00/1.00 addr 1
usb3 at uhci1: USB revision 1.0
uhub3 at usb3 "Intel UHCI root hub" rev 1.00/1.00 addr 1
usb4 at uhci2: USB revision 1.0
uhub4 at usb4 "Intel UHCI root hub" rev 1.00/1.00 addr 1
usb5 at uhci3: USB revision 1.0
uhub5 at usb5 "Intel UHCI root hub" rev 1.00/1.00 addr 1
usb6 at uhci4: USB revision 1.0
uhub6 at usb6 "Intel UHCI root hub" rev 1.00/1.00 addr 1
isa0 at ichpcib0
isadma0 at isa0
com0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pms0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pms0 mux 0
pms0: ALPS Glidepoint, version 0x7321
pcppi0 at isa0 port 0x61
spkr0 at pcppi0
npx0 at isa0 port 0xf0/16: reported by CPUID; using exception 16
vscsi0 at root
scsibus2 at vscsi0: 256 targets
softraid0 at root
scsibus3 at softraid0: 256 targets
root on wd0a (abd27361b43df756.a) swap on wd0b dump on wd0b
umass0 at uhub1 port 1 configuration 1 interface 0 "Kingston
DataTraveler 2.0" rev 2.00/1.00 addr 2
umass0: using SCSI over Bulk-Only
scsibus4 at umass0: 2 targets, initiator 0
sd0 at scsibus4 targ 1 lun 0:  SCSI4
0/direct removable serial.09511665BE81491E5F0A
sd0: 14888MB, 512 bytes/sector, 30490624 sectors


On 2/29/16, Josh Grosse  wrote:
> On 2016-02-29 13:14, Alan Corey wrote:
>
>> I didn't send this to the list but I did cell phone pictures of the
>> screen.
>> Stopped in pmap_extract both times. And it gets through checking my
>> OpenBSD
>> partitions OK. Doesn't usually try to fsck the Linux partitions. I've
>> seen
>> booting after unclean shutdowns many times  but it's always done the
>> fsck
>> and recovered.
>
> Thanks for the photo.  Along with the pmap_extract fault, I see a
> significant
> USB event.  Since UVM is the BSD virtual memory manager, I assume the
> USB
> event affects virtual memory maps, hence the fault.
>
> A pmap is the physical map of a virtual memory page to a physical
> location.
> (That's really all I know about them.)
>
> The ACPI event is probably related to the USB event, as it occurs
> between
> the last successful fsck and USB detaching.
>
> I can't tell you what the system is trying to do at the moment of
> failure;
> fsck executions during boot are based on fstab(5) fs_passno values (the
> sixth
> column in /etc/fstab), and its possible it's attempting the next fsck,
> or
> doing the next set of services in the rc(8) script -- which is mounting
> your
> filesystems.
>
> Either post a link to the photo, or type up the captured screen and send
> it to misc@.  With a dmesg, which you should be able to get by starting
> in single-user mode.
>
> -Josh-
>
>


-- 
Credit is the root of all evil.  - AB1JX

Re: uvm fault booting

[Alan Corey]

Transcribing from a cell phone picture of the screen on another machine later:

/dev/rwd0a fscked OK
/dev/rwd0e fscked OK (both of my OpenBSD partitions)
acpivideo1 unknown event 0x00
uhubs 0-6 detached
uhubs 0-6 at usb0-6 "Intel UHCI (and EHCI) root hub" rev 1.00 (2.0 on the EHCI)
uvm_fault (0xdaa7a1b0, 0xcfe08000,0,1) -> e
kernel: page fault trap code = 0
Stopped at   pmap_extract+0x3f:   mov1  [camera lat/lon stamp on top of this]

-- 
Credit is the root of all evil.  - AB1JX

uvm fault booting

[Alan Corey]

Everything's been stable for months.  Then I was running badblocks on an SD
card plugged into a USB card reader.  The screen went white,  I couldn't
ssh to it, so I killed the power. It boots to just past the file system
checks and hangs, disk access light on.  Did that before so I went away for
an hour, came back to find the uvm fault message. 5.7 i386

Sent from my Motorola XT1505

Thousands separator in printf?

[Alan Corey]

I'm not dyslexic but I have trouble looking at big numbers like 10
digits and telling which is bigger.  So I Googled and with most GCC
versions with a working locale you put something like %'u in a printf.
I also saw mention of a SATSEP macro, but I'm not getting anything to
work.  I know I could write a formatting function, I just wondered if
there's an easier way.  And I'm counting bytes, not dollars. :)

-- 
Credit is the root of all evil.  - AB1JX

stubborn athn

[Alan Corey]

I have an Atheros AR5418 mini-PCI card in my laptop, which I think
came from an eBay seller in China.  It used to work under OpenBSD, but
that may have been back about 5.2 or 4.7.  Now I'm using 5.7.  It
works now in Kismet under OpenBSD, but an ifconfig scan comes up not
finding anything.  If I set the nwid and chan in ifconfig and do a
dhclient I get no link.

If I boot the same machine into Linux it works, but mostly only using
WiCD to configure it.  One thing I notice is that Linux dmesg shows
the regdomain as 0x6a, which could be a concern since it came directly
from China.

OpenBSD dmesg says:

athn0 at pci2 dev 0 function 0 "Atheros AR5418" rev 0x01: apic 2 int 17
athn0: MAC AR5418 rev 2, RF AR5133 (2T3R), ROM rev 8, address
00:21:63:30:0e:4c

ifconfig says:

athn0: flags=8802 mtu 1500
lladdr 00:21:63:30:0e:4c
priority: 4
groups: wlan
media: IEEE802.11 autoselect
status: no network
ieee80211: nwid ""

lspci says:

0c:00.0 Network controller: Qualcomm Atheros AR5418 Wireless Network
Adapter [AR5008E 802.11(a)bgn] (PCI-Express) (rev 01)
Subsystem: Askey Computer Corp. Device 7125
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr-
Stepping- SERR+ FastB2B- DisINTx-
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort-
SERR- 

Re: Daily digest, Issue 3715 (21 messages)

[Alan Corey]

Re: dhcrelay: send_packet: No buffer space available

If it's easy to do try a different network card.  The only time I've
ever seen that error came from a urtwn card under OpenBSD 5.7 and
earlier.  But Stuart knows a lot more about it than I do.

On 2/13/16, owner-m...@openbsd.org  wrote:
> The pre-dawn daily digest
> Volume 1 : Issue 3715 : "text" Format
>
> Messages in this Issue:
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   Re: Kernel panic during installation
>   OpenBGPd RTBH peer with match clause on community
>   Re: OpenBGPd RTBH peer with match clause on community
>   Re: OpenBGPd RTBH peer with match clause on community
>   Re: NVM Express (NVMe) support status
>   Re: Problems using squid as transparent proxy for SSL/TLS
>   Re: dhcrelay: send_packet: No buffer space available
>   Re: dhcrelay: send_packet: No buffer space available
>   Re: OpenBSD5.7, hangs on ppb6 "Intel 5000 PCIE" Dell poweredge 1950
>   how to mount a *dmg in -current
>   Re: how to mount a *dmg in -current
>   Re: how to mount a *dmg in -current
>   pfsync and table
>   Re: pfsync and table
>   Re: sshfs man page, -o idmap=user
>
> --
>
> Date: Fri, 12 Feb 2016 07:51:34 -0500
> From: Donald Allen 
> To: misc 
> Subject: Re: Kernel panic during installation
> Message-ID:
> 
>
> On Feb 12, 2016 05:08, "Stefan Sperling"  wrote:
>>
>> On Thu, Feb 11, 2016 at 08:42:21PM -0500, Donald Allen wrote:
>> > When attempting to install the 2/8 snapshot on my Thinkpad x-250, I
> chose
>> > to configure the wireless network interface (iwm). This resulted in the
>> > following:
>> >
>> > iwm0: could not read firmware iwm-7265-9 (error 2)
>> > panic: attempt to execute user address 0x0 in supervisor mode
>>
>> Do you have a trace from ddb please?
>
> There was no entry to ddb. There was one additional message after the
> above:
>
> syncing disks... done
>
> and that was all she wrote. (I took a photo of the screen.)
>
> If you have a suggestion for how to get the trace, I will certainly try to
> help. Or maybe build a kernel with some additional printfs to try to
> isolate where this is happening?
>
>
> --
>
> Date: Fri, 12 Feb 2016 07:45:35 -0800
> From: Chris Cappuccio 
> To: Donald Allen 
> Cc: misc 
> Subject: Re: Kernel panic during installation
> Message-ID: <20160212154535.gb5...@ref.nmedia.net>
>
> Donald Allen [donaldcal...@gmail.com] wrote:
>> On Feb 12, 2016 05:08, "Stefan Sperling"  wrote:
>> >
>> > On Thu, Feb 11, 2016 at 08:42:21PM -0500, Donald Allen wrote:
>> > > When attempting to install the 2/8 snapshot on my Thinkpad x-250, I
>> chose
>> > > to configure the wireless network interface (iwm). This resulted in
>> > > the
>> > > following:
>> > >
>> > > iwm0: could not read firmware iwm-7265-9 (error 2)
>> > > panic: attempt to execute user address 0x0 in supervisor mode
>> >
>> > Do you have a trace from ddb please?
>>
>> There was no entry to ddb. There was one additional message after the
>> above:
>>
>> syncing disks... done
>>
>> and that was all she wrote. (I took a photo of the screen.)
>>
>> If you have a suggestion for how to get the trace, I will certainly try
>> to
>> help. Or maybe build a kernel with some additional printfs to try to
>> isolate where this is happening?
>
> sysctl ddb.panic=1 ??
>
>
> --
>
> Date: Fri, 12 Feb 2016 10:47:16 -0500
> From: Donald Allen 
> To: Chris Cappuccio 
> Cc: misc 
> Subject: Re: Kernel panic during installation
> Message-ID:
> 
>
> On Fri, Feb 12, 2016 at 10:45 AM, Chris Cappuccio  wrote:
>
>> Donald Allen [donaldcal...@gmail.com] wrote:
>> > On Feb 12, 2016 05:08, "Stefan Sperling"  wrote:
>> > >
>> > > On Thu, Feb 11, 2016 at 08:42:21PM -0500, Donald Allen wrote:
>> > > > When attempting to install the 2/8 snapshot on my Thinkpad x-250, I
>> > chose
>> > > > to configure the wireless network interface (iwm). This resulted in
>> the
>> > > > following:
>> > > >
>> > > > iwm0: could not read firmware iwm-7265-9 (error 2)
>> > > > panic: attempt to execute user address 0x0 in supervisor mode
>> > >
>> > > Do you have a trace from ddb please?
>> >
>> > There was no entry to ddb. There was one additional message after the
>> above:
>> >
>> > syncing disks... done
>> >
>> > and that was all she wrote. (I took a photo of the screen.)
>> >
>> > If you have a suggestion for how to get the trace, I will certainly try
>> to
>> > help. Or maybe build a kernel with some additional printfs to try to
>> > isolate where this is happening?
>>
>> sysctl ddb.panic=1 ??
>>
>
> Ah .. thank you. I'll give it a try.
>
>
> --
>
> Date: Fri, 12 Feb 2016 17:17:31 +0100
> From: Stefan Sperling 
> To: Donald Allen 
> Cc: Chris Cappuccio , misc 
> Su

Re: text-mode gui

[Alan Corey]

I can't resist jumping in.  I see what y'all are saying but have you
considered the possibility of multiple installers?  The standard
officially supported and maintained one that's guaranteed to work.
Others, clearly labelled as contributed works, buried in sysutils and
maybe on github that have colors and fonts and themes and music.
Video for all I care.

I don't know or care what softdep is and if you're going to impose
full-disk encryption I'll find a different operating system.  KISS,
too much to go wrong.

Back about 2008 I had my own way of downloading and installing, still
with the stock installer.  I'd download some files and put them on a
CD, using the install floppy image as a boot image.  Boot the CD as a
floppy, shell out and mount it as a CD, then go back and install from
a mounted drive.  Worked fine for years until somebody had a "bright
idea" that broke it.  I don't mind reading man pages but re-reading
them to see if they've changed gets tedious.  And you can rarely
change something without breaking something else.

I've been stuck in Linux-land lately because there's no BSD for my
phones.  Mostly it leaves a bad taste in my mouth but I can't help
noticing how dpkg and apt-get and Synaptic (Debian) are different ways
of doing the same things and they don't conflict with each other.
There are a few variations of fdisk plus parted and gparted, fixparts
and testdisk.  There are several ways of installing, GUI and not, then
there's debootstrap.  Still dealing with that, installing a minimal
ARM system from an i386 machine.  But it'll fit in my pocket and I've
done it before.  Over time the tools broke because somebody changed
something.

-- 
Credit is the root of all evil.  - AB1JX

Re: QIV is faster in Linux

[Alan Corey]

If I could get logs from the mplayers would that help?  I tried but it
didn't work so I skipped it.  I don't know how to profile on Linux.
The machine has 4 gigs of RAM.

On 12/14/15, Michael McConville  wrote:
> Alan Corey wrote:
>> I'm thinking this is graphics slowness or filesystem slowness.  Both
>> OpenBSD and Debian both have working mplayers, I don't know if that
>> could log something useful about throughput.
>
> Filesystems slowness seems unlikely to me. I'd guess that it's because
> of the rendering acceleration. Maybe we don't support what Linux uses in
> that case, or there's an ifdef somewhere that gives us a sluggish
> fallback.
>
> It could also be because of a pathological memory allocation pattern.
> They have a much bigger performance impact on OpenBSD because of the
> memory sanitization.
>
> The list goes on. You really can't know without profiling.
>
> That said, remember that things like this are almost always a little
> faster on Linux.
>


-- 
Credit is the root of all evil.  - AB1JX

Re: QIV is faster in Linux

[Alan Corey]

Stuck them on devio, simpler than that GUI crap

  http://devio.us/~ab1jx/files/slowness/

Outputs from time(1) too

time qiv dsc_2258.jpg

Copied and pasted, not redirected

Debian:
0.572u 0.316s 0:03.28 26.8% 0+0k 0+0io 0pf+0w
OpenBSD:
1.920u 0.510s 0:05.97 40.7% 0+0k 182+19io 762pf+0w


-- 
Credit is the root of all evil.  - AB1JX

Re: QIV is faster in Linux

[Alan Corey]

> Can you share a photo taken with your camera and the dmesg of your
> computer?

Sure, but where?  The jpegs are 6 - 15 megs each.  I don't know
anything about pastebin-type services yet except that they exist.

I'm thinking this is graphics slowness or filesystem slowness.  Both
OpenBSD and Debian both have working mplayers, I don't know if that
could log something useful about throughput.

-- 
Credit is the root of all evil.  - AB1JX

QIV is faster in Linux

[Alan Corey]

Just an observation on my multiboot laptop.  My current Nikon takes 24
megapixel images, 6000x4000 and I almost dread looking at them in QIV
under OpenBSD.  It's not so bad in Debian, same hardware.  I don't
know how to localize or quantify that.  I guess I'd need to build a
profiled version of QIV, maybe later.  2 profiled versions.

-- 
Credit is the root of all evil.  - AB1JX

Thanks guys

[Alan Corey]

I'm still on my project to multiboot 2 machines so I have Linux to try
Android development but I've got one working.  I just reinstalled Lilo
from a Debian install CD, and I can do it again if I break it again.
So XP, OpenBSD, Debian, all working on my 2008 Dell Latitude D530
laptop.

I had heard that Arch Linux was the most like OpenBSD but it's nowhere
near as mature.  I spent about a week on it, downloaded the live CD
and installed from it, only to find the live CDs aren't up to date
really and installing took another 24 hours of downloading current
stuff.  X isn't one nice tidy package there like it is with OpenBSD.
I finally got it working but FVWM never looked quite right, some stuff
missing, some other stuff I didn't bother to remove.  Never got to
1920x1080 video like I have here.  My motherboard video from Intel
(different machine) is really made by Nvidia so I tried about 3
drivers including one from Nvidia that wouldn't compile.  The same
hardware has been working without a glitch under OpenBSD for years, I
never knew there was anything odd about it, it just works.  Installing
OpenBSD has never been much harder than installing Windows.  Arch was
too much like Slackware from 1995, tracking down drivers, calculating
modelines.  Some strange rationale for using network interface names
like enp0s29f7u2 too.  Nothing good about the experience at all, I
could have lived with it, 20 years ago I might have, but Debian and
OpenBSD got me used to an easier life.  And I've used FreeBSD and Red
Hat before they got so commercial.  Ubuntu, tried it, replaced it with
Debian in a week. Linux is still sorta Mickey Mouse in my opinion, but
sometimes it works.

Device Boot  StartEndSectors   Size Id Type
/dev/sda1  *63   65545199   65545137  31.3G  c W95 FAT32 (LBA)
/dev/sda2 65545200  198659789  133114590  63.5G a6 OpenBSD
/dev/sda3198659790  264188924   65529135  31.3G 83 Linux
/dev/sda4264188986 1953520064 1689331079 805.5G  5 Extended
/dev/sda5264188988  329734124   65545137  31.3G  b W95 FAT32
/dev/sda6329734188  395279324   65545137  31.3G  b W95 FAT32
/dev/sda7395279388  4034724748193087   3.9G 82 Linux swap / Solaris
/dev/sda8403476480 1427476479 102400 488.3G a6 OpenBSD
/dev/sda9   1427487768 1953520064  526032297 250.9G 83 Linux

  Alan Corey, ab1jx
-- 
Credit is the root of all evil.  - AB1JX

Re: Hitting the bootable cylinder limit?

[Alan Corey]

Another possibility is that something like a bootloader was looking
for a file at a particular location (sector, inode) on the disk.  In
installing Grub under Arch Linux there's mention of that.  Linux has
an immutable flag on files, with a utility chattr for setting and
unsetting it, when it's set supposedly the file can't be moved.

-- 
Credit is the root of all evil.  - AB1JX

Re: disklabel suggestion

[Alan Corey]

For posterity the problem was that I had a shared swap partition which
of course didn't show in the list of mounted stuff.  Added it a week
or two ago and forgot it.

But my original point was that this was an editor which lost what was
being edited just because it couldn't save.  With no warning.  Would
anyone tolerate that in a word processor?  I had about 1/2 hour into
that session, copying and pasting, changing between starting and
ending points and starting points and lengths.  Math/calc was really
handy.

Weird cursor problem

[Alan Corey]

I don't see anything in the archives about this.  On 5.7 i386 with
fvwm several times a day my cursor changes to the left-pointing finger
arrow that Firefox uses to point to links and clicking on things has
no effect.  I can't change the focus, if an rxvt window has the focus
I can type in it. Today it happened when Firefox hadn't been running
in at least a couple hours.  If I can manage to kill Firefox that
usually stops it.  Today I discovered that ctrl-arrow to change panes
of the desktop stops it, 100% of the time so far.  Repaint window,
change cursor I guess.  Not a huge problem, just thought I'd mention
it.  Just a curiosity.

-- 
Credit is the root of all evil.  - AB1JX

Re: disklabel suggestion

[Alan Corey]

I understand what it's saying but I can't figure out which one it's
complaining about.  All I have mounted is:

freebie# mount
/dev/wd0a on / type ffs (local)
/dev/wd0i on /win_c type msdos (local)
/dev/wd0l on /win_d type msdos (local)
/dev/wd0m on /win_e type msdos (local)
/dev/wd0n on /usr type ffs (local, nodev)
fusefs on /root/.gvfs type fuse (local)
freebie#

I'm trying to go from:
# /dev/rwd0c:
type: ESDI
disk: ESDI/IDE disk
label: ST31000340AS
duid: f1f9d8681047d339
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 255
sectors/cylinder: 16065
cylinders: 121601
total sectors: 1953525168
boundstart: 65545200
boundend: 196619535
drivedata: 0

16 partitions:
#size   offset  fstype [fsize bsize  cpg]
  a: 41945712 65545200  4.2BSD   2048 163841 # /
  b:  2088453107490912swap   # none
  c:   19535251680  unused
  d: 87040128109579392  4.2BSD   2048 163841
  i: 65545137   63   MSDOS   # /win_c
  j:32130196619535  ext2fs   # /grubpart
  k: 10249407262148733 unknown   # none
  l: 65545137272398203   MSDOS   # /win_d
  m: 65545137337943403   MSDOS   # /win_e
  n:   1023999102403488603  4.2BSD   2048 163841 # /usr
  o:526032297   1427487768  ext2fs

to  (boundend is wrong here, it might as well be the whole drive)

# /dev/rwd0c:
type: ESDI
disk: ESDI/IDE disk
label: ST31000340AS
duid: f1f9d8681047d339
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 255
sectors/cylinder: 16065
cylinders: 121601
total sectors: 1953525168
boundstart: 65545200
#boundend: 196619535
boundend: 1427487704
drivedata: 0

16 partitions:
#size   offset  fstype [fsize bsize  cpg]
  a: 41945712 65545200  4.2BSD   2048 163841 # /
  b:  2088453107490912swap   # none
  c:   19535251680  unused
#  d: 87040128109579392  4.2BSD   2048 163841
  e: 4096196671488  ext2fs   # /debroot
  f: 24515127237633543  ext2fs   # /archroot
  g:32130196619535  ext2fs   # /grubpart
  i: 65545137   63   MSDOS   # /win_c
  j: 10244096262152192swap   # shared swap
#  k: 10249407262148733 unknown   #
none (?) moved
# k was my /usr as installed, I moved the files to n
  l: 65545137272398203   MSDOS   # /win_d
  m: 65545137337943403   MSDOS   # /win_e
  n:   1023999102403488603  4.2BSD   2048 163841 # /usr
  o:327677742   1427487768  ext2fs   # /debdata
  p:184313856   1769209856  ext2fs   # /archdata

I did some rearranging in gparted but I was careful to not move
anything except empty partitions.  Everything except the little grub
partition was empty.  o doesn't exist anymore, there was never
anything in there, but it's not mounted either.



On 12/2/15, Theo de Raadt  wrote:
>> I'm trying to make several changes to my disklabel at once.  If I try
>> to do it with -R to read in a file I get disklabel: ioctl DIOCWDINFO:
>> Open partition would move or shrink
>
> You are attempting to change the position or size of a mounted partition.
> You can't do that.  The filesystem will attempt to write out to the disk
> and scribble somewhere unhealthy.  The kernel therefore refuses to
> perform the action.
>


-- 
Credit is the root of all evil.  - AB1JX

disklabel suggestion

[Alan Corey]

I'm trying to make several changes to my disklabel at once.  If I try
to do it with -R to read in a file I get disklabel: ioctl DIOCWDINFO:
Open partition would move or shrink

So I used -E and used the interactive editor, which let me get through
the same edits without complaining about anything, then when I did q
gave the same error and dumped everything.  It doesn't say which
partition would change and I can't spot it, doesn't give you the
option to stay in the editor or save to a file.  I guess I have to
make one change at a time or make one and write it with w before the
next one.  I thought I was getting away with something until the last
minute.  Seems like on that error it could stay in the editor, maybe
even narrow down the problem better in the error message.  Don't quit
unless the write was successful?

-- 
Credit is the root of all evil.  - AB1JX

Is OpenBSD dump/restore compatible with Linux?

[Alan Corey]

More of my running out of BIOS-accessible space saga.

My Linux partitions with a few gigs of Android development stuff are
intact on my laptop, I just need to rearrange the boot partition (I
think).  I've got empty partitions set aside for installing Linux on
my desktop machine.  Can I do a dump from OpenBSD over my lan to a
Linux restore (I've now got an Arch Linux live cd)?  Or should I do
the restore to OpenBSD on the desktop writing to the Linux partitions?
 I don't know about permissions and ownerships of files, specifically.

I set that up differently.  I used 2 cylinders to make a grub-only
partition at the start of the Linux space.  I'm hoping to load grub as
an option from the XP bootloader then having it boot Debian and
probably Arch.  If I get it working I'll set my laptop up the same way
and clone it back.  I could use rsync or make tarballs but I think
dump/restore is probably best.

Oh, the grub in ports/sysutils seems to be the original, not grub2
which development has also stopped on.  The original got to be too
patched together, so grub2 is a complete rewrite.  stage1 and stage2
is old stuff.  A Grub partition needs to be at least 12 megs.

-- 
Credit is the root of all evil.  - AB1JX

Re: Daily digest, Issue 3641 (37 messages)

[Alan Corey]

re: bootable cylinder limit?

>The point is that if you use Windows you must use its boot menu, and it's
> easier to configure it to boot multiple OS than grub or lilo. EasyBSD
> handles all the boot blocks for you.

One problem is that everything is old (including me).  The computers
are from 2003 and 2008, they get hard drives replaced every few years
before they wear out.  Windows is XP (Pro, sp 3) because I don't spend
money on Microsoft products.  The computers came with COA stickers for
XP, that's what I use.  EasyBCD does nothing, it won't even install
under XP.

I've been multibooting for 15 years, that in itself isn't the problem,
I've done it many times.  I probably won't try FreeBSD and OpenBSD
together again.  Mostly I use one tool to make my partitions and a
second one to check up on it, looking for things like overlapped
partitions and partitions not ending on cylinder boundaries.  As disks
get bigger the tools have to be replaced.  My favorite until this
drive was gdisk by Symantek, part of the Ghost package.  This time I
used Gparted http://gparted.org/ and it looks OK so far, I used the
live Linux CD version.  Make the partitions right and even Windows
will live with them, but I always install it first because it has a
history of obliterating other operating systems on the same drive.
And let it run the show, use its boot menu, start other operating
systems from boot.ini.

The problem is that every computer's BIOS can only allocate X number
of bits for storing things like a cylinder number or (worse) an LBA.
Historically there have always been limits, sometimes you can update
the motherboard's BIOS and that helps a little.  There was one at 540
MB.  But these limits are rarely published because partly they depend
on the BIOS and partly the drive geometry.  Drive geometry isn't
absolute, you don't have to stick with hardware reality.  Too many
heads?  Cut it in half and double the cylinders instead.  That's not
always wise but it's been done.  The manufacturers put the limit so
high it won't be a problem, but then drives get bigger.  My first hard
drive was 20 megabytes, now I have some 6(?) orders of magnitude
bigger,

Write something in assembly using only BIOS interrupts (no operating
system) that gets hooked by the BIOS when booting to start it, it
tries to access higher and higher LBAs until it fails, meanwhile
logging.  I taught myself x86 assembly language about 1994, haven't
used it since, but it sounds possible.  I used the A86, Masm, Tasm
assemblers, not nasm, but to me the thrill of assembly is that things
happen instantly when the program's running.  My new fast machine was
a 40 MHz 386 SX motherboard with a brand new 200 megabyte hard drive.
I pounded a lot of nails, did a lot of grunt work, to earn it.

On 11/25/15, Peter Kay  wrote:
>  Yes, it is possible for grub to boot Windows. LILO too, it can even boot
> Xen if you use mbootpack (otherwise it doesn't support initrd).
>
> The point is that if you use Windows you must use its boot menu, and it's
> easier to configure it to boot multiple OS than grub or lilo. EasyBSD
> handles all the boot blocks for you.
> --
> Sent from my Android device with K-9 Mail. Please excuse my brevity.


-- 
Credit is the root of all evil.  - AB1JX

Re: Daily digest, Issue 3641 (37 messages)

[Alan Corey]

re: bootable cylinder limit?

All manner of things seem to have broken when I went from a 500 gig
drive to 1 TB, or maybe it's because I added Linux.  For years I've
been using the method that used to be in the OpenBSD FAQ of using dd
to write out the first sector of the partition you want to boot to a
file, copying that into the Windows partition, then setting it up in
Windows boot.ini.  It worked this time for a week or so, and only
Linux broke, OpenBSD and Windows still work.

I used lilo because it was willing to install into the Linux
partition, not the MBR.  That might be possible with grub, I'm now
reading http://www.gnu.org/software/grub/manual/grub.html.  Seems like
I might need to chain load grub from the Windows bootloader.  I wanted
each OS self-contained so as a last resort if I flagged that partition
bootable the OS installed there would boot, or I could link a copied
bootsector from boot.ini.

I've used lilo (and loadlin) before, not grub.  Grub seemingly won't
boot Windows, it has to be the other way around. I did get lilo up by
putting the Debian install CD back in and it seems limited to LBA32,
not LBA48 as dmesg shows my drive using.  Yes, the problem with LBA,
not CHS, is that you need really big (unsigned) integers.

I hate it when you want to return to a simpler way of life and find it
doesn't work anymore.  I have a bootable floppy image from Windows 95
so I just tried to set that up as the bootable part of a CD (worked
before) so I could run Norton Utilities to look at the MBR.  Comes up
not finding command.com.  Same thing happens with a Dell Diagnostics
CD I made in 2008.  All this fancy crap...

--
Credit is the root of all evil.  - AB1JX

Re: Hitting the bootable cylinder limit?

[Alan Corey]

I don't think that's it, I've never had me-tv running.  Not too
serious about watching TV, I just have a Diamond TVW750U USB dongle
tuner I couldn't get to work under Windows, wanted to try it under
Linux.  I think that Ubuntu bug relates to having a colon in the file
name.

I'm using Lilo in that partition, which I might be able to fix by
reinstalling it, but the same thing is likely to happen at some random
point in the future.

On 11/23/15, Joel Rees  wrote:
> Theo, if want to suggest that the OP look at Ubuntu-bugs bug 491602, should
> I do so off-list?
>
> ;-)
>
> 2015/11/24 1:55 "Alan Corey" :
>>
>> It seems like there should be a better way to detect this other than
>> trial and error.  I put a new 1 TB drive in my laptop (Seagate
>> ST1000LM024) about a month ago.  Being aware there was such a limit I
>> made small boot partitions at the beginning of the drive (I thought):
>> 32 GB Windows, 64 GB OpenBSD, 32 GB Linux.  As predicted everything
>> worked at first, then installing MeTV keys made my Linux unbootable
>> with an error from Lilo about the key file being corrupt and I suspect
>> it's related to this limit.  The original position of the file was
>> probably OK, the new file got made in an unreachable position.
>>
> [...]
>
> Joel Rees
>
> Computer memory is just fancy paper,
> CPUs just fancy pens.
> All is a stream of text
> flowing from the past into the future.
>


-- 
Credit is the root of all evil.  - AB1JX

Hitting the bootable cylinder limit?

[Alan Corey]

It seems like there should be a better way to detect this other than
trial and error.  I put a new 1 TB drive in my laptop (Seagate
ST1000LM024) about a month ago.  Being aware there was such a limit I
made small boot partitions at the beginning of the drive (I thought):
32 GB Windows, 64 GB OpenBSD, 32 GB Linux.  As predicted everything
worked at first, then installing MeTV keys made my Linux unbootable
with an error from Lilo about the key file being corrupt and I suspect
it's related to this limit.  The original position of the file was
probably OK, the new file got made in an unreachable position.

So I've probably got some storage-only partitions that won't boot, but
I want to avoid the same thing happening when I put a 1 TB drive
(Seagate
ST31000340AS) in my laptop machine (Dell Optiplex GX270) because I
really would like Linux working somewhere since I want to play with
Android stuff.  I need to be able to build kernels for my phones and
use Android Studio.

So on the laptop:
Disk: wd0   geometry: 121601/255/63 [1953525168 Sectors]
Offset: 0   Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
*0: 0C  0   1   1 -   4079 254  63 [  63:65545137 ] Win95 FAT32L
 1: A6   4080   0   1 -  12365 254  63 [65545200:   133114590 ] OpenBSD
 2: 83  12366   0   1 -  16444 254  63 [   198659790:65529135 ] Linux files*
 3: 05  16445   0  62 - 121600 254  63 [   264188986:  1689331079 ] Extended DOS
Offset: 264188986   Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
 0: 0B  16445   1   1 -  20524 254  63 [   264188988:65545137 ] Win95 FAT-32
 1: 05  20525   0   1 -  24604 254  63 [   329734125:65545200 ] Extended DOS
 2: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 3: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
Offset: 329734125   Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
 0: 0B  20525   1   1 -  24604 254  63 [   329734188:65545137 ] Win95 FAT-32
 1: 05  24605   0   1 -  25114 254  63 [   395279325: 8193150 ] Extended DOS
 2: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 3: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
Offset: 395279325   Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
 0: 82  24605   1   1 -  25114 254  63 [   395279388: 8193087 ] Linux swap
 1: 05  25115   0   1 -  88856  76  52 [   403472475:  1024004005 ] Extended DOS
 2: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 3: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
Offset: 403472475   Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
 0: A6  25115  63  37 -  88856  76  52 [   403476480:  102400 ] OpenBSD
 1: 05  88857   0   1 - 121600 254  63 [  1427487705:   526032360 ] Extended DOS
 2: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 3: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
Offset: 1427487705  Signature: 0xAA55
Starting Ending LBA Info:
 #: id  C   H   S -  C   H   S [   start:size ]
---
 0: 83  88857   1   1 - 121600 254  63 [  1427487768:   526032297 ] Linux files*
 1: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 2: 00  0   0   0 -  0   0   0 [   0:   0 ] unused
 3: 00  0   0   0 -  0   0   0 [   0:   0 ] unused

# /dev/rwd0c:
type: ESDI
disk: ESDI/IDE disk
label: ST1000LM024 HN-M
duid: abd27361b43df756
flags:
bytes/sector: 512
sectors/track: 63
tracks/cylinder: 255
sectors/cylinder: 16065
cylinders: 121601
total sectors: 1953525168
boundstart: 65545200
boundend: 198659790
drivedata: 0

16 partitions:
#size   offset  fstype [fsize bsize  cpg]
  a:  4194288 65545200  4.2BSD   2048 163841 # /
  b:   524288 69739488swap   # none
  c:   19535251680  unused
  d:  6291456 70263776  4.2BSD   2048 163841
  e:

dpb - um, priority?

[Alan Corey]

dpb's great, especially since I rtfm'd enough to find the -I flag.
But I'm trying to build and install ports on a machine and use it for
something else at the same time.  It's not like I've got a lot of
machines.

It seems like priority or niceness would need to get set on each
process that's spawned.  As far as I can see the same number would
work everywhere.  Or maybe have it take the number from an environment
variable so it could be changed easily during the run.

I always liked the ports system. it's always amazed me that all the
pieces fit together and it works.  I've rarely had trouble with it
except when I try to circumvent it by using a newer version of
something.  I guess in that case I should make the newer version into
a port.

-- 
Credit is the root of all evil.  - AB1JX

Re: Anyone experienced with 4G/LTE modems?

[Alan Corey]

re: Anyone experienced with 4G/LTE modems?

Comparing the man page and eBay I found a used Huawei 156g for $10
right off the bat.  This could be fun.  Need to make a spreadsheet or
something of specs.

The phone I'm using now is a Motorola Moto-e2 and the wifi hotspot
seems to work sometimes.  Some Androids turn off WiFi when they sleep.
Some are low powered to try to improve the battery life.  And I mostly
hate Android 5.

On 11/3/15, Chris Cappuccio  wrote:
> Alan Corey [alan01...@gmail.com] wrote:
>>
>> I have used USB tethering from a phone to an OpenBSD machine and
>> bridged that to a WiFi card set up as an AP.  I'm not impressed with
>> the reliability of WiFi after using it a few months.  It's convenient
>> when it works but I'd rather run wires then maybe hang WiFi APs off
>> them, like one at each end of the house.
>>
>
> The net80211 infrastructure and related drivers need help. Apply within.
>


-- 
Credit is the root of all evil.  - AB1JX

Re: Anyone experienced with 4G/LTE modems?

[Alan Corey]

re: Anyone experienced with 4G/LTE modems?

Froogle sees the RUT950 at $278 which is a little steep.  I was using
an $88 Xgody phone that mostly worked.  I used it for a couple months
and I think the prolonged use of the WiFi hotspot maybe overheated it.
The WiFi and GPS both stopped working after being intermittent.  Both
are sometimes built into the CPUs on those I've read.

I have used USB tethering from a phone to an OpenBSD machine and
bridged that to a WiFi card set up as an AP.  I'm not impressed with
the reliability of WiFi after using it a few months.  It's convenient
when it works but I'd rather run wires then maybe hang WiFi APs off
them, like one at each end of the house.

I didn't know about umsm, I'll try to find one of the supported
devices.  That's what I was looking for.

On 11/3/15, Tati Chevron  wrote:
> On Tue, Nov 03, 2015 at 03:31:07PM -0500, Alan Corey wrote:
>>What I'd like to get is something I can connect to an OpenBSD machine
>>over USB or go for one of the slightly better ones that offers an
>>ethernet connection.  Firewall it, connect it through my wired LAN,
>>run a WiFi AP for portable devices.  I'm on Straight Talk, so AT&T
>>towers, I'm looking for something unlocked.
>
> Have you looked at the 3G routers by teltonika?  Specifically the RUT950.
>
> They support operation in bridge mode, so you can basically disable all of
> the
> internal routing and firewalling, and dedicate an OpenBSD machine to that
> task.
>
> They have decent antennas, too.
>
> Unlike DSL routers, most 3G routers don't have an option to run in bridge
> mode.
>
> --
> Tati Chevron
> Perl and FORTRAN specialist.
> SWABSIT development and migration department.
> http://www.swabsit.com
>


-- 
Credit is the root of all evil.  - AB1JX

Anyone experienced with 4G/LTE modems?

[Alan Corey]

Anybody have good experiences with any of the currently available
4G/LTE modems that start around $30 on eBay, mostly by Huawei?  I
won't have a real internet connection for at least a year.  Right now
I'm using cell phones, connecting to them by WiFi or urndis over USB.
Even for cell phones I'm in a fringe area.  There's a local cell tower
2 miles away which is rumored to be LTE-only.  With my GSM-only phone
I typically see signals about -100 dbm, 0-4 bars.  I had another
(Chinese) phone that I think was doing 3G, but I fried the WiFi in it
and sent it back under warranty.   Mailed it to China over a month
ago, probably still stuck in US customs because a week ago they hadn't
gotten it yet.

What I'd like to get is something I can connect to an OpenBSD machine
over USB or go for one of the slightly better ones that offers an
ethernet connection.  Firewall it, connect it through my wired LAN,
run a WiFi AP for portable devices.  I'm on Straight Talk, so AT&T
towers, I'm looking for something unlocked.

Huawei e3372 ($30 range)

Ebay listing: 
http://www.ebay.com/itm/121801557242?_trksid=p2055119.m1438.l2649&ssPageName=STRK%3AMEBIDX%3AIT

specs:
http://consumer.huawei.com/en/mobile-broadband/dongles/tech-specs/e3372.htm


Aircard 763S ($60 range)

manual: 
http://www.downloads.netgear.com/files/aircard/4112126_AirCard%20763S%20User%20Guide_r1.pdf

typical eBay listing:
http://www.ebay.com/itm/Unlocked-763S-Sierra-Wireless-4G-LTE-100-Mbps-GSM-Mobile-Broadband-Hotspot-New/121363925255?_trksid=p2055119.c100022.m2048&_trkparms=ao%3D1%26asc%3D20140122153519%26meid%3Ddc84d9bb5d134b25ac5dbc8ec13f5bea%26pid%3D100022%26

These look better but I have seen one mention of them being unlocked
only until the next time you have to reset them.   Maybe just from one
seller.  The manual looks good, I've never had a problem with Netgear
stuff.

There are other models and prices.

-- 
Credit is the root of all evil.  - AB1JX

Re: dpb - wow!

[Alan Corey]

re: dpb - wow!

Yeah, well, if I started on 5.8 now I might have it running before 5.9
came out.  I started at 5.6, before I got it all downloaded and
running to clone 5.7 came out.  I've got that on my laptop and
installing on my desktop.  Gotta draw the line somewhere.  I download
through my cell phone, it's the best internet connection I've got.

On 11/2/15, Josh Grosse  wrote:
> On 2015-11-02 13:33, Alan Corey wrote:
>> I'd never tried it before but it's saving me a lot of babysitting to
>> start the next build compared to doing them one at a time.
>>
>> There's probably a way to do this but I'd never tried dpb because I
>> didn't have a list of pkgpaths to feed it.  I could query sqlports I
>> guess, but a command-line flag to pkg_info to have it give full
>> pkgpaths would be good.  My goal is to be able to make a list of
>> pkgpaths on a machine, save it, put in a new hard drive, do an OpenBSD
>> install, then run dpb on the saved list of pkgpaths.
>
> Hey, Alan.  I've been an end-user of dpb() for some years. It's the
> bees' knees. A couple of hints which may help:
>
> * out-of-date(1) produces pkgpath output, which I use with dpb -R for
>-stable package builds.
>
> * pkg_info(1) has a -P option, which along with -mq produces a nice list
> of
>manually installed pkgpaths.
>
>> And my hyperthreaded P4 now gets detected as MP?  Neat.  Just jumping
>> it from 5.0 to 5.7.
>
> 5.8 was released October 18.  :) :)
>


-- 
Credit is the root of all evil.  - AB1JX

dpb - wow!

[Alan Corey]

I'd never tried it before but it's saving me a lot of babysitting to
start the next build compared to doing them one at a time.

There's probably a way to do this but I'd never tried dpb because I
didn't have a list of pkgpaths to feed it.  I could query sqlports I
guess, but a command-line flag to pkg_info to have it give full
pkgpaths would be good.  My goal is to be able to make a list of
pkgpaths on a machine, save it, put in a new hard drive, do an OpenBSD
install, then run dpb on the saved list of pkgpaths.

And my hyperthreaded P4 now gets detected as MP?  Neat.  Just jumping
it from 5.0 to 5.7.

-- 
Credit is the root of all evil.  - AB1JX

Re: urtwn driver has problems

[Alan Corey]

Any chance of the stability fixes getting released as an official
patch?  The problems go back to at least 5.0.  I just mostly finished
a clean install of 5.7 on a new drive in my 5.2 machine.  I've got
another new 1 TB drive for my 5.0 machine.  The Firefox 5.0 in this is
getting ancient anyway.

urtwn driver has problems

[Alan Corey]

I don't know how to document this but my 5.7 laptop about once a week
has a hard crash when running ifconfig or dhclient on urtwn0.  I have
to hold down the power switch, turn it off, reboot.

My 5.0 machine with a different urtwn adapter is slightly worse, I
stopped using it and use an old Atheros card.  This would occasionally
also crash if I unplugged the adapter and plugged it back in.

Both give the "no buffer space available" message at times from the
driver.  Sometimes unplugging the adapter, putting it back and doing
ifconfig and dhclient over again fixes it, sometimes it crashes it.

Both machines are i386.  On a quick look I don't see anything in
/var/log/messages about it.

-- 
Credit is the root of all evil.  - AB1JX

Re: Why does my 5.7 laptop suspend when I close the lid?

[Alan Corey]

Yes it was harsh, but I still don't know how to resume and I didn't
appreciate having to do an unclean shutdown by holding down the power
button for 10 seconds.  Neither the apm or acpi man pages says how to
resume (or mentions lid), I guess it depends on the hardware.  I've
been using OpenBSD since 2.7 and I don't like surprises.  I close the
lids on my laptops to carry them or use them as a writing surface.
With my past few Dell laptops sound didn't work after suspend so I
stopped using it.  My newest is from 2008 so it didn't seem worth
mentioning.

I don't remember seeing the option to enable or disable in the
install, it reminded me of something Windows might do.

Yes, it's machdep.lidsuspend=0 to turn it off.  I thought there used
to be an apm.con or acpi.conf, I looked for those first.

On 6/30/15, Eric Furman  wrote:
> A lot of people worked very hard to add this "feature",
> because most people wanted it.
> Search the archives
>
> On Tue, Jun 30, 2015, at 02:38 PM, Alan Corey wrote:
>> I didn't ask it to do that and I don't know how to unsuspend.  As far
>> as I'm concerned this is an undocumented "feature".  If I want to
>> suspend I'll type zzz.  I haven't found a way to turn this off.
>>
>> --
>> Credit is the root of all evil.  - AB1JX
>>
>


-- 
Credit is the root of all evil.  - AB1JX

Why does my 5.7 laptop suspend when I close the lid?

[Alan Corey]

I didn't ask it to do that and I don't know how to unsuspend.  As far
as I'm concerned this is an undocumented "feature".  If I want to
suspend I'll type zzz.  I haven't found a way to turn this off.

-- 
Credit is the root of all evil.  - AB1JX

Re: Robustness in ports fetch program?

[Alan Corey]

Oh, I apparently can't do FTP, but that's a recent thing so I'm not
sure.  I'm using a cell phone data connection.

On 5/20/15, Alan Corey  wrote:
> I didn't override it because I didn't know how.  I've defined
> FETCH_CMD in the environment before but I've never messed with
> mk.conf.  I put in literally what you said, but  _PROGRESS and
> FTP_KEEPALIVE seem to be undefined.
>
> I'm experimenting now using wget but here's a run from trying to
> install qiv.  I was able to get firefox installed without much
> problem.
>
> ===>  Checking files for qiv-2.3.1
>>> Fetch http://spiegl.de/qiv//download/qiv-2.3.1.tgz
> Trying 2a03:2500:1:7:5054:ff:fe95:55b1...
> ftp: connect: No route to host
>>> Fetch http://ftp.openbsd.org/pub/OpenBSD/distfiles//qiv-2.3.1.tgz
> Trying 129.128.5.191...
> Requesting http://ftp.openbsd.org/pub/OpenBSD/distfiles//qiv-2.3.1.tgz
>
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  | 0   --:--
> ETA
>   0% |  

Re: Robustness in ports fetch program?

[Alan Corey]

ed -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 17% |  |   128 KB  - stalled -
 28% |**|   203 KB00:00
208751 bytes received in 0.00 seconds (12442.53 MB/s)
>> Size does not match for file-5.22.tar.gz
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2935
'/usr/ports/distfiles/file-5.22.tar.gz': @lock=file-...)
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2307 '_internal-fetch')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2556
'/usr/ports/pobj/libmagic-5.22/.extract_done')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:1940
'/usr/ports/packages/i386/all/libmagic-5.22.tgz')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2493 '_internal-package')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2473 'package')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:1957
'/var/db/pkg/libmagic-5.22/+CONTENTS')
*** Error 1 in /usr/ports/devel/libmagic
(/usr/ports/infrastructure/mk/bsd.port.mk:2473 'install')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2100
'/usr/ports/pobj/qiv-2.3.1/.dep-devel-libmagic')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2556
'/usr/ports/pobj/qiv-2.3.1/.extract_done')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:1940
'/usr/ports/packages/i386/all/qiv-2.3.1.tgz')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2493
'_internal-package')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:2473 'package')
*** Error 1 in . (/usr/ports/infrastructure/mk/bsd.port.mk:1957
'/var/db/pkg/qiv-2.3.1/+CONTENTS')
*** Error 1 in /usr/ports/graphics/qiv
(/usr/ports/infrastructure/mk/bsd.port.mk:2473 'install')


On 5/18/15, Marc Espie  wrote:
> On Sun, May 17, 2015 at 08:18:06AM -0400, Alan Corey wrote:
>> I don't think it did this back in 5.0 days or maybe earlier.  I started
>> with OpenBSD 2.7, I just usually attributed problems to being my fault.
>> And I've always used the ports tree, not packages. Distfiles are often
>> useful across OpenBSD versions, sometimes in FreeBSD, I've even built
>> some
>> under Linux.
>>
>> I didn't look at what FETCH_CMD was defined as by default, I just assumed
>> defining something non-null changed it.  I did notice that when it
>> retries
>> it's wrongly assumed there's a problem with the first source and gone to
>> another.
>>
>> Does every developer have perfect internet?  That's very frustrating,
>> maybe
>> counterproductive in testing.  Try a modem, you can probably find a free
>> one.  Connection interruptions and resets happen many times a day.
>> On May 17, 2015 1:22 AM, "Marc Espie"  wrote:
>
> Why are you ranting instead of providing the info I'm asking for ?!!!
>
> JUST OVERRIDE THE DAMN FETCH_CMD!!!
>
> put
> FETCH_CMD = /usr/bin/ftp -v ${_PROGRESS} -k ${FTP_KEEPALIVE} -C
>
> in /etc/mk.conf

Re: 5.7: size for ninja distfile?

[Alan Corey]

I seem to be getting a lot of size does not match errors (unusual)
leading to error 1. But when I up-arrow and replay the line it goes
past it.

On 5/18/15, Alan Corey  wrote:
> I kept getting "size does not match for ninja-1.5.3p0" and several of
> the alternate sources were giving errors like 404.
>
> I have a file size of 168829 for
> 3309498174411e02e7680ea8b470bb7d1d70bdb8.tar.gz and the archive tests
> OK with gunzip -t.  This size does match the distinfo, but there are
> notes in the makefile about a toolchain error.
>
> I tried downloading ninja-1.5.3.tar.gz from
> https://pypi.python.org/pypi/ninja but that didn't help.
>
> I finally installed the package instead.  Back to building cmake for
> firefox.
>
> --
> Credit is the root of all evil.  - AB1JX
>


-- 
Credit is the root of all evil.  - AB1JX

5.7: size for ninja distfile?

[Alan Corey]

I kept getting "size does not match for ninja-1.5.3p0" and several of
the alternate sources were giving errors like 404.

I have a file size of 168829 for
3309498174411e02e7680ea8b470bb7d1d70bdb8.tar.gz and the archive tests
OK with gunzip -t.  This size does match the distinfo, but there are
notes in the makefile about a toolchain error.

I tried downloading ninja-1.5.3.tar.gz from
https://pypi.python.org/pypi/ninja but that didn't help.

I finally installed the package instead.  Back to building cmake for firefox.

-- 
Credit is the root of all evil.  - AB1JX

Re: Robustness in ports fetch program?

[Alan Corey]

I don't think it did this back in 5.0 days or maybe earlier.  I started
with OpenBSD 2.7, I just usually attributed problems to being my fault.
And I've always used the ports tree, not packages. Distfiles are often
useful across OpenBSD versions, sometimes in FreeBSD, I've even built some
under Linux.

I didn't look at what FETCH_CMD was defined as by default, I just assumed
defining something non-null changed it.  I did notice that when it retries
it's wrongly assumed there's a problem with the first source and gone to
another.

Does every developer have perfect internet?  That's very frustrating, maybe
counterproductive in testing.  Try a modem, you can probably find a free
one.  Connection interruptions and resets happen many times a day.
On May 17, 2015 1:22 AM, "Marc Espie"  wrote:

> On Sat, May 16, 2015 at 10:31:24PM -0400, Alan Corey wrote:
> > I'd seen this happen in 5.6 too, but I just caught an example of it in
> > 5.7.  My connection leaves a lot to be desired, but there's nothing I
> > can do about that.  I normally have FETCH_CMD set to use wget once I
> > get it installed but this was in doing a standard make install of a
> > port.
> >
> > The first time the connection gets interrupted, but something thinks
> > it should be done and checks the size.  That's wrong so it downloads
> > it over again instead of just resuming the download.  It should only
> > download it over again if the size matches but the CRC is wrong.
> > Seems like anyway.
> >
> > ===>  Verifying install for tcl-8.5.16 in lang/tcl/8.5
> > ===>  Checking files for tcl-8.5.16p0
> > >> Fetch
> > >> http://downloads.sourceforge.net/sourceforge/tcl/tcl8.5.16-src.tar.gz
> > tcl8.5.16-src.tar.gz  60% |*|  2696 KB
> 00:00
> > >> Size does not match for tcl8.5.16-src.tar.gz
> > >> Fetch
> http://ftp.openbsd.org/pub/OpenBSD/distfiles//tcl8.5.16-src.tar.gz
> > tcl8.5.16-src.tar.gz  23% |**   |  1024 KB
> 00:03 ETA
>
> The problem lies in ftp(1).
>
> Logic in the ports tree is fine. But there's nothing it can do there:
> somehow
> your ftp returns 0 (e.g., success), so the partial file gets removed.
>
> If you want to get it fixed, you may have to provide more input, as we
> obviously do not see that problem... First thing would be to override
> FETCH_CMD to remove the -V, so that you can show us what ftp says about
> things.  Tracing the code thru the program would help.

Robustness in ports fetch program?

[Alan Corey]

I'd seen this happen in 5.6 too, but I just caught an example of it in
5.7.  My connection leaves a lot to be desired, but there's nothing I
can do about that.  I normally have FETCH_CMD set to use wget once I
get it installed but this was in doing a standard make install of a
port.

The first time the connection gets interrupted, but something thinks
it should be done and checks the size.  That's wrong so it downloads
it over again instead of just resuming the download.  It should only
download it over again if the size matches but the CRC is wrong.
Seems like anyway.

===>  Verifying install for tcl-8.5.16 in lang/tcl/8.5
===>  Checking files for tcl-8.5.16p0
>> Fetch
>> http://downloads.sourceforge.net/sourceforge/tcl/tcl8.5.16-src.tar.gz
tcl8.5.16-src.tar.gz  60% |*|  2696 KB00:00
>> Size does not match for tcl8.5.16-src.tar.gz
>> Fetch http://ftp.openbsd.org/pub/OpenBSD/distfiles//tcl8.5.16-src.tar.gz
tcl8.5.16-src.tar.gz  23% |**   |  1024 KB00:03 ETA

-- 
Credit is the root of all evil.  - AB1JX

Dealing with a mix of DHCP and fixed addresses

[Alan Corey]

I'm pretty sure I haven't seen this in the FAQ.

I don't like DHCP, I consider it useful in temporary situations only.
I've had a few machines on a LAN with fixed IPs for years and it all
works fine.

Along comes a cell phone and it becomes my internet gateway, at least
most of the time.  Typical Android anyway, I've seen 2 of them work
the same way.  When I turn on the WiFi hotspot it becomes
192.168.43.1.  It assigns IPs from a pool that includes 192.168.43.34,
192.168.43.72, 192.168.43.134

I am running Debian on this one but I'm not sure to what degree I can
circumnavigate the Android hotspot.  I could just turn it off I
suppose and do my sharing from within Linux.  If I turn on Android's
WiFi it expects to connect to an AP, which I could set up.  Trying to
bridge to a WiFi interface on an OpenBSD machine that's in a DHCP
arrangement gives an error.  The first machine to connect to the phone
always gets assigned 192.168.43.34 and the phone's IP (and gateway &
DNS) is always 192.168.43.1

I'd like to have some fixed IPs just because I want to be able to FTP
and SSH from one machine to another.  The phone's rooted, I suppose I
could try to find and modify its equivalent of dhcpd.conf.  There
isn't one but I can fiddle with how it's set up at least to some
degree.

WiFi seems the only reasonable way in and out of this thing, the USB
hardware doesn't support "host mode" and only works for some things.

So anyway, anybody else deal with this situation?

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: disabled voice and video in Pidgin?

[Alan Corey]

I've been working for months on 5.6 to clone, by dialup it's slow. I
only upgrade every few years.  Going to 1 TB drives.

I'm ab1jx on devio btw.

On 3/5/15, Jiri B  wrote:
> On Thu, Mar 05, 2015 at 03:49:33AM -0500, Alan Corey wrote:
>> This is under OpenBSD 5.2.  I built pidgin from ports, it worked, I
>> decided I want sound so I removed it and libpurple, set FLAVOR to
>> audio.  It built libpurple-audio and pidgin-audio and installed them
>> but when I go into Help -> Build Information it says audio and video
>> are disabled.  Why?
>
> ...and nobody would care about 5.2 here, time to upgrade?
>
> j.
>


-- 
Credit is the root of all evil.  - AB1JX

disabled voice and video in Pidgin?

[Alan Corey]

This is under OpenBSD 5.2.  I built pidgin from ports, it worked, I
decided I want sound so I removed it and libpurple, set FLAVOR to
audio.  It built libpurple-audio and pidgin-audio and installed them
but when I go into Help -> Build Information it says audio and video
are disabled.  Why?

I have pulseaudio, portaudio (apparently not jack at the moment)
installed, maybe it needs some codec?  Is there some dependency that
the port didn't build?
Can there be something in ~/.purple from my old installation that's
turning it off?  Since it's listed under build information it seems
more serious than that.  Sound works in other things.


Pidgin 2.10.6 (libpurple 2.10.6)
4cfe697ea3ae39a4fb3dad8e3ed1c70855901095

Build Information
  Arguments to ./configure:   '--disable-avahi' '--disable-cap'
'--disable-doxygen' '--disable-farstream' '--disable-gevolution'
'--disable-gnutls' '--disable-nm' '--disable-nss' '--disable-perl'
'--disable-tcl' '--disable-tk' '--disable-vv'
'--with-python=/usr/local/bin/python2.7' '--disable-schemas-install'
'--disable-gtkspell'
'--with-gconf-schema-file-dir=/usr/local/share/schemas/pidgin'
'--prefix=/usr/local' '--sysconfdir=/etc' '--mandir=/usr/local/man'
'--infodir=/usr/local/info' '--localstatedir=/var'
'--disable-silent-rules' 'CC=cc' 'CFLAGS=-I/usr/local/include'
'LDFLAGS=-L/usr/local/lib' 'CPPFLAGS=-I/usr/local/include'
  Print debugging messages: No
  Plugins: Enabled
  SSL: SSL support is present.

  Library Support
Cyrus SASL: Disabled
D-Bus: Enabled
Evolution Addressbook: Disabled
Gadu-Gadu library (libgadu): Enabled
GtkSpell: Disabled
OpenSSL: Enabled
GnuTLS: Disabled
GStreamer: Enabled
Mono: Disabled
NetworkManager: Disabled
Network Security Services (NSS): Disabled
Perl: Disabled
Tcl: Disabled
Tk: Disabled
UTF-8 DNS (IDN): Enabled
Voice and Video: Disabled
X Session Management: Enabled
XScreenSaver: Enabled
Zephyr library (libzephyr): Internal
Zephyr uses Kerberos: No

-- 
Credit is the root of all evil.  - AB1JX

Short path to making Android APKs under OpenBSD?

[Alan Corey]

Somewhere I read that Android now only officially supports their own
Android Studio.  Eclipse might work but needs to be newer than even
OpenBSD 5.6's version.  Gradle might work but seems to need Groovy,
there seems to be no end to the bloatware.  Android Studio is a 1 gig
download and wants 4 gigs of RAM to run.

Can't it be done with make or cmake?  This reminds me of the gobs of
software for etext publishing when an epub file is just a zip file
with a certain directory structure inside.  An APK file is just a zip
file.  I don't need easy and GUI, I can follow a checklist or write
something that does it for me.  I've written some Java but I'd rather
use c and maybe GTK.

-- 
Credit is the root of all evil.  - AB1JX

Re: Help needed: pkg_add dropps connections

[Alan Corey]

This is probably unrelated but I've noticed that the fetching that
happens with "make install" in ports seems less robust than it used to
be.  If my internet provider disconnects or the connection gets reset
beyond that, it doesn't resume the download.  And I've tried setting
FETCH_CMD to wget -c, it doesn't help much (in 5.6, that's what I have
my 5.2 machine set to).

So I do a make install, wait until I've got a working URL, then ctrl-c
to stop it, copy the url, open another rxvt in the distfiles dir, type
wget, paste the URL.  wget very rarely fails.

I've got portsql installed and was able to make myself some partial
fetchlists from that but my query didn't find dependencies of
dependencies.  A scratch install of 5.6 still took a couple months.

On 2/18/15, owner-m...@openbsd.org  wrote:

Credit is the root of all evil.  - AB1JX

Re: Using WiFi hotspots with OpenBSD?

[Alan Corey]

The advice I got works, in principle.  What's screwy is that it
doesn't work with my Atheros cards, either in my laptop or a desktop
machine.  I've got a couple of urtwn generic RTL8188 USB adapters and
those work fine.  One Atheros is an ath and one's an athn.  It's
almost like they're stuck in that proprietary Atheros mode, but they
do seem to recognize other APs.  Come to think of it they're
recognizing other Atheros cards.  If I do an ifconfig scan (on an
Atheros) it doesn't see my phone's AP.  If I scan with Kismet using
those interfaces I see the phone. Doing an ifconfig to set the nwid
then running dhclient on them doesn't work (to the phone).  I fairly
routinely connect with the athn in my laptop to an AP I'm running with
another athn.  They work, they just don't see my phone's AP.

-- 
Credit is the root of all evil.  - AB1JX

Using WiFi hotspots with OpenBSD?

[Alan Corey]

OK, I'm new to the concept of hotspots and why they're different from
regular APs.  I just got a phone that I can set to be a hotspot to
relay the phone's data stream over WiFi.  It works fine when I'm
booted into Windows, works like any other AP.  Under OpenBSD no luck
so far.  I can set the nwid, the bssid, the gateway (copied from
Windows) but when I run dhclient nothing.  It's like there's no DHCP
server out there, dhclient times out looking.  The phone is an Android
(4.4.2) so if I knew what I was doing I could look there.  It is
rooted and I could poke around in it.

-- 
Credit is the root of all evil.  - AB1JX

setting WiFi txpower with ifconfig

[Alan Corey]

I'm trying to do some antenna work so I want a weak signal from the
other side of the basement.  So I try stuff like ifconfig athn0
txpower 1 and get "ifconfig: SIOCS80211TXPOWER: Invalid argument".
Any number I've tried gives the same thing.  If I leave out the number
it tells me I need one.  Worse, the same thing happens with athn, ath,
urtwn, in OpenBSD 5.6, 5.2, 5.0.  Looking in ieee80211_ioctl.h the
power (dBm) value is an int16_t but there's also an i_mode int which
controls whether the value can be set manually or it's in automatic
mode. The ifconfig man page says:

txpower dBm
Set the transmit power.  The driver will disable any auto level
 and transmit power controls in this mode.

Implying that setting txpower should work.  It seems stuck in auto
mode or something.  I've Googled this and other people have had
similar problems for years but no solution seems to be posted.

Using -txpower which should (by the man page) put it back in auto mode
gives the same invalid argument error.  It seems to be something wrong
with ifconfig since it works the same way with 3 different cards, or
it's something wrong in the man page so I'm using it wrong.  I'm
probably the only one who wants to turn the power down.  I should
probably stick in a printf to see exactly what's being fed to the
ioctl. Or something in setiftxpower() in ifconfig.c

  Alan
-- 
Credit is the root of all evil.  - AB1JX

FAQ entry for printing?

[Alan Corey]

Could we have an FAQ entry for how to set up printing with lpr and/or
cups?  I had lpr working once years ago with a text printer.  Now I
want to print (mostly JPEGs) to an HP color laser printer (cp2025dn).
I've got a PPD file I found on the web (it's Postscript) for the
printer which has its own IP address.

I'm not using color management I don't think but the files I'm trying
to print are sRGB.  Under Windows 2000 they worked fine but XP added
color management so now a purple flower prints blue.  I'd like to
print under OpenBSD as an alternative.

If somebody's got this working and could describe the setup that would
be a good start.  I'm trying to do photo quality prints which I used
to be able to do under Windows 2000.

Thanks,

  Alan

-- 
Credit is the root of all evil.  - AB1JX

Re: Getting unswapped?

[Alan Corey]

Mostly so when I switch to a different application, maybe on a
different page of the FVWM desktop, it isn't sitting there swapped out
and it's responsive. I've usually got 20 or more applications open at
once (most just RXVT windows) and reboot about once a week.  If I
invest in RAM I expect it to get used.  Seems like Linux and FreeBSD
are better about this but I don't use them often. Now I've got 864
free, 25 swapped out (restarted Firefox).

On 5/27/14, Philip Guenther  wrote:
> On Mon, May 26, 2014 at 9:49 PM, Philip Guenther 
> wrote:
>
>> On Mon, May 26, 2014 at 9:36 PM, Alan Corey  wrote:
>>
>>> Several hours ago I edited a few big images in The Gimp so there was
>>> some swapping.  I still have about 60 megs swapped out even though
>>> I've got 600 megs of RAM free. I've seen this before, sometimes it'll
>>> stay swapped out overnight until I reboot to clear it. The Gimp was
>>> closed hours ago.
>>>
>>> Is there any command to cause the swap system to do a HUP or something
>>> to re-evaluate the situation?
>>>
>>
> [Stupid gmail control-enter]
>
> If the data has remained swapped out, it's because it hasn't been needed
> yet.  Perhaps its the process memory for a daemon which isn't being
> connected to and doesn't need to do anything.  Why would you *want* to swap
> that in?
>
>
> Philip Guenther
>


-- 
Credit is the root of all evil.  - AB1JX

Getting unswapped?

[Alan Corey]

Several hours ago I edited a few big images in The Gimp so there was
some swapping.  I still have about 60 megs swapped out even though
I've got 600 megs of RAM free. I've seen this before, sometimes it'll
stay swapped out overnight until I reboot to clear it. The Gimp was
closed hours ago.

Is there any command to cause the swap system to do a HUP or something
to re-evaluate the situation?

Thanks,

Alan

-- 
Credit is the root of all evil.  - AB1JX

Re: Ralink mystery usb mini WiFi adapter

[Alan Corey]

I didn't buy Ralink on purpose. I've had issues with other products
from them and generally prefer Atheros.

If you want, I'll stick it back in its padded envelope and send it to
you to experiment on. I think I'd like it back someday but if it won't
work under OpenBSD it's useless.  I hope to know by tomorrow if it
works under FreeBSD 10 on my Raspberry Pi but otherwise I could only
use it under Windows. Email me a snail mail address if you want it.

On 4/20/14, Stefan Sperling  wrote:
> On Sat, Apr 19, 2014 at 10:23:06PM -0400, Alan Corey wrote:
>> So it does need a different driver, it's not just a matter of tweaking
>> a device ID somewhere?
>
> Looking closer, it seems to be a run(4) variant.
> At least the vendor driver groups it with other run(4) devices.
>
> That doesn't mean it will work without modifications, though.
> It seems to need a different firmware at least. Whether or not
> it is backwards compatible to older devices is hard to tell
> without spending a lot of time digging around in the vendor
> sources... But there are other run devices we don't yet support
> without code changes:
> http://marc.info/?l=openbsd-tech&m=138903287819764&w=2h
>


-- 
Credit is the root of all evil.  - AB1JX

Re: Ralink mystery usb mini WiFi adapter

[Alan Corey]

On 4/19/14, Stefan Sperling  wrote:
> On Sat, Apr 19, 2014 at 02:12:23AM -0400, Alan Corey wrote:
>> I just got a cheap USB WiFi adapter that I thought was a Realtek for
>> some reason, turned out to be Ralink. I was interested in small
>> because I want to mount it at the focal point of a TVRO satellite
>> dish. If I'd known it was Ralink I wouldn't have bought it.
>>
>> So I plug it into my laptop running OpenBSD 5.2 and it just gets a
>> ugen in dmesg, nothing in ifconfig at all, no drivers attach to it.
>> ugen0 at uhub1 port 3 "MediaTek 802.11 n WLAN" rev 2.01/0.00 addr 2
>>
>> On a Linux box lsusb says
>> Bus 001 Device 071: ID 148f:7601 Ralink Technology, Corp.
>>
>> The page at Rakuten where I got it:
>> http://www.rakuten.com/prod/usb-mini-wifi-wireless-adapter-network-card-802-11n-150m/236363146.html
>> There's a UPC code on that page.
>>
>> Any new Ralink driver since 5.2? I couldn't find anything.  No
>> paperwork came with it, just a mini-cd with Windows, Mac and
>> supposedly Linux drivers for kernels 2.4 & 2.6 (ancient).  It seems to
>> use just "802 11N" as a model number.
>>
>> Any way to get it working or do I set it aside and wait for a driver
>> someday?  It was only $7.49 with free shipping. I've got a Realtek
>> RTL8188 coming from China.
>
> There is a GPL'd linux driver for it from Ralink.
> http://www.mediatek.com/en/downloads/mt7601u-usb/
> So it's possible that OpenBSD will support it at some point.
> Just keep it and wait.
>

So it does need a different driver, it's not just a matter of tweaking
a device ID somewhere?  Looks like it's not being recognized at the
USB layer, I've had to fudge device ids of scanners for SANE.

And yes, this is a generic kernel.  The standard 4 Ralink drivers are
present but ignore it.


-- 
Credit is the root of all evil.  - AB1JX

Ralink mystery usb mini WiFi adapter

[Alan Corey]

I just got a cheap USB WiFi adapter that I thought was a Realtek for
some reason, turned out to be Ralink. I was interested in small
because I want to mount it at the focal point of a TVRO satellite
dish. If I'd known it was Ralink I wouldn't have bought it.

So I plug it into my laptop running OpenBSD 5.2 and it just gets a
ugen in dmesg, nothing in ifconfig at all, no drivers attach to it.
ugen0 at uhub1 port 3 "MediaTek 802.11 n WLAN" rev 2.01/0.00 addr 2

On a Linux box lsusb says
Bus 001 Device 071: ID 148f:7601 Ralink Technology, Corp.

The page at Rakuten where I got it:
http://www.rakuten.com/prod/usb-mini-wifi-wireless-adapter-network-card-802-11n-150m/236363146.html
There's a UPC code on that page.

Any new Ralink driver since 5.2? I couldn't find anything.  No
paperwork came with it, just a mini-cd with Windows, Mac and
supposedly Linux drivers for kernels 2.4 & 2.6 (ancient).  It seems to
use just "802 11N" as a model number.

Any way to get it working or do I set it aside and wait for a driver
someday?  It was only $7.49 with free shipping. I've got a Realtek
RTL8188 coming from China.

  Alan
-- 
Credit is the root of all evil.  - AB1JX

xnecview broken?

[Alan Corey]

This should probably go to ports@ but I don't belong to that.

I'm running 5.2 as the latest, but 5.4 looks the same (has the same
setjmp patch).  There was an old problem with xnecview under OpenBSD
that caused it to crash if you tried to use it on more than about 6
frequencies, but this is new since 5.0.

Run necpp on any of the samples that come with it, then try to feed to
xnecview and you get a crash like this:

The program 'xnecview' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadMatch (invalid parameter attributes)'.
  (Details: serial 194 error_code 8 request_code 70 minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the --sync command line
   option to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)
XNECVIEW 1.35

#  freq.   Zr   Zi  SWR gain  f/b  phitheta
-

Run the example with a line like
nec2++ -i 36dip.nec -o 36dip.out
then xnecview:
xnecview 36dip.nec 36dip.out

It does it with my antennas too, works right under Linux.

  Alan

-- 
Credit is the root of all evil.  - AB1JX

Re: Deleted everything in /

[Alan Corey]

Got it.  Thanks.  I burned a 5.2 install and used the ramdisk
/usr/mdec/installboot from that. I don't have 5.5 and it would take
weeks by modem to get it.

On 3/6/14, Chris Cappuccio  wrote:
> Alan Corey [alan01...@gmail.com] wrote:
>> I'm at 5.2. Booting from a 5.4 install image I mounted my / as /mnt
>> then my /usr as /mnt2. Then I did:
>>
>> /mnt2/mdec/installboot -n -v /mnt/boot /mnt2/mdec/biosboot /dev/wd0c
>> and get: Bad system call
>>
>> There's a /mnt/boot in place copied from /mnt2/mdec
>>
>
> You need to run installboot from /usr/mdec (or /usr/sbin on 5.5)
> on the install image ramdisk, not the 5.2 host.
>
> And you really need to use the installer and let it do all
> this for you, or else you should read the install/upgrade scripts
> and figure out the stuffs.
>


-- 
Credit is the root of all evil.  - AB1JX

Re: Deleted everything in /

[Alan Corey]

I'm at 5.2. Booting from a 5.4 install image I mounted my / as /mnt
then my /usr as /mnt2. Then I did:

/mnt2/mdec/installboot -n -v /mnt/boot /mnt2/mdec/biosboot /dev/wd0c
and get: Bad system call

There's a /mnt/boot in place copied from /mnt2/mdec

On 3/5/14, Chris Cappuccio  wrote:
> Alan Corey [alan01...@gmail.com] wrote:
>> I wasn't too concerned until I discovered the machine wouldn't boot,
>> comes up to ERR M or something like that.  I put a kernel and a copy
>> of boot in place from /usr/mdec but it still comes up and just says
>> loading then nothing.  Do I need to do installboot?
>>
>
> Yeah
>
>> I was half asleep and did rm tempdir /* when it should have been rm
>> tempdir/*
>>
>
> You're better off just using the fs or iso images from the 55 snapshot
> directory and using the 'U'pgrade because you probably have a bunch
> of other broken/missing stuff as well.
>


-- 
Credit is the root of all evil.  - AB1JX

Deleted everything in /

[Alan Corey]

I wasn't too concerned until I discovered the machine wouldn't boot,
comes up to ERR M or something like that.  I put a kernel and a copy
of boot in place from /usr/mdec but it still comes up and just says
loading then nothing.  Do I need to do installboot?

I was half asleep and did rm tempdir /* when it should have been rm tempdir/*

I did this once years ago too.  All my stored FAQs are on the machine
I can't boot.

  Alan

-- 
Credit is the root of all evil.  - AB1JX

backdir: a simple little backup/versioning tool

[Alan Corey]

For about a year now when I'm working on some little program I make a
"backups" subdir and whenever I get around to it I copy the file I'm
working on in there with the date and time on the file as part of the
file name.  I finally wrote a little C program that does the same
thing.

It makes a "backups" subdir if there isn't one, then copies every
normal, non-executable file in the current directory using the last
modified date/time on the files as part of the filename.  file.c
becomes backups/file_2013-05-04_1548.c, etc.  If a file hasn't been
changed since last time it gets skipped.

Pure C, has no dependencies as far as I know, a 1.7 k tarball with
make file.  Tested under 4.7, 5.0, 5.2.  There's a page at
  http://ab1jx.webs.com/calcs/backdir/index.html

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: rtl_sdr (was: Re: CVS: cvs.openbsd.org: ports)

[Alan Corey]

> Date: Wed, 1 May 2013 09:25:52 + (UTC)
> From: Stuart Henderson 
> To: misc@openbsd.org
> Subject: Re: rtl_sdr (was: Re: CVS: cvs.openbsd.org: ports)
> Message-ID: 
>
> On 2013-04-29, noah pugsley  wrote:
>> So glad to see this. Receiving broadcast FM isn't even the half of it.
>>
>> From rtlsdr.org:
>>
>>
>>- FM: both narrow band and wideband. The former is used on two way
>> radio
>>systems such as emergency services and private radio networks (like
>>couriers and taxis) and UHF CB and the latter is the usual broadcast FM
>> the
>>likes of which you have in the kitchen and car. Aircraft and boats and
>>ships also use narrow band FM which you can use RTLSDR to listen to.
>> The
>>SDR# software can receive both narrowband and wideband FM and the
>> latter do
>>stereo FM too!
>>- AM: Most AM transmissions are below the bottom frequency of RTLSDR
>>dongles. You will need a translator (I'll deal with translators soon)
>> to
>>get these frequency bands.
>>- Upper/Lower Sideband (USB/LSB). See AM above.
>>- CW: Continuous wave for morse code enthusiasts.
>>- With GNURadio you can receive and demodulate digital modes such as
>>pagers (POCSAG), ADS-B (aircraft positions), AIS (ship positions), AP25
>> and
>>TETRA (digital trunk radio) and many others.
>>- GPS reception is currently being worked on but should be do-able.
>>- Satellite reception including receiving ham transmissions from the
>>International Space Station are possible to. I have seen some screen
>> shots
>>of someone using RTLSDR and a 2.5m dish to track the carrier signal on
>> deep
>>space robots such as Voyager and the Mars missions.
>>- This
>> postalludes
>> to the tuner chip being disabled and the RTL chip being used to
>>receive transmissions at 30MHz and below.
>>
>>
>> Of course getting samples is one thing, doing something with them is
>> another. Anybody working on a gnuradio port? :-(
>>
>>
>
> I've made a start at a gnuradio port, as has bentley@, the result of
> merging them together is in openbsd-wip, but I won't have much time
> to look at it further for a bit.
>
> https://github.com/jasperla/openbsd-wip/tree/master/comms/gnuradio

AM is used in the aircraft band from about 108 to about 144 MHz for
pilots talking to towers, for historical reasons I suppose.  Also the
CB band around 27 MHz is mostly AM, partly SSB.

I looked at your openbsd-wip stuff.  You've got gnuradio-companion in
your disabled list still, but you kinda need that.
qtgui also: both wxgui and qtgui

One big shock to me was that lxml has to be lxml from lxml.de
pyqwt is also not from ports

Libusb without async may be trouble too.

I've got my disabled list down to comedi, uhd, shd, fcd, which is all
hardware I don't have.  Everything builds, nothing works.

I think I've got Jack working, but I've got like 30 failures in make
test among the qa tests.  All of them are seqfaults.  I just checked
and my last post to discuss-gnuradio around April 15 never got any
answers.  There was a post since about tracking down segfaults but to
someone else.  Haven't gotten back to it.

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: open source laptop battery repair?

[Alan Corey]

> Date: Mon, 29 Apr 2013 02:32:28 +0200
> From: Paolo Aglialoro 
> To: misc@openbsd.org
> Subject: Re: open source laptop battery repair?
> Message-ID:
> 
>
> If the battery doesn't want(/show) to charge, sometimes a sleazy trick can
> do the job: detach it while pc is running on AC and reattach it, sometimes
> you need to do that more than once. Sometimes after this you gotta keep'em
> charging some hours and then try again (it all depends on batteries, this
> is more likely to work on oem equipment).

That's a very neat trick, especially since it worked on one of the
batteries.  I'd always been careful to shut down before changing
batteries.  It took a few times to get it out of the critical area,
then it went on to 100% charge.  I just popped it out and back in a
few times while plugged in.

One down, one to go, and two more in the mail to me.  At least I can
move it from the downstairs charger to the upstairs charger without
shutting down first.

Thank you ,

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: open source laptop battery repair?

[Alan Corey]

On 4/28/13, Riccardo Mottola  wrote:
> Hi Alan,
> Although the document contains interesting "internal details" about the
> hardware, I don't think these are very useful when dealing with broken
> batteries. Most of the communication is shielded by the APM/ACPI/other
> power manager. Being the author of BatMon for gnustep and owner of
> several laptops with different operating systems, I have some empirical
> experience.
> You may try to use http://gap.nongnu.org/batmon/ and depending on the
> OS/BIOS/Battery you might get a little interesting information

sysctl | grep bat shows me
hw.sensors.acpibat1.volt0=11.10 VDC (voltage)
hw.sensors.acpibat1.volt1=9.30 VDC (current voltage)
hw.sensors.acpibat1.current0=0.00 A (rate)
hw.sensors.acpibat1.amphour0=3.04 Ah (last full capacity)
hw.sensors.acpibat1.amphour1=0.43 Ah (warning capacity)
hw.sensors.acpibat1.amphour2=0.13 Ah (low capacity)
hw.sensors.acpibat1.amphour3=0.00 Ah (remaining capacity), CRITICAL
hw.sensors.acpibat1.raw0=2 (battery charging), OK
by reading from the acpi.

the hw.sensors.acpibat1.volt1 was at 10.97 2 days ago. That's the only
change I've seen.  This is on a CD/DVD bay battery, the other one is
not recognized at all since the Windows crash, so I took it back out.

What the be2works setup uses is an adapter designed by Philips to
connect from a USB or parallel port to a battery pack after you take
the battery pack out.  It talks to the gas gauge chip directly,
without acpi getting in the way.  smbus is just 2 wires (plus ground),
and you don't hurt anything if you get the clock and data backwards,
it just doesn't work.  The adapter is mostly a 74LS05 set of
inverters, but you could use a few transistors to do the same job if
you don't have a 74LS05.  Just level translation mostly.  The be2works
download includes a schematic of what to build, but it's a copy of
what Philips designed.

I've seen that 80% figure before, it's what lithium ion batteries go
to after about 400 charge/discharge cycles.  NiMH would be about
useless by then.  This is my 4th Dell laptop.  I've also got a C610
that takes the same battery as the CPIa before it, and I haven't
bought a new battery for that in several years.  It's down to the 80%
point but still working.  I don't think that's a smart battery, and
this one did come from eBay.

But why would Windows crashing because of a wrong driver for a WiFi
card kill a battery?  Neither Windows or Openbsd recognizes it, they
both say the battery is absent.  I think the internal fuse blew.

The CD/DVD bay battery is recognized, supposedly it's charging, but it
never gets anywhere.  After 2 days charging it still won't boot the
laptop.  And both of these batteries are less than 1 month old.  Or at
least I bought them that recently on eBay.  I don't know what's in
them exactly.  Could be old cells or cheap cells.

I don't believe in trying to rejuvenate a battery pack any more than a CRT.

There's some good information (and a lot of batteries) at
http://www.batteryspace.com/

If you look at what the be2works program does and what's in the TI
PDFs I think it's possible to write something that talks directly to a
gas gauge chip in (an unplugged) battery with an adapter.

  Alan
>
> I think that when a battery goes bad, everything which is problematic is
> inside the battery.
> I also do think that the powermanagment more than often does a bad job,
> but there isn't much you can do. I have batteries which sometimes do run
> for 1 hour or more, but the power manager reports them as dead (= little
> internal capacity).
>
> The chip inside tries to know: 1) the design capacity 2) the maximum
> capacity reached after the last charge 3) current capacity. Furthermore
> usually it tries to count the cycles
>
> 1) is always correct for original equipment batteries. I have seen
> "cheap oem" batteries with wrong values and it might be wrong if you
> susbstitute the elements inside with wrong
>
> 2) and 3) are what go wrong. Usually, you should see 2) slowly
> decreasing with each cycle. A "sane" but "old" battery drops about a
> small percent each time, so that perhaps after 400 cycles you are to 80%
> capacity. Nothing you can do about it!
>
> Sometimes the battery "thinks" it has a too low capacity. In this case,
> weird things happen. THee best case is that you will get like "5 minutes
> left" but actually your computer continues to operate. Most often you
> will get an abrupt drop and you rcomptuer goes off before giving a
> meaningful warning... or whatever else.
>
> Sometimes this might happen if just one of the elements goes bad. Laptop
> batteries do not have a balancing method, as far as I know.
>
> Anyway, the battery should recalibrate itself after a full discharge
> cycle (Li-Ion don't have memory, but their chip drifts). But I have
> noticed that more recent and smarter batteries essentially fail to do that.
>
> But essentially, other than trying to reset and force a recalibration
> with a tool from the host, I don't kn

Re: open source laptop battery repair?

[Alan Corey]

The repair I was talking about is mostly physically replacing cells
with new ones and making the pack work with the laptop again.  There's
enough difference between the prices of cells (lithium ion) and the
packs (Dell wants $110 for mine) to make that worthwhile.

But there are other repairs.  You can buy batteries on eBay cheap that
have been "refilled" only sometimes they aren't that great.  I'd like
to know what's going on inside.  Also other things: I installed a new
WiFi card and in the process of finding the right Windows driver for
it Windows bluescreened a few times.  Now one battery isn't recognized
at all, the other is permanently discharged and won't charge, either
under Windows or OpenBSD.  Something got written to an EEPROM
somewhere, unless something popped the thermal fuse inside the pack or
fried some hardware in the laptop.  The 2 batteries have different
symptoms.

Most of these schemes use an adapter with a few transistors or an IC
to connect to a disconnected battery pack over a USB or LPT port, they
don't disrupt what's happening in the computer.  The bus is related to
iic or i2c with slight differences. There are 2 wires (clock and data)
plus ground.

Try "sysctl | grep bat" (on a laptop).  How do the analog cell
voltages get converted to digital values you read on the screen?  It's
in the battery pack itself.  There's a TI "gas gauge" chip or similar
in the pack.  There's more at work than most people suspect, some are
even passworded.

See 
http://media.blackhat.com/bh-us-11/Miller/BH_US_11_Miller_Battery_Firmware_Public_WP.pdf
until it gets moved.

  Alan



On 4/28/13, Francois Pussault  wrote:
>
>
> Battery repair soft-tools are just fake or legend for laptops,
> because most of batteries are using LI-ON technology
>
> so when battery has been badly used (for example still in the machine while
> running 100% of time on sector for about a year),
> the battery is chemically modified inside of itself.
>
> so it can never be repaired, you can get a few more battery time for example
> a battery that take charge of 50% its nominal capacity, will go up back to
> 60 or 75%  for a short period of time (few month) (software tools will show
> 100% but reality will be 60 to 75%).
>
> but then will decrease it capacity very fast , faster by far, than if you
> didn't try to force its repair...
>
> & this is worst again, for older batteries technologies.
>
> hardware tools exist that can do the job with better results but they are
> not cheap.
>
>
>> 
>> From: Alan Corey 
>> Sent: Sun Apr 28 07:16:46 CEST 2013
>> To: 
>> Subject: open source laptop battery repair?
>>
>>
>> Just wondering if anyone knows about tools for laptop battery repair
>> that might run under OpenBSD.  The "smart batteries" have a
>> microprocessor that interfaces to the cells and talks to the cpu over
>> an smbus.  ACPI talks to that bus, but it can't help with broken
>> batteries or replacing cells.
>>
>> Google for a pdf called BH_US_11_Miller_Battery_Firmware_Public_WP.pdf
>> if you want to read more.  There's a semi-commercial program called
>> be2works designed for cell replacement and such, but the full version
>> is $300.  Is there anything open source?
>>
>> BTW: Miller's bibliography at the end of the pdf above is quite good.
>> I was able to download all the pdfs he mentions.  Most are from TI.
>> Unfortunately he was working with Apple hardware, I've got Dell.  But
>> he got in there with logic analyzers and the whole bit.
>>
>>   Alan
>> --
>> Credit is the root of all evil.  - AB1JX
>>
>
>
> Cordialement
> Francois Pussault
> 3701 - 8 rue Marcel Pagnol
> 31100 Toulouse
> France
> +33 6 17 230 820   +33 5 34 365 269
> fpussa...@contactoffice.fr
>


-- 
Credit is the root of all evil.  - AB1JX

open source laptop battery repair?

[Alan Corey]

Just wondering if anyone knows about tools for laptop battery repair
that might run under OpenBSD.  The "smart batteries" have a
microprocessor that interfaces to the cells and talks to the cpu over
an smbus.  ACPI talks to that bus, but it can't help with broken
batteries or replacing cells.

Google for a pdf called BH_US_11_Miller_Battery_Firmware_Public_WP.pdf
if you want to read more.  There's a semi-commercial program called
be2works designed for cell replacement and such, but the full version
is $300.  Is there anything open source?

BTW: Miller's bibliography at the end of the pdf above is quite good.
I was able to download all the pdfs he mentions.  Most are from TI.
Unfortunately he was working with Apple hardware, I've got Dell.  But
he got in there with logic analyzers and the whole bit.

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: mixing ports and non-ports programs

[Alan Corey]

On 4/14/13, Marc Espie  wrote:
> On Sat, Apr 13, 2013 at 11:40:09PM -0400, Alan Corey wrote:
>> I'm not sure how to look this up, and it doesn't seem to be in the FAQ.
>
> There does not seem to be a question in your message...

I'm just wordy, I'll get to it.

>
>> I frequently add stuff that isn't in ports by building from sources.
>> Sometimes this real world stuff needs newer versions of other things
>> than what's in ports.  But the port is already installed and has
>> things depending on it.  So I have to uninstall the dependencies,
>> uninstall the port, then install the newer version.  Sometimes I can
>> take the distfile from the uninstalled dependency and build it as a
>> generic tarball outside the ports system, sometimes it has too many
>> patches to make it worthwhile.
>
> sometimes there is an update in the works. sometimes some of those pesky
> other ports have to be fixed as well. It can get complicated.

The question is what's the best way to mix current stuff in, I wasn't
criticizing.  It's actually handy to do an install and be able to
reuse a fair percentage of distfiles.  Can't do that with packages.
There are things in my 5.2 box that use the same distfiles as my 4.7
box.

For the most part I don't mind old - there probably aren't many bugs
or it would have been replaced.  But fltk is one of those that
applications are mostly written for a certain version.  It's like jdk
or gcc that way, except you can only have 1 version installed at a
time.

>
>> For example: I've just gotten fldigi running under 5.0 and 5.2, but
>> the fltk 1.x in ports is too old to work.  I had to uninstall it,
>> which meant aqsis had to be uninstalled.  Turns out there's a newer
>> aqsis too but I haven't started on that yet.
>>
>> I'm not in a position to make updated ports because I don't run current.
>
> Why not ?  I mean you say you have a 5.0 installation and a 5.2 one. That
> sounds complicated enough already. So why not have a -current install ?
> it's not harder than 5.0/5.2.

I don't try to install new things on my 4.7 box often but it still
works fine.  The native gcc is 3.3.5 is the main reason.  It also has
gcc 4.2.4 from ports but it doesn't work quite as well that way.

>
> (besides 5.0 is no longer supported, so  you'll have to update that one
> anyways).
>

This is ridiculous.  A whole year and a half and it's been abandoned.
Look at how long FreeBSD or Debian supports their versions.  Back in
the early 90's I used to play with Slackware, downloading each version
onto floppies, bringing it home and installing it, usually just in
time to do it all over with the next version.  I didn't know how to do
much else with it, but I was learning.

Now I actually /use/ OpenBSD, every day, on 3-4 machines.  Consider
them production machines even though I'm retired.  I do experimental
things with the likes of Gnuradio and the Osmocom suite lately, not
the operating system.  I might replace an operating system once in the
3-5 year expected life of a hard drive.

I could understand if Microsoft stopped supporting Vista, because it
was so bad many places wouldn't even use it, but OpenBSD 5.0 isn't
that different from 5.2.
Some things don't work under 5.2, just as some things don't work under
5.0.  You fix bugs, you introduce new ones, it isn't always an
improvement from the user's perspective.  We used to have a policy of
never buying  a Windows version until the first service pack came out.

I ran -current once when I had a job with an internet connection, but
only to get some feature.  As soon as the next release came out I
wiped it clean and installed the release.

-

Once again we're off on a tangent and I never got an answer to my
question of how to mix ports and non-ports versions of things.
Something like a way to uninstall a port without having to uninstall
everything that depends on it.  Or replace a port from sources and
leave everything else in place.

  Alan

Credit is the root of all evil.  - AB1JX

mixing ports and non-ports programs

[Alan Corey]

I'm not sure how to look this up, and it doesn't seem to be in the FAQ.

I frequently add stuff that isn't in ports by building from sources.
Sometimes this real world stuff needs newer versions of other things
than what's in ports.  But the port is already installed and has
things depending on it.  So I have to uninstall the dependencies,
uninstall the port, then install the newer version.  Sometimes I can
take the distfile from the uninstalled dependency and build it as a
generic tarball outside the ports system, sometimes it has too many
patches to make it worthwhile.

For example: I've just gotten fldigi running under 5.0 and 5.2, but
the fltk 1.x in ports is too old to work.  I had to uninstall it,
which meant aqsis had to be uninstalled.  Turns out there's a newer
aqsis too but I haven't started on that yet.

I'm not in a position to make updated ports because I don't run current.

BTW it's easy to install fldigi right now.  Just download and unpack
it, cd into it, do setenv LIBS "-lossaudio -lexecinfo" then
./configure, gmake, gmake install.  Great program for hams, also does
wefax.  You need fltk 1.3.0 though.  The ports hamlib is old so if
you've got a recent radio you need to update that.  You also need
libexecinfo, libsamplerate, portaudio, xmlrpc.  I have gnuradio built
but not working yet: troubleshooting via the gnuradio list.

  Alan

-- 
Credit is the root of all evil.  - AB1JX

Re: i965 DRI crashes with 5.2

[Alan Corey]

Forget it.  I'm almost finished a clean install of 5.2 that's taken
about 2 months by modem.  I'll live with a reduced number of
screensavers.  I just wanted to mention that there is a problem.  I
don't have the bandwidth to fool around with current and replacing it
every week.

  Alan

On 3/21/13, Chris Cappuccio  wrote:
> Alan Corey [alan01...@gmail.com] wrote:
>>
>> Not much to go on probably, but anyone else seeing this?
>>
>
> OpenBSD 5.3-current (post 5.3 release) now supports the latest
> Intel XF86 driver with KMS. It's worth trying before you do
> look at much else. See the snapshots/i386 or snapshots/amd64
> directory. Do a full upgrade.
>


-- 
Credit is the root of all evil.  - AB1JX

i965 DRI crashes with 5.2

[Alan Corey]

I don't know that it didn't happen before 5.2.  I bought a used Dell
Latitude D530 to load 5.2 onto, and most things work except when I
installed xscreensaver I find that about 1/3 of the individual
screensaver programs crash.  The main program just says "no preview
available" but when I quit that there were dozens of .core files left
behind in the current directory.

Obviously having xscreensaver isn't a life or death issue but it's
kind of fun once in a while.  They terminate with a signal 11 (I
forget where I found that out, it's in my notes from yesterday).
There's also a longish error message on a graphics screen that points
to /usrX11R6/lib/modules/dri/i965_dri.so no matter which one crashes.

Not much to go on probably, but anyone else seeing this?

  Alan

-- 
Credit is the root of all evil.  - AB1JX

Re: automake 1.11.5: never mind

[Alan Corey]

Changing my .cshrc to define AUTOMAKE_VERSION 1.11 instead of 1.11.5
and rebooting cured the problem.  Not sure why since I don't have
1.11.0 installed.  I have 1.10.3p6, 1.11.5p1, 1.9.6p10.  Oh well.

End of story for now.

  Alan

> The distfile name is automake-1.11.5.tar.gz and pkg_info reports
> 1.11.5, but sqlports rejects it.

> In the ports tree there are 6 versions:
> d530# cd automake
> d530# ls
> 1.10 1.12 1.8  CVS  Makefile.inc
> 1.11 1.4  1.9  Makefile

>   Alan




On 3/4/13, Brad Smith  wrote:
> On Mon, Mar 04, 2013 at 01:04:51AM -0500, Alan Corey wrote:
>> I'm defining
>> setenv AUTOMAKE_VERSION 1.11.5
>> In my .cshrc, I don't know why exactly.
>
> The value should be 1.11 not 1.11.5.
>

-- 
Credit is the root of all evil.  - AB1JX



-- 
Credit is the root of all evil.  - AB1JX

Re: automake 1.11.5: never mind

[Alan Corey]

The distfile name is automake-1.11.5.tar.gz and pkg_info reports
1.11.5, but sqlports rejects it.

In the ports tree there are 6 versions:
d530# cd automake
d530# ls
1.10 1.12 1.8  CVS  Makefile.inc
1.11 1.4  1.9  Makefile

  Alan




On 3/4/13, Brad Smith  wrote:
> On Mon, Mar 04, 2013 at 01:04:51AM -0500, Alan Corey wrote:
>> I'm defining
>> setenv AUTOMAKE_VERSION 1.11.5
>> In my .cshrc, I don't know why exactly.
>
> The value should be 1.11 not 1.11.5.
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>
>


-- 
Credit is the root of all evil.  - AB1JX

Automake 1.11.5 IS an issue

[Alan Corey]

This is probably where it came from:

d530# pkg_info | grep automake
automake-1.10.3p6   GNU Standards-compliant Makefile generator
automake-1.11.5p1   GNU Standards-compliant Makefile generator
automake-1.9.6p10   GNU Standards-compliant Makefile generator

I was defining AUTOMAKE_VERSION to the latest one I had installed, and
pkg_info reports a 1.11.5p1.  I don't know why.

  Alan

-- 
Credit is the root of all evil.  - AB1JX

automake 1.11.5: never mind

[Alan Corey]

I'm defining
setenv AUTOMAKE_VERSION 1.11.5
In my .cshrc, I don't know why exactly.
-- 
Credit is the root of all evil.  - AB1JX

Broken dependency: automake 1.11.5 in sqlports

[Alan Corey]

While installing sqlports from ports under openbsd 5.2 I got

pass #2
+++ converters/ruby-json,rbx
+++ databases/db/v4,bootstrap,no_java,no_tcl
>> Broken dependency: devel/automake/1.11.5 non existent
+++ databases/ruby-activerecord,ruby19
Died at /usr/ports/databases/sqlports/files/mksqlitedb line 114.
*** Error code 1

Stop in /usr/ports/databases/sqlports (line 25 of Makefile).
*** Error code 1

Stop in /usr/ports/databases/sqlports (line 2496 of
/usr/ports/infrastructure/mk/bsd.port.mk).
*** Error code 1

Stop in /usr/ports/databases/sqlports (line 1718 of
/usr/ports/infrastructure/mk/bsd.port.mk).
*** Error code 1

Of course there isn't any automake 1.11.5.
Maybe this should go to po...@openbsd.org, but I'm not signed up there.

  Alan
-- 
Credit is the root of all evil.  - AB1JX

Re: Are pthreads hopeless in 5.0?

[Alan Corey]

I have bought official CDs, they haven't been all that useful.  Even
the stickers faded when I put them on my car.  I also supported by
regular Paypal donation when I was working.

I like to build from distfiles.  I'd like to see a set of DVDs that
had distfiles for everything in a particular OpenBSD version, so all
the dependencies were met.  As far as I know the distfiles are generic
enough so they work on multiple architectures.

I'd make a list, drive about 2 miles, download everything on my list,
come home and start building.  When I found some dependency missing it
went on the next day's list and I went on to something else.
Eventually I got all that working and thought of more stuff I wanted.
I have 515 ports by pkg_info, 552 on my laptop under 4.7, plus a lot
of things that aren't in the ports collection just built from sources,
like Fldigi, Gnuradio and snd.  There isn't a hamradio category like
in FreeBSD.  Just about any unix program with a decent configure
script and a lot without I've managed to get working.

I've been using OpenBSD since 2.7 and rarely download binaries, except
maybe Firefox.  I like OpenBSD because it's small and simple (compared
to FreeBSD or most Linuxes) and it doesn't insist on installing Gnome
or KDE or something else to run your life.  Fvwm is fine, typing
startx is fine, and don't tell me I can't log in as root if I want to.
 It doesn't try to work like Windows (or Ubuntu).  My first unix
experiences were with Data General and DEC unix in the 90s and this is
more like those than anything else I've tried.  It's about all I've
used since I retired a few years ago.

But occasionally I find a few rough edges like pthreads. I'd heard
there were pthreads issues and I knew they'd been replaced in newer
versions, I'm just wondering if I've run into some of those problems.
Most of the Osmocom group's programs like rtl_adsb, rtl_tcp don't work
quite right under OpenBSD even though they compile fine (once you get
rid of a -lrt here and there). There's probably a common cause and
pthreads could be it. 6 different programs, 6 different problems.
Very nice set of programs other than that.  Command line software
defined radio, originally for ARM machines like the Raspberry pi.
About 1% of the size of Gnuradio, which is nice in it's own way but a
little more bloatware than most people need or want.

  Alan



On 1/20/13, Rogier Krieger  wrote:
> On Sun, Jan 20, 2013 at 5:11 PM, Alan Corey  wrote:
>> I'm on dialup, that's the only thing available in this area.
>> Installing 5.0 took me about 3 months by the time I got everything the
>> way I wanted it.
>
> Assuming the 3 months is spent on fiddling with a system rather than
> downloading: It may pay off investing time in tools such as puppet or
> cfengine (available through ports) that allow you to re-install more
> easily, from a set of recipes/changes created by you.
>
> Additionally, there's features such as siteXX.tgz, as desribed in FAQ
> 4.15 [1]. I've found that rather helpful in bootstrapping a system.
>
> Regards,
>
> Rogier
>
>
> References
> 1. OpenBSD FAQ - 4.15 - How can I install a number of similar systems?
> http://www.openbsd.org/faq/faq4.html#Multiple
>


-- 
Credit is the root of all evil.  - AB1JX