Re: dovecot issues

2014-06-21 Thread Otto Moerbeek
On Fri, Jun 20, 2014 at 11:37:07PM -0500, Edgar Pettijohn wrote:

 On 06/20/2014 11:20 PM, Edgar Pettijohn wrote:
  On 06/20/2014 10:58 PM, Mike Bregg wrote:
  On 2014-06-20 21:14, Edgar Pettijohn wrote:
  On 06/20/2014 10:05 PM, Mike Bregg wrote:
  On 2014-06-20 20:48, Edgar Pettijohn wrote:
  I'm trying to setup dovecot with opensmtpd and I'm getting the
  following
  errors for dovecot.
 
  Jun 20 21:41:04 auth-worker(10932): Fatal:
  pool_system_realloc(268435456): Out of memory
  Jun 20 21:41:04 auth-worker(10932): Fatal: master:
  service(auth-worker):
  child 10932 returned error 83 (Out of memory (service auth-worker {
  vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
  environment to get core dump)
  Jun 20 21:41:04 auth: Error: auth worker: Aborted request: Worker
  process died unexpectedly
  Jun 20 21:41:10 auth-worker(12071): Fatal:
  pool_system_realloc(268435456): Out of memory
  Jun 20 21:41:10 auth: Error: auth worker: Aborted request: Worker
  process died unexpectedly
  Jun 20 21:41:10 auth-worker(12071): Fatal: master:
  service(auth-worker):
  child 12071 returned error 83 (Out of memory (service auth-worker {
  vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
  environment to get core dump)
  You could try increasing vsz_limit to something like 512 MB.  Do you
  have mailboxes with a large quantity of emails (as in thousands) that
  256 MB isn't enough?
  Mike
 
  I felt like that was a clue, but have no idea how to set vsz_limit,
  however there is no mail on the system.  I got those errors just testing
  imap login from mutt.
  Have a look through the config files in /etc/dovecot/conf.d
 
  Mike
 
  Yeah I found it hiding out in 10-master.conf upped it to 512MB as
  suggested with no luck.  Then just for fun I upped it to 5000M and still
  no luck.
 
  Edgar
 
 I just noticed in the log even after raising vsz_limit the error remains
 the same claiming vsz_limit=256MB

This is a know bug. If an unknown user tries to log in, the login code
goes into a loop expanding a buffer until it runs out of mem. The
cause of the bug is also known, but nobody (including myself) came up
with a diff yet.

The login process gets rsstarted, so there is no immediate functional
problem, apart from the log being filled. 

-Otto



Re: relayd url redirection

2014-06-21 Thread Antoine Jacoutot
On Fri, Jun 20, 2014 at 10:35:13PM -0400, Predrag Punosevac wrote:
 I am seriously reading realyd man pages for the first time in my life.
 Namely I am after url suffix redirection. I will try to explain little
 bit better.
 
 I have close to 10 virtual hosts running behind OpenBSD firewall which
 has a single public IP address. My boss insists that all virtual hosts
 be reachable on the port 80. Can relayd do url suffix redirection? What
 I mean is a map:
 
 www.autonlab.org   10.0.0.0 host
 www.autonlab.org/web1  10.0.0.1 host
 www.autonlab.org/web2  10.0.0.2 host
 www.autonlab.org/web3  10.0.0.3 host
 
 and so on. Until now I was using nginx as a proxy and port redirection

You can do that with nginx without the need for port redirection if I am not 
mistaken.
Using proxy_pass in your Location.

-- 
Antoine



mdoc(7) - .Os on other systems

2014-06-21 Thread Jan Stary
The mandatory .Os macro of an mdoc(7) manpage
makes the rendered manpage have OpenBSD Reference Manual
at the top center, and OpenBSD X.Y at the bottom.
As the mdoc(7) manual says,

Left unspecified, it defaults to the local
operating system version. This is the suggested form.

Viewing the same page on Linux with mandoc,
it renders General Commands Manual and Linux 3.2.0-4-amd64.

Now, vieweing the same manpage on Linux with groff -mdoc,
it renders BSD General Commands Manual at the top center
and BSD at the bottom.

I suppose there is a difference between how mandoc and groff
figure out the OS information to use for .Os; groff_mdoc(7) says

.Os [operating system] [release]
If the first parameter is empty, the default `' is used.
This may be overridden in the local configuration file, mdoc.local.

Could someone please eleborate on this difference between
mandoc and groff regarding how they get the OS information?
Of course my Linux users wonder why the manpage says BSD.

See an example page below.
Thank you.


Jan


.Dd June 19, 2014
.Dt SNDFILE-CMP 1
.Os
.Sh NAME
.Nm sndfile-cmp
.Nd compare two audio files
.Sh SYNOPSIS
.Nm sndfile-cmp
.Ar file1
.Ar file2
.Sh DESCRIPTION
.Nm
compares the audio data of two sound files.
For two files to compare as being the same, their channel counts, sample rate,
audio data lengths and actual audio data must match.
Other differences such as string metadata like song title, artist etc and their
presence or absence are ignored.
.Sh EXIT STATUS
.Bl -tag -width 0 -compact
.It 0
The audio data is the same
.It 1
The audio data differs
.El
.Sh SEE ALSO
.Lk http://www.mega-nerd.com/libsndfile
.Sh AUTHORS
.An Conrad Parker Aq Mt con...@metadecks.org
.An Erik de Castro Lopo Aq Mt er...@mega-nerd.com



Re: mdoc(7) - .Os on other systems

2014-06-21 Thread Jason McIntyre
On Sat, Jun 21, 2014 at 11:43:57AM +0200, Jan Stary wrote:
 The mandatory .Os macro of an mdoc(7) manpage
 makes the rendered manpage have OpenBSD Reference Manual
 at the top center, and OpenBSD X.Y at the bottom.
 As the mdoc(7) manual says,
 
   Left unspecified, it defaults to the local
   operating system version. This is the suggested form.
 
 Viewing the same page on Linux with mandoc,
 it renders General Commands Manual and Linux 3.2.0-4-amd64.
 
 Now, vieweing the same manpage on Linux with groff -mdoc,
 it renders BSD General Commands Manual at the top center
 and BSD at the bottom.
 
 I suppose there is a difference between how mandoc and groff
 figure out the OS information to use for .Os; groff_mdoc(7) says
 
   .Os [operating system] [release]
   If the first parameter is empty, the default `' is used.
   This may be overridden in the local configuration file, mdoc.local.
 
 Could someone please eleborate on this difference between
 mandoc and groff regarding how they get the OS information?
 Of course my Linux users wonder why the manpage says BSD.
 

mandoc uses the code in mdoc_validate.c (ingo will correct me if i'm
wrong) and it's commented thus:

Set the operating system by way of the `Os' macro.
* The order of precedence is:
* 1. the argument of the `Os' macro, unless empty
* 2. the -Ios=foo command line argument, if provided
* 3. -DOSNAME=\foo\, if provided during compilation
* 4. sysname release from uname(3)

groff uses the stuff in the tmac files in doc-common. it is not a pretty
sight, but you can see where it sets stuff like General Commands
Manual. i don;t have access to a linux system, but maybe groff classes
mdoc pages as BSD by default, or maybe your linux distro does.

jmc



Re: relayd url redirection

2014-06-21 Thread Sebastian Benoit
Predrag Punosevac(punoseva...@gmail.com) on 2014.06.20 22:35:13 -0400:
 I am seriously reading realyd man pages for the first time in my life.
 Namely I am after url suffix redirection. I will try to explain little
 bit better.

relayd cannot do this currently unfortunatly.

/Benno



Video card fan speed temperature

2014-06-21 Thread Julian Andrej
Hello,

is it possible to show/change the current fan speed, temperature and
clock of my video card? (ATI Radeon HD7770)

`sysctl hw` shows sensors of my cpu and north bridge which seem
correct. I'm using the 5.5 release.

I'm asking because i want to control the fan speed, which is much
lower with the proprietary driver for example on linux.

Regards



Re: Video card fan speed temperature

2014-06-21 Thread Jonathan Gray
On Sat, Jun 21, 2014 at 06:19:51PM +0200, Julian Andrej wrote:
 Hello,
 
 is it possible to show/change the current fan speed, temperature and
 clock of my video card? (ATI Radeon HD7770)
 
 `sysctl hw` shows sensors of my cpu and north bridge which seem
 correct. I'm using the 5.5 release.
 
 I'm asking because i want to control the fan speed, which is much
 lower with the proprietary driver for example on linux.

There is a temperature sensor on some models but it isn't exposed via
the sensors framework at the moment.

The code has a notion of profiles as mentioned here:
http://xorg.freedesktop.org/wiki/RadeonFeature/#index3h2

The choice of profile isn't exposed to userland currently.

Newer upstream versions of the radeon code default to dynamic
power control managed by the hardware (dpm) which apparently
helps reduce noise quite a bit but that will have to wait until
we have newer drm code.



System Hangs with Intel i7 3920XM

2014-06-21 Thread Gustav Fransson Nyvell

Hi,

I found something about this before, something old, would like to hear 
if there's some update.


Well, I've had 3 total freezes since I installed OpenBSD recently. This 
is on a laptop and I can't suspend by closing the screen even, that it 
will just black the screen. Everything is frozen, sound is stuck at 
repeat, fans are whizzing. And I'm running X11 and it's happened with 
just X11+Firefox and X11+Firefox+mplayer. It's not temperature; last 
time I'm sure it was around 65-68°C and it can handle 95°C for sustained 
time without hanging (hours of compiling firefox.) System was pretty 
idle, actually, this last time.


I want to ask 1) Is there something I can do? 2) Is there some logging I 
can enable that will dump stuff useful for reading after a hang? 3) Is 
this a kernel driver/module/something problem or caused by an 
application? (Applications should be able to do this, right?)


Thanks!

Well, I'll go on to list specifics here.

Intel 3920XM 2.9GHz Ivy Bridge with Intel HD Graphics 4000 GPU in the 
same chip. System has Optimus hardware GPU switching with an extra 
nVidia GeForce GTX 680M (unsupported, I know.)

32 GB RAM
120GB Intel SSD

dmesg
--
OpenBSD 5.5-current (GENERIC.MP) #4: Thu Jun 19 19:17:56 CEST 2014
gus...@uncouth.nyvell.se:/usr/src/sys/arch/amd64/compile/GENERIC.MP
real mem = 34226692096 (32641MB)
avail mem = 33306812416 (31763MB)
mpath0 at root
scsibus0 at mpath0: 256 targets
mainbus0 at root
bios0 at mainbus0: SMBIOS rev. 2.7 @ 0xeb420 (33 entries)
bios0: vendor American Megatrends Inc. version 4.6.5 date 08/22/2012
bios0: CLEVO P15xEMx
acpi0 at bios0: rev 2
acpi0: sleep states S0 S3 S4 S5
acpi0: tables DSDT FACP APIC FPDT ASF! MCFG HPET SSDT SSDT SSDT DMAR 
SSDT SSDT
acpi0: wakeup devices P0P1(S4) USB1(S3) USB2(S3) USB3(S3) USB4(S3) 
USB5(S3) USB6(S3) USB7(S3) PXSX(S4) RP01(S3) PXSX(S4) RP02(S4) PXSX(S4) 
RP03(S4) PXSX(S4) RP04(S4) [...]

acpitimer0 at acpi0: 3579545 Hz, 24 bits
acpimadt0 at acpi0 addr 0xfee0: PC-AT compat
cpu0 at mainbus0: apid 0 (boot processor)
cpu0: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2794.01 MHz
cpu0: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu0: 256KB 64b/line 8-way L2 cache
cpu0: smt 0, core 0, package 0
mtrr: Pentium Pro MTRR support, 10 var ranges, 88 fixed ranges
cpu0: apic clock running at 99MHz
cpu0: mwait min=64, max=64, C-substates=0.2.1.1.2, IBE
cpu1 at mainbus0: apid 2 (application processor)
cpu1: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2793.66 MHz
cpu1: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu1: 256KB 64b/line 8-way L2 cache
cpu1: smt 0, core 1, package 0
cpu2 at mainbus0: apid 4 (application processor)
cpu2: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2793.66 MHz
cpu2: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu2: 256KB 64b/line 8-way L2 cache
cpu2: smt 0, core 2, package 0
cpu3 at mainbus0: apid 6 (application processor)
cpu3: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2793.66 MHz
cpu3: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu3: 256KB 64b/line 8-way L2 cache
cpu3: smt 0, core 3, package 0
cpu4 at mainbus0: apid 1 (application processor)
cpu4: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2793.66 MHz
cpu4: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu4: 256KB 64b/line 8-way L2 cache
cpu4: smt 1, core 0, package 0
cpu5 at mainbus0: apid 3 (application processor)
cpu5: Intel(R) Core(TM) i7-3920XM CPU @ 2.90GHz, 2793.66 MHz
cpu5: 
FPU,VME,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,SSE3,PCLMUL,DTES64,MWAIT,DS-CPL,VMX,SMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,PCID,SSE4.1,SSE4.2,x2APIC,POPCNT,DEADLINE,AES,XSAVE,AVX,F16C,RDRAND,NXE,LONG,LAHF,PERF,ITSC,FSGSBASE,SMEP,ERMS

cpu5: 256KB 

Re: System Hangs with Intel i7 3920XM

2014-06-21 Thread Dimitris Papastamos
On Sat, Jun 21, 2014 at 07:35:22PM +0200, Gustav Fransson Nyvell wrote:
 I want to ask 1) Is there something I can do? 2) Is there some logging I can
 enable that will dump stuff useful for reading after a hang? 3) Is this a
 kernel driver/module/something problem or caused by an application?
 (Applications should be able to do this, right?)

I am no expert but have you tried disabling nvidia from the bios and booting
only with your intel graphics card?

does this happen without X?

have you tried the bsd.sp kernel?



Re: System Hangs with Intel i7 3920XM

2014-06-21 Thread Gustav Fransson Nyvell

On 06/21/14 19:55, Dimitris Papastamos wrote:

On Sat, Jun 21, 2014 at 07:35:22PM +0200, Gustav Fransson Nyvell wrote:

I want to ask 1) Is there something I can do? 2) Is there some logging I can
enable that will dump stuff useful for reading after a hang? 3) Is this a
kernel driver/module/something problem or caused by an application?
(Applications should be able to do this, right?)

I am no expert but have you tried disabling nvidia from the bios and booting
only with your intel graphics card?
No, AFAIK, this can't be done sadly. I could possibly remove it 
physically but I rather not.


does this happen without X?

Unknown, not in my experience, I'm always in X.


have you tried the bsd.sp kernel?

No.





--
This e-mail is confidential and may not be shared with anyone other than 
recipient(s) without written permission from sender.



Re: System Hangs with Intel i7 3920XM

2014-06-21 Thread Dimitris Papastamos
If you have any means of hooking up a serial cable (docking station?)
you might be able to see the ddb prompt and go from there.

Does the freeze happen instantly or does it slowly become unresponsive?
For the latter you might have time to switch to a VT and wait for ddb.
You can force a crash dump via 'boot dump' or 'boot crash'.



Re: LAN vs VLAN interface performance

2014-06-21 Thread Chris Cappuccio
Adam Thompson [athom...@athompso.net] wrote:
 Yes, OT... But unless you've chosen to do something silly (like enabling 
 MVRP, or blindly allowing all VLANs to an untrusted host) saying VLANs 
 aren't secure is about as useful as ICMP isn't secure.
 Please explain how VLANs are not secure when you have control of the devices 
 on both ends of an 802.1Q-tagged link?  That's no more or less secure than 
 having multiple links to a switch running un-tagged ports on different VLANs. 
  Or are you saying I should have a separate physical switch for each subnet?

When OpenBSD first impelemented if_vlan, not all vlan tagging switches
were created equal. I ported and improved if_vlan (we had it working in
an OS release _before_ FreeBSD did.) So I was paying attention to this
area fairly closely. Vlan hopping bugs with 802.1Q (and ISL) were not
entirely unheard of!

Certain switches had designs that paid attention to the MAC address
tables and didn't properly filter based on the vlan tag, even when
ports were set to only pass packets with a certain vlan tag.
This would depend on certain configuration options or combinations
of options.

Some would only exhibit this behavior under certain stress
situations. For instance, when the MAC address table was full, the
switch would basically turn into a sort of full-duplex hub.

This is well documented by security researchers who were proving these
bugs at the time. And this was some 14 years ago. If you're still using
a 14+ year old switch that hasn't failed by now, (even a nice, high-end
one) you are doing better than many others. Realize that these issues
were taken fairly seriously by vendors because vlans were being used
as a security mechanism. 

This situation is made more complicated because plenty of low-end
switches still exhibit odd behavior under stress situations. Don't
use them if you are using vlans for privacy. (If you are being paid
for co-location or ISP services, don't buy shit. You are being
paid for privacy.) Hopefully Broadcom and whoever else makes
switch ASICs has fixed their bugs by now, too.

Many networks use vlans for privacy very extensively these days.
Open source implementatons rule on 802.11 based wireless (Ubiquiti,
Mikrotik, Deliberant, and all the companies that use their
hardware in similar products with some Linux derivative kernel),
also there is a variety of Linux and BSD in firewall and router
gear. I hope most gear configures their vlan tagging kernel
interfaces and bridges correctly, configures their switch ASICs
or FPGAs correctly, and so on. Don't forget the NSA loves your
routers :)

If you want to test your gear, there are various tools
out there that you can test with. There are some MAC-
address-table flooding tools. You can also use if_vlan, arp -s
and ping to test for MAC address forwarding (for instance, will
a software bridge or switch send a packet to a MAC address
destination that in a different vlan that is learned on a
port configured to only pass traffic for a certain vlan.)

Basically anything that your users could possibly do, you can
test for. I'd be curious if anyone who has done this kind of
testing with recent hardware and software implementations can
share their results.

 
 I recall the performance question coming up at BSDCan, and Henning saying 
 that under ideal circumstances, yes, OpenBSD could usefully fill a 10GE pipe, 
 but that non-ideal circumstances would degrade performance faster than the 
 equivalent $250k+ hardware 10GE router.
 
 Reyk@ is, I believe, running an entire company predicated on high-performance 
 routing, including multiple-10GE but that's second-hand knowledge.
 
 Certainly OpenBSD is easily capable of pushing many gigabits per second of 
 traffic, I just don't know what the absolute upper limits are today.
 

Reyk's new company is Esdenera Networks. He hasn't released much in
the way of technical details but he did recently commit if_vxlan which
just might be one component of the Esdenera secret sauce :)

OpenBSD's networking capabilities are constantly being improved.

Some people are looking at improving OpenBSD's MP networking 
performance these days. Right now all routers and firewalls should
be on SP kernels or you will actually have worse performance.
There is a chance than this situation may improve in later
releases, a remote chance for 5.6.

Chris



Re: System Hangs with Intel i7 3920XM

2014-06-21 Thread Chris Cappuccio
Dimitris Papastamos [s...@2f30.org] wrote:
 If you have any means of hooking up a serial cable (docking station?)
 you might be able to see the ddb prompt and go from there.
 

He should be getting a ddb prompt even within X with inteldrm.

 Does the freeze happen instantly or does it slowly become unresponsive?
 For the latter you might have time to switch to a VT and wait for ddb.
 You can force a crash dump via 'boot dump' or 'boot crash'.

There are some known hangs in current for i386 (that might be fixed)
but amd64 is not known. 



Re: Video card fan speed temperature

2014-06-21 Thread Julian Andrej
Do you mind telling me where to find the power profile in the sources?
Or is it not possible at the moment even with changing and recompiling
the kernel?

On 21 June 2014 19:03, Jonathan Gray j...@jsg.id.au wrote:
 On Sat, Jun 21, 2014 at 06:19:51PM +0200, Julian Andrej wrote:
 Hello,

 is it possible to show/change the current fan speed, temperature and
 clock of my video card? (ATI Radeon HD7770)

 `sysctl hw` shows sensors of my cpu and north bridge which seem
 correct. I'm using the 5.5 release.

 I'm asking because i want to control the fan speed, which is much
 lower with the proprietary driver for example on linux.

 There is a temperature sensor on some models but it isn't exposed via
 the sensors framework at the moment.

 The code has a notion of profiles as mentioned here:
 http://xorg.freedesktop.org/wiki/RadeonFeature/#index3h2

 The choice of profile isn't exposed to userland currently.

 Newer upstream versions of the radeon code default to dynamic
 power control managed by the hardware (dpm) which apparently
 helps reduce noise quite a bit but that will have to wait until
 we have newer drm code.



Re: System Hangs with Intel i7 3920XM

2014-06-21 Thread Gustav Fransson Nyvell

On 06/21/14 20:02, Dimitris Papastamos wrote:

If you have any means of hooking up a serial cable (docking station?)
you might be able to see the ddb prompt and go from there.

Does the freeze happen instantly or does it slowly become unresponsive?
For the latter you might have time to switch to a VT and wait for ddb.
You can force a crash dump via 'boot dump' or 'boot crash'.
Sorry, no docking station or serial port. I could get one but it does 
freeze instantly so. And I've never seen a ddb prompt on screen.


--
This e-mail is confidential and may not be shared with anyone other than 
recipient(s) without written permission from sender.



Re: mdoc(7) - .Os on other systems

2014-06-21 Thread Ingo Schwarze
Hi Jan,

adding to what Jason already said...

Jan Stary wrote on Sat, Jun 21, 2014 at 11:43:57AM +0200:

 The mandatory .Os macro of an mdoc(7) manpage makes the rendered
 manpage have OpenBSD Reference Manual at the top center,

That is not true.

Try

  mandoc -Ios=FOOBAR /usr/share/man/man1/true.1

It still has OpenBSD Reference Manual at the top center.

What is shown at the top center is the optional volume argument
of the .Dt macro, see the mdoc(7) manual for details.

Try

  sed 's/Dt TRUE 1/Dt TRUE 1 FOOBAR/' /usr/share/man/man1/true.1 | mandoc

and

  sed 's/Dt TRUE 1/Dt TRUE LOCAL FOOBAR/' /usr/share/man/man1/true.1 | \
  groff -mdoc -Tascii -P-c

Groff only allows overriding the volume for non-standard sections,
otherwise the default volume name of the standard section takes precedence.
Mandoc always uses a volume name provided in .Dt and only falls back to
the section's default volume name when .Dt doesn't provide one.

 and OpenBSD X.Y at the bottom.

That is true.

 As the mdoc(7) manual says,
 
   Left unspecified, it defaults to the local
   operating system version. This is the suggested form.

Yes.  That is somewhat unfortunate but cannot easily be improved.

On the one hand, if you find old, dusty manuals somewhere, you
would like to know where they came from, so having an explicit
OpenBSD 5.5 inside each individual manual would be nice:
Even formatting them on Linux, you could still see, this is an
OpenBSD 5.5 manual.

However, that would require one commit to each individual manual
page for each release, which would be an unreasonable burden.

 Viewing the same page on Linux with mandoc,
 it renders General Commands Manual

In order to be closer (even though not identical) to what groff does,
we chose that as our default for mandoc-portable, see

  http://mdocml.bsd.lv/cgi-bin/cvsweb/msec.in?annotate=1.6cvsroot=mdocml

The OpenBSD msec.in is different, see

  /usr/src/usr.bin/mandoc/msec.in

 and Linux 3.2.0-4-amd64.
 
 Now, viewing the same manpage on Linux with groff -mdoc,
 it renders BSD General Commands Manual

As Jason already noted, that comes from tmac/doc-common
in the groff distribution.  In the -current groff git, we have:

  .ds doc-volume-operating-system BSD
  .ds doc-volume-ds-1 General Commands Manual
  .ds doc-volume-ds-2 System Calls Manual
  .ds doc-volume-ds-3 Library Functions Manual
  .ds doc-volume-ds-4 Kernel Interfaces Manual
  .ds doc-volume-ds-5 File Formats Manual
  .ds doc-volume-ds-6 Games Manual
  .ds doc-volume-ds-7 Miscellaneous Information Manual
  .ds doc-volume-ds-8 System Manager's Manual
  .ds doc-volume-ds-9 Kernel Developer's Manual

and then, inside the definition of .Dt,
simplifying the code a bit,

  .ds doc-volume \*[doc-volume-operating-system]
  .as doc-volume  \*[doc-volume-ds-\$2]

So you effectively end up with \*[doc-volume]
equal to BSD General Commands Manual.

 at the top center and BSD at the bottom.

That BSD is not coming form the same place as the other BSD.

Still in -current groff git, the file tmac/doc-common has:

  .ds doc-default-operating-system BSD

and then inside the definition of the .Os macro:

  .ie \$1 \
  .ds doc-operating-system \*[doc-default-operating-system]

and then the internal .doc-footer macro later
uses \*[doc-operating-system] for the page footer.

So, the packager for your Linux distribution should maybe
customize \*[doc-volume-operating-system]
and \*[doc-default-operating-system] via the mdoc.local
mechanism but didn't do so.

Oh well, i'm just relizing the names of these variables
in groff -current differ from the latest groff release, 1.22.2,
where they don't have the doc-* prefix yet.

In the OpenBSD groff-1.22.2 port, we are patching doc-common
like this:

  .ds volume-operating-system OpenBSD
  .ds default-operating-system OpenBSD\~ports

The former is for the page header, the latter for the page footer,
and the latter can be overridden by .Os.

Yours,
  Ingo



Re: Video card fan speed temperature

2014-06-21 Thread Jonathan Gray
You'd have to build a new kernel to modify the profile
at the moment.  I believe what we have currently is the profile
where the clocks are left untouched from what the bios sets them to
on boot.

In terms of code it is sys/dev/pci/drm/radeon/radeon_pm.c

adding something like the the following to the end of radeon_pm_init()
might be enough to set a different profile but I've not tested it.

rw_enter_write(rdev-pm.rwlock);
rdev-pm.profile = PM_PROFILE_LOW;
radeon_pm_update_profile(rdev);
radeon_pm_set_clocks(rdev);
rw_exit_write(rdev-pm.rwlock)

On Sat, Jun 21, 2014 at 08:08:33PM +0200, Julian Andrej wrote:
 Do you mind telling me where to find the power profile in the sources?
 Or is it not possible at the moment even with changing and recompiling
 the kernel?
 
 On 21 June 2014 19:03, Jonathan Gray j...@jsg.id.au wrote:
  On Sat, Jun 21, 2014 at 06:19:51PM +0200, Julian Andrej wrote:
  Hello,
 
  is it possible to show/change the current fan speed, temperature and
  clock of my video card? (ATI Radeon HD7770)
 
  `sysctl hw` shows sensors of my cpu and north bridge which seem
  correct. I'm using the 5.5 release.
 
  I'm asking because i want to control the fan speed, which is much
  lower with the proprietary driver for example on linux.
 
  There is a temperature sensor on some models but it isn't exposed via
  the sensors framework at the moment.
 
  The code has a notion of profiles as mentioned here:
  http://xorg.freedesktop.org/wiki/RadeonFeature/#index3h2
 
  The choice of profile isn't exposed to userland currently.
 
  Newer upstream versions of the radeon code default to dynamic
  power control managed by the hardware (dpm) which apparently
  helps reduce noise quite a bit but that will have to wait until
  we have newer drm code.



Re: dovecot issues

2014-06-21 Thread Edgar Pettijohn
On 06/21/2014 02:17 AM, Otto Moerbeek wrote:
 On Fri, Jun 20, 2014 at 11:37:07PM -0500, Edgar Pettijohn wrote:

 On 06/20/2014 11:20 PM, Edgar Pettijohn wrote:
 On 06/20/2014 10:58 PM, Mike Bregg wrote:
 On 2014-06-20 21:14, Edgar Pettijohn wrote:
 On 06/20/2014 10:05 PM, Mike Bregg wrote:
 On 2014-06-20 20:48, Edgar Pettijohn wrote:
 I'm trying to setup dovecot with opensmtpd and I'm getting the
 following
 errors for dovecot.

 Jun 20 21:41:04 auth-worker(10932): Fatal:
 pool_system_realloc(268435456): Out of memory
 Jun 20 21:41:04 auth-worker(10932): Fatal: master:
 service(auth-worker):
 child 10932 returned error 83 (Out of memory (service auth-worker {
 vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
 environment to get core dump)
 Jun 20 21:41:04 auth: Error: auth worker: Aborted request: Worker
 process died unexpectedly
 Jun 20 21:41:10 auth-worker(12071): Fatal:
 pool_system_realloc(268435456): Out of memory
 Jun 20 21:41:10 auth: Error: auth worker: Aborted request: Worker
 process died unexpectedly
 Jun 20 21:41:10 auth-worker(12071): Fatal: master:
 service(auth-worker):
 child 12071 returned error 83 (Out of memory (service auth-worker {
 vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
 environment to get core dump)
 You could try increasing vsz_limit to something like 512 MB.  Do you
 have mailboxes with a large quantity of emails (as in thousands) that
 256 MB isn't enough?
 Mike

 I felt like that was a clue, but have no idea how to set vsz_limit,
 however there is no mail on the system.  I got those errors just testing
 imap login from mutt.
 Have a look through the config files in /etc/dovecot/conf.d

 Mike

 Yeah I found it hiding out in 10-master.conf upped it to 512MB as
 suggested with no luck.  Then just for fun I upped it to 5000M and still
 no luck.

 Edgar

 I just noticed in the log even after raising vsz_limit the error remains
 the same claiming vsz_limit=256MB
 This is a know bug. If an unknown user tries to log in, the login code
 goes into a loop expanding a buffer until it runs out of mem. The
 cause of the bug is also known, but nobody (including myself) came up
 with a diff yet.

 The login process gets rsstarted, so there is no immediate functional
 problem, apart from the log being filled. 

   -Otto

Thanks for all the replies.  Not sure what I was doing wrong I was
trying to use bsdauth, but could never get it to let me login.  Switched
to MySql and its working.  No more memory errors in the log.



Re: relayd url redirection

2014-06-21 Thread vadimou
On 6/20/14, Predrag Punosevac punoseva...@gmail.com wrote:
 and so on. Until now I was using nginx as a proxy and port redirection

I agree with Antoine. Read the nginx manual carefully. All you need is there.



Re: dovecot issues

2014-06-21 Thread Brad Smith

On 21/06/14 3:21 PM, Edgar Pettijohn wrote:

On 06/21/2014 02:17 AM, Otto Moerbeek wrote:

On Fri, Jun 20, 2014 at 11:37:07PM -0500, Edgar Pettijohn wrote:


On 06/20/2014 11:20 PM, Edgar Pettijohn wrote:

On 06/20/2014 10:58 PM, Mike Bregg wrote:

On 2014-06-20 21:14, Edgar Pettijohn wrote:

On 06/20/2014 10:05 PM, Mike Bregg wrote:

On 2014-06-20 20:48, Edgar Pettijohn wrote:

I'm trying to setup dovecot with opensmtpd and I'm getting the
following
errors for dovecot.

Jun 20 21:41:04 auth-worker(10932): Fatal:
pool_system_realloc(268435456): Out of memory
Jun 20 21:41:04 auth-worker(10932): Fatal: master:
service(auth-worker):
child 10932 returned error 83 (Out of memory (service auth-worker {
vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
environment to get core dump)
Jun 20 21:41:04 auth: Error: auth worker: Aborted request: Worker
process died unexpectedly
Jun 20 21:41:10 auth-worker(12071): Fatal:
pool_system_realloc(268435456): Out of memory
Jun 20 21:41:10 auth: Error: auth worker: Aborted request: Worker
process died unexpectedly
Jun 20 21:41:10 auth-worker(12071): Fatal: master:
service(auth-worker):
child 12071 returned error 83 (Out of memory (service auth-worker {
vsz_limit=256 MB }, you may need to increase it) - set CORE_OUTOFMEM=1
environment to get core dump)

You could try increasing vsz_limit to something like 512 MB.  Do you

have mailboxes with a large quantity of emails (as in thousands) that
256 MB isn't enough?

Mike


I felt like that was a clue, but have no idea how to set vsz_limit,
however there is no mail on the system.  I got those errors just testing
imap login from mutt.

Have a look through the config files in /etc/dovecot/conf.d

Mike


Yeah I found it hiding out in 10-master.conf upped it to 512MB as
suggested with no luck.  Then just for fun I upped it to 5000M and still
no luck.

Edgar


I just noticed in the log even after raising vsz_limit the error remains
the same claiming vsz_limit=256MB

This is a know bug. If an unknown user tries to log in, the login code
goes into a loop expanding a buffer until it runs out of mem. The
cause of the bug is also known, but nobody (including myself) came up
with a diff yet.

The login process gets rsstarted, so there is no immediate functional
problem, apart from the log being filled.

-Otto


Thanks for all the replies.  Not sure what I was doing wrong I was
trying to use bsdauth, but could never get it to let me login.  Switched
to MySql and its working.  No more memory errors in the log.


Of course there are no more memory errors when you changed the
authentication mechanism. Otto mentioned to you the reason for the
errors.

--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.



Re: dovecot issues

2014-06-21 Thread Edgar Pettijohn
On 06/21/2014 06:37 PM, Brad Smith wrote:
 On 21/06/14 3:21 PM, Edgar Pettijohn wrote:
 On 06/21/2014 02:17 AM, Otto Moerbeek wrote:
 On Fri, Jun 20, 2014 at 11:37:07PM -0500, Edgar Pettijohn wrote:

 On 06/20/2014 11:20 PM, Edgar Pettijohn wrote:
 On 06/20/2014 10:58 PM, Mike Bregg wrote:
 On 2014-06-20 21:14, Edgar Pettijohn wrote:
 On 06/20/2014 10:05 PM, Mike Bregg wrote:
 On 2014-06-20 20:48, Edgar Pettijohn wrote:
 I'm trying to setup dovecot with opensmtpd and I'm getting the
 following
 errors for dovecot.

 Jun 20 21:41:04 auth-worker(10932): Fatal:
 pool_system_realloc(268435456): Out of memory
 Jun 20 21:41:04 auth-worker(10932): Fatal: master:
 service(auth-worker):
 child 10932 returned error 83 (Out of memory (service
 auth-worker {
 vsz_limit=256 MB }, you may need to increase it) - set
 CORE_OUTOFMEM=1
 environment to get core dump)
 Jun 20 21:41:04 auth: Error: auth worker: Aborted request: Worker
 process died unexpectedly
 Jun 20 21:41:10 auth-worker(12071): Fatal:
 pool_system_realloc(268435456): Out of memory
 Jun 20 21:41:10 auth: Error: auth worker: Aborted request: Worker
 process died unexpectedly
 Jun 20 21:41:10 auth-worker(12071): Fatal: master:
 service(auth-worker):
 child 12071 returned error 83 (Out of memory (service
 auth-worker {
 vsz_limit=256 MB }, you may need to increase it) - set
 CORE_OUTOFMEM=1
 environment to get core dump)
 You could try increasing vsz_limit to something like 512 MB. 
 Do you
 have mailboxes with a large quantity of emails (as in thousands)
 that
 256 MB isn't enough?
 Mike

 I felt like that was a clue, but have no idea how to set vsz_limit,
 however there is no mail on the system.  I got those errors just
 testing
 imap login from mutt.
 Have a look through the config files in /etc/dovecot/conf.d

 Mike

 Yeah I found it hiding out in 10-master.conf upped it to 512MB as
 suggested with no luck.  Then just for fun I upped it to 5000M and
 still
 no luck.

 Edgar

 I just noticed in the log even after raising vsz_limit the error
 remains
 the same claiming vsz_limit=256MB
 This is a know bug. If an unknown user tries to log in, the login code
 goes into a loop expanding a buffer until it runs out of mem. The
 cause of the bug is also known, but nobody (including myself) came up
 with a diff yet.

 The login process gets rsstarted, so there is no immediate functional
 problem, apart from the log being filled.

 -Otto

 Thanks for all the replies.  Not sure what I was doing wrong I was
 trying to use bsdauth, but could never get it to let me login.  Switched
 to MySql and its working.  No more memory errors in the log.

 Of course there are no more memory errors when you changed the
 authentication mechanism. Otto mentioned to you the reason for the
 errors.

And then I confirmed that that was the reason by switching to Mysql.



Re: OpenBSD GSoC 2014 accepted projects status

2014-06-21 Thread ian kremlin
hi!

i'm a student working on four DBus daemons that emulate the behavior of
systemd ones as to allow porting code that depends on systemd less of a
hassle

i've set up gitweb to track my progress, you can find it here:
https://uglyman.kremlin.cc/gitweb/gitweb.cgi?p=systemd-utl.git;

the 'master' branch should always compile with strict flags, while the
'devel' one is more of my own day-to-day sandbox.

very excited to be working with you all and hope to continue after GSoC
ends!
ian



Re: Slow performance on Radeon (HD7770) video card

2014-06-21 Thread Jonathan Gray
On Sat, Jun 21, 2014 at 10:32:55PM +0200, Julian Andrej wrote:
 Hello,
 
 i'm getting really low performance on my ATI Radeon HD7770 video card.
 glxgears runs at poor 27 fps and videos are stuttering (playback with
 mplayer and different -vo options).

We don't do acceleration on southern islands or newer Radeon
parts because it depends on LLVM, glamor and drm backed EGL.
This also requires the gbm part of Mesa which until very
recently has only supported Linux and udev/systemd.

Yes, even basic 2d acceleration requires this mess because
xf86-video-ati only has OpenGL backed glamor acceleration
for these parts, they didn't write any normal X style acceleration.