Re: 6.1 syspatch installed SP kernel on MP system

[Stephane HUC "PengouinBSD"]

Hi,

Please reboot your machine, and at the prompt bsd>,
specify 'boot bsd.mp' ...

As you see, with my station:
$ uname -a
OpenBSD *** 6.1 GENERIC.MP#2 amd64

;)


Le 05/03/17 à 03:00, Aaron Riekenberg a écrit :
> Have a 6.1 amd64 MP system running the generic MP kernel from the
> installation.  MP was automatically chosen by the installer - this is an
> Intel Atom 330 dual core box.
> 
> Ran syspatch today which installed 4 new patches:
> 
> $ syspatch -l
> 001_dhcpd
> 002_vmmfpu
> 003_libressl
> 004_softraid_concat
> 
> 
> One issue is, after installing patches with syspatch and rebooting I'm now
> running the SP kernel instead of MP:
> 
> $ uname -a
> OpenBSD server.localdomain 6.1 GENERIC#4 amd64
> 
> $ sysctl hw.ncpu
> hw.ncpu=1
> 
> 
> After the default installation - hw.ncpu was 2 and I was running the MP
> kernel.
> 
> Looking at rollback files in /var/syspatch (see below), it looks like 2
> patches updated the kernel - 002_vmmfpu and 004_softraid_concat.
> Interestingly 002_vmmfpu only seems to have updated /bsd (not /bsd.mp), but
> 004_softraid_concat updated both.
> 
> So I *think* if I moved /bsd.mp to /bsd manually I would have a fully
> patched MP kernel - since presumably the 004 kernel includes the 002
> patch.  Have not tried this yet.
> 
> This manual step seems like a bug.  Would imagine more people have MP
> machines than SP these days.
> 
> /var/syspatch$ tar ztvf ./61-002_vmmfpu/rollback.tgz
> -rw-r--r--  1 root wheel 10751939 Apr 22 05:26 ./bsd
> -r--r--r--  1 root bin   7321 Sep  4  2016
> ./usr/include/amd64/cpufunc.h
> -r--r--r--  1 root bin   1832 Mar 26  2015
> ./usr/include/amd64/fpu.h
> -r--r--r--  1 root bin  20593 Mar 23 12:58
> ./usr/include/amd64/vmmvar.h
> -r--r--r--  1 root bin   7321 Sep  4  2016
> ./usr/include/machine/cpufunc.h
> -r--r--r--  1 root bin   1832 Mar 26  2015
> ./usr/include/machine/fpu.h
> -r--r--r--  1 root bin  20593 Mar 23 12:58
> ./usr/include/machine/vmmvar.h
> 
> /var/syspatch$ tar ztvf ./61-004_softraid_concat/rollback.tgz
> -rw-r--r--  1 root wheel 10684150 May  2 19:34 ./bsd
> -rw-r--r--  1 root wheel 10756231 May  2 19:34 ./bsd.mp
> 
> $ ls -latrh /bsd*
> -rw-r--r--  1 root  wheel  10.2M Apr 22 05:26 /bsd.sp
> -rw-r--r--  1 root  wheel   9.0M Apr 22 05:26 /bsd.rd
> -rw-r--r--  1 root  wheel  10.2M May  2 19:35 /bsd
> -rw-r--r--  1 root  wheel  10.3M May  2 19:35 /bsd.mp
> 

-- 
~ " Fully Basic System Distinguish Life! " ~ " Libre as a BSD " +=<<<

Stephane HUC as PengouinBSD or CIOTBSD
b...@stephane-huc.net



signature.asc
Description: OpenPGP digital signature

6.1 syspatch installed SP kernel on MP system

[Aaron Riekenberg]

Have a 6.1 amd64 MP system running the generic MP kernel from the
installation.  MP was automatically chosen by the installer - this is an
Intel Atom 330 dual core box.

Ran syspatch today which installed 4 new patches:

$ syspatch -l
001_dhcpd
002_vmmfpu
003_libressl
004_softraid_concat


One issue is, after installing patches with syspatch and rebooting I'm now
running the SP kernel instead of MP:

$ uname -a
OpenBSD server.localdomain 6.1 GENERIC#4 amd64

$ sysctl hw.ncpu
hw.ncpu=1


After the default installation - hw.ncpu was 2 and I was running the MP
kernel.

Looking at rollback files in /var/syspatch (see below), it looks like 2
patches updated the kernel - 002_vmmfpu and 004_softraid_concat.
Interestingly 002_vmmfpu only seems to have updated /bsd (not /bsd.mp), but
004_softraid_concat updated both.

So I *think* if I moved /bsd.mp to /bsd manually I would have a fully
patched MP kernel - since presumably the 004 kernel includes the 002
patch.  Have not tried this yet.

This manual step seems like a bug.  Would imagine more people have MP
machines than SP these days.

/var/syspatch$ tar ztvf ./61-002_vmmfpu/rollback.tgz
-rw-r--r--  1 root wheel 10751939 Apr 22 05:26 ./bsd
-r--r--r--  1 root bin   7321 Sep  4  2016
./usr/include/amd64/cpufunc.h
-r--r--r--  1 root bin   1832 Mar 26  2015
./usr/include/amd64/fpu.h
-r--r--r--  1 root bin  20593 Mar 23 12:58
./usr/include/amd64/vmmvar.h
-r--r--r--  1 root bin   7321 Sep  4  2016
./usr/include/machine/cpufunc.h
-r--r--r--  1 root bin   1832 Mar 26  2015
./usr/include/machine/fpu.h
-r--r--r--  1 root bin  20593 Mar 23 12:58
./usr/include/machine/vmmvar.h

/var/syspatch$ tar ztvf ./61-004_softraid_concat/rollback.tgz
-rw-r--r--  1 root wheel 10684150 May  2 19:34 ./bsd
-rw-r--r--  1 root wheel 10756231 May  2 19:34 ./bsd.mp

$ ls -latrh /bsd*
-rw-r--r--  1 root  wheel  10.2M Apr 22 05:26 /bsd.sp
-rw-r--r--  1 root  wheel   9.0M Apr 22 05:26 /bsd.rd
-rw-r--r--  1 root  wheel  10.2M May  2 19:35 /bsd
-rw-r--r--  1 root  wheel  10.3M May  2 19:35 /bsd.mp

Re: 82599ES support

[Evgeniy Sudyr]

I'm using AOC-STGN-i2S and can confirm it works perfect with ix(4)

On Tue, May 2, 2017 at 11:13 PM, Zeljko Jovanovic
 wrote:
> On 02.05.2017. 19:57, Lyndon Nerenberg wrote:
>
>> We're looking to buy some 10-gig SFP+ boards, and are eyeing up
>> Supermicro's 2-port boards (listed as the 'Intel 82599ES - AOC-STGN-i2S').
>>
>> ix(4) doesn't list the ES variant of the chip, and a quick grep through
>> the driver source doesn't mention it explicitly, either.  Are any of you
>> running this board under >= 6.0 ?
>>
>> (We need to buy these boards as part of a single lease, so I'm constrained
>> on what's available.  Otherwise I'd just buy some X520s.)
>
>
>
> 82599ES is the the "standard" two-interface chip, which is used in Intel
> X520 and many other boards.
>
> 82599EB is also a two-interface chip, but with XAUI interface only (no 10
> Gb/s serial capability for directly connecting to SFP+). I didn't see any
> boards which use it, but it would be suitable for XENPAK or X2 transceivers.
> So 82599ES is essentialy 82599EB + serdes.
>
> There is also 82599EN, which is a single port 82599ES.
>
>
> These data can be found in the 82599 datasheet. I am familiar with them,
> because five years ago I worked on a 82599ES board design, but sadly my
> company didn't mananage to finish it on time. There was suddenly more and
> more boards on the market, the prices dropped, and we abandoned the
> almost-completed project.
>



-- 
--
With regards,
Eugene Sudyr

Re: Ubuntu in vmd and avx

[Mike Larkin]

On Tue, May 02, 2017 at 10:31:41PM +0200, Karl Pettersson wrote:
> On Mon, Apr 24, 2017 at 11:56:25AM -0700, Mike Larkin wrote:
> > On Mon, Apr 24, 2017 at 11:50:07AM -0700, Mike Larkin wrote:
> > > On Mon, Apr 24, 2017 at 08:18:17PM +0200, Karl Pettersson wrote:
> > > > I run 64-bit Ubuntu as a vmm guest, according to:
> > > > https://gist.github.com/reyk/6d369c5c0bd0c76f4906f83933f3bb71
> > > > 
> > > > It works well for the most part; I have removed cloud-init becaue I do
> > > > not need it. However, I used Python 2.7 with numpy 1.12 (installed via 
> > > > Pip),
> > > > and experienced crashes which seemed to be related to AVX2
> > > > instructions, even though the cpu should support this, according to
> > > > /proc/cpuinfo. (cf. https://github.com/numpy/numpy/issues/8128). For
> > > > example, when I tried to reproduce this simple chart,
> > > > https://matplotlib.org/examples/lines_bars_and_markers/fill_demo.html,
> > > > this made Python exit with illegal instruction, according to gdb:
> > > > 
> > > > Program received signal SIGILL, Illegal instruction.  0x73ce5d56
> > > > in UBYTE_less_equal_avx2 (args=0x7fffba50, dimensions= > > > out>, steps=, __NPY_UNUSED_TAGGEDfunc=) at
> > > > numpy/core/src/umath/loops.c.src:899
> > > > 
> > > > Downgrading to numpy 1.11.1~rc1 (the version available from as an Ubuntu
> > > > package) seems to have fixed the issue.
> > > >
> > > 
> > > we filter out avx2 from the guest, for now.  
> > > 
> > 
> > actually after looking at this it appears we don't, and we should. the 
> > filter
> > mask is only for avx (and not the later ones). I'll add suppression of avx2,
> > that may force your software to downgrade to something we do allow. I'll 
> > commit
> > that tonight.
> >
> 
> Was this commit ever made? I only find references to AVX in the CVS log
> for vmm.c.
> 

Thanks for reminding me, I must have fatfingered 'd'elete on that email.

I'll get to this tonight.

-ml

Re: 82599ES support

[Zeljko Jovanovic]

On 02.05.2017. 19:57, Lyndon Nerenberg wrote:

We're looking to buy some 10-gig SFP+ boards, and are eyeing up Supermicro's 
2-port boards (listed as the 'Intel 82599ES - AOC-STGN-i2S').


ix(4) doesn't list the ES variant of the chip, and a quick grep through the 
driver source doesn't mention it explicitly, either.  Are any of you running 
this board under >= 6.0 ?


(We need to buy these boards as part of a single lease, so I'm constrained on 
what's available.  Otherwise I'd just buy some X520s.)



82599ES is the the "standard" two-interface chip, which is used in Intel X520 
and many other boards.


82599EB is also a two-interface chip, but with XAUI interface only (no 10 Gb/s 
serial capability for directly connecting to SFP+). I didn't see any boards 
which use it, but it would be suitable for XENPAK or X2 transceivers. So 82599ES 
is essentialy 82599EB + serdes.


There is also 82599EN, which is a single port 82599ES.


These data can be found in the 82599 datasheet. I am familiar with them, because 
five years ago I worked on a 82599ES board design, but sadly my company didn't 
mananage to finish it on time. There was suddenly more and more boards on the 
market, the prices dropped, and we abandoned the almost-completed project.

Re: Ubuntu in vmd and avx

[Karl Pettersson]

On Mon, Apr 24, 2017 at 11:56:25AM -0700, Mike Larkin wrote:
> On Mon, Apr 24, 2017 at 11:50:07AM -0700, Mike Larkin wrote:
> > On Mon, Apr 24, 2017 at 08:18:17PM +0200, Karl Pettersson wrote:
> > > I run 64-bit Ubuntu as a vmm guest, according to:
> > > https://gist.github.com/reyk/6d369c5c0bd0c76f4906f83933f3bb71
> > > 
> > > It works well for the most part; I have removed cloud-init becaue I do
> > > not need it. However, I used Python 2.7 with numpy 1.12 (installed via 
> > > Pip),
> > > and experienced crashes which seemed to be related to AVX2
> > > instructions, even though the cpu should support this, according to
> > > /proc/cpuinfo. (cf. https://github.com/numpy/numpy/issues/8128). For
> > > example, when I tried to reproduce this simple chart,
> > > https://matplotlib.org/examples/lines_bars_and_markers/fill_demo.html,
> > > this made Python exit with illegal instruction, according to gdb:
> > > 
> > > Program received signal SIGILL, Illegal instruction.  0x73ce5d56
> > > in UBYTE_less_equal_avx2 (args=0x7fffba50, dimensions= > > out>, steps=, __NPY_UNUSED_TAGGEDfunc=) at
> > > numpy/core/src/umath/loops.c.src:899
> > > 
> > > Downgrading to numpy 1.11.1~rc1 (the version available from as an Ubuntu
> > > package) seems to have fixed the issue.
> > >
> > 
> > we filter out avx2 from the guest, for now.  
> > 
> 
> actually after looking at this it appears we don't, and we should. the filter
> mask is only for avx (and not the later ones). I'll add suppression of avx2,
> that may force your software to downgrade to something we do allow. I'll 
> commit
> that tonight.
>

Was this commit ever made? I only find references to AVX in the CVS log
for vmm.c.

/etc/rc.d/nsd start/stop/restart fails in nsd.conf does not enable "Remote Control"

[Peter Fraser]

/etc/rc.d/nsd uses nsd-control to start/stop/restart nsd.

nsd.conf tells you that "Remote Control" is by default disabled.
It would be nice if some part of the documentation pointed out that it must be 
enabled.

82599ES support

[Lyndon Nerenberg]

We're looking to buy some 10-gig SFP+ boards, and are eyeing up 
Supermicro's 2-port boards (listed as the 'Intel 82599ES - AOC-STGN-i2S').


ix(4) doesn't list the ES variant of the chip, and a quick grep through 
the driver source doesn't mention it explicitly, either.  Are any of you 
running this board under >= 6.0 ?


(We need to buy these boards as part of a single lease, so I'm constrained 
on what's available.  Otherwise I'd just buy some X520s.)


--lyndon

Intel's Management Technology is indeed vulnerable

[I love OpenBSD]

INTEL-SA-00075
There is an escalation of privilege vulnerability in Intel® Active Management 
Technology (AMT), Intel® Standard Manageability (ISM), and Intel® Small 
Business Technology versions firmware versions 6.x, 7.x, 8.x 9.x, 10.x, 11.0, 
11.5, and 11.6 that can allow an unprivileged attacker to gain control of the 
manageability features provided by these products.

Can I preview a bitlink before clicking on it?
https://support.bitly.com/hc/en-us/articles/230650447-Can-I-preview-a-bitlink-before-clicking-on-it-

Arstechnica:
http://bit.ly/2qyHCQn

Semiaccurate:
http://bit.ly/2pB2MjO

Intel's PDF:
http://intel.ly/2qAK4G0

Re: Kernel panic on Dell R210 with OpenBSD 6.0 (relayd related ?)

[Stuart Henderson]

On 2017-05-02, Mathieu BLANC  wrote:
> On Wed, Mar 29, 2017 at 02:06:23PM +0200, Mathieu BLANC wrote:
>> It also kernel panics with just this pf rules :
>> # cat pf_minimal.conf 
>> set limit { states 10 }  
>> set skip on lo   
>> anchor "relayd/*"
>> pass 
>> 
>
> I upgraded the system to 6.1 release last week, the kernel panic is still here
> (with the same logs).

Probably the best thing to do at this point is to write a mail to bugs@:

1. describe what the machine is doing in detail. carp? ipsec? pfsync?
what sort of relays? include config (sanitized if necessary, but do that
consistently).

2. copy in the panic message and stack trace as text (re-type it,
don't attach a picture or send a link to a picture).

3. make it a self-contained report with description etc all in the one
message, don't rely on people having message history.

4. include dmesg.

Re: What is "activity" on a relayed SSH connection? (Re: (solved) relayd splice timeout)

[Markus Rosjat]

Am 02.05.2017 um 14:50 schrieb Andreas Kusalananda Kähäri:

On Tue, May 02, 2017 at 12:35:22PM +0200, Markus Rosjat wrote:

Hi Hiltjo,

just to let you know seesion timeout did the try and works like a charm



On a related note:  What constitutes "activity" on a relayed SSH
connection?

I'm also using relayd on a router to relay SSH connections to an
internal host, and it seems as if "relactl show sessions" will never say
that "idle" is anything other than the same as "age".

$ relayctl show sessions
session 0:447 192.168.1.4:63327 -> 10.0.0.10:22 RUNNING
age 01:13:47, idle 01:13:47, relay 1, pid 84257


Regards,
Kusalananda


I also noticed that the timeout seems to be fixed that even I do traffic 
over the relayed connection a timeout appears after the amound of 
seconds I defined in my relayd.conf

There  not reset in any way, like Andreas stated above.

but thats a thing I could live with I just need a big enough value :)

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

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

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

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

Re: Kernel panic on Dell R210 with OpenBSD 6.0 (relayd related ?)

[Mathieu BLANC]

On Tue, May 02, 2017 at 03:44:43PM +0200, Andre Ruppert wrote:
> Hi,
> 
> Im running 6.0 amd64 on a pair of R210 with relayd, but these are R210 (II).
> 
> No kernel panics at all, and these systems are working in a live
> environment...
> 
> Regards
> Andre

Hi,

Yes, i have also several OpenBSD on R210 + 6.0 (or 6.1) + relayd and it works
like a charm. 

The only problem appeared when an admin did a REJECT (iptables) on one on the
host checked by relayd with check tcp (i tried to put all the details i could
in the previous mails).

The next step is to try with current (until now i've waited for the 6.1 release
which was very close to be released).

-- 
Mathieu

Re: Kernel panic on Dell R210 with OpenBSD 6.0 (relayd related ?)

[Andre Ruppert]

Hi,

Im running 6.0 amd64 on a pair of R210 with relayd, but these are R210 (II).

No kernel panics at all, and these systems are working in a live 
environment...


Regards
Andre



Am 02.05.17 um 15:03 schrieb Mathieu BLANC:

On Wed, Mar 29, 2017 at 02:06:23PM +0200, Mathieu BLANC wrote:

It also kernel panics with just this pf rules :
# cat pf_minimal.conf
set limit { states 10 }
set skip on lo
anchor "relayd/*"
pass



I upgraded the system to 6.1 release last week, the kernel panic is still here
(with the same logs).





smime.p7s
Description: S/MIME Cryptographic Signature

Re: Kernel panic on Dell R210 with OpenBSD 6.0 (relayd related ?)

[Mathieu BLANC]

On Wed, Mar 29, 2017 at 02:06:23PM +0200, Mathieu BLANC wrote:
> It also kernel panics with just this pf rules :
> # cat pf_minimal.conf 
> set limit { states 10 }  
> set skip on lo   
> anchor "relayd/*"
> pass 
> 

I upgraded the system to 6.1 release last week, the kernel panic is still here
(with the same logs).

-- 
Mathieu

What is "activity" on a relayed SSH connection? (Re: (solved) relayd splice timeout)

[Andreas Kusalananda Kähäri]

On Tue, May 02, 2017 at 12:35:22PM +0200, Markus Rosjat wrote:
> Hi Hiltjo,
> 
> just to let you know seesion timeout did the try and works like a charm


On a related note:  What constitutes "activity" on a relayed SSH
connection?

I'm also using relayd on a router to relay SSH connections to an
internal host, and it seems as if "relactl show sessions" will never say
that "idle" is anything other than the same as "age".

$ relayctl show sessions
session 0:447 192.168.1.4:63327 -> 10.0.0.10:22 RUNNING
age 01:13:47, idle 01:13:47, relay 1, pid 84257


Regards,
Kusalananda




> 
> Regards
> 
> Markus
> 
> Am 28.04.2017 um 11:34 schrieb Hiltjo Posthuma:
> > On Thu, Apr 27, 2017 at 07:11:56PM +0200, Markus Rosjat wrote:
> > > Hi there,
> > > 
> > > I was playing arround wit relayd just to get a feeling for it. So I 
> > > started
> > > with relaying a ssh connection to a machine behind my gateway.
> > > 
> > > But it seems there is some kind of config value I miss because after like 
> > >  8
> > > minutes the open ssh connection gets suddenly closed. Running relayd in
> > > foreground shows a splice timeout.
> > > 
> > > So question is, can I and if so where can I adjust the timeout value.
> > > 
> > > SSH might be a bad example for relayd use but its the easiest starting 
> > > point
> > > thought. Better to discover stuff befor a setup gets more complicated.
> > > 
> > > Regards
> > > 
> > > --
> > > Markus Rosjatfon: +49 351 8107223mail: ros...@ghweb.de
> > > 
> > > G+H Webservice GbR Gorzolla, Herrmann
> > > Königsbrücker Str. 70, 01099 Dresden
> > > 
> > > http://www.ghweb.de
> > > fon: +49 351 8107220   fax: +49 351 8107227
> > > 
> > > Bitte prüfen Sie, ob diese Mail wirklich ausgedruckt werden muss! Before 
> > > you
> > > print it, think about your responsibility and commitment to the 
> > > ENVIRONMENT
> > > 
> > 
> > Hey,
> > 
> > Have you tried "session timeout"?
> > 
> > They can be used for relays and redirections.
> > 
> > See the RELAYS and REDIRECTIONS section in relayd.conf(5).
> > 
> 
> -- 
> Markus Rosjatfon: +49 351 8107223mail: ros...@ghweb.de
> 
> G+H Webservice GbR Gorzolla, Herrmann
> Königsbrücker Str. 70, 01099 Dresden
> 
> http://www.ghweb.de
> fon: +49 351 8107220   fax: +49 351 8107227
> 
> Bitte prüfen Sie, ob diese Mail wirklich ausgedruckt werden muss! Before you
> print it, think about your responsibility and commitment to the ENVIRONMENT
> 

(solved) relayd splice timeout

[Markus Rosjat]

Hi Hiltjo,

just to let you know seesion timeout did the try and works like a charm

Regards

Markus

Am 28.04.2017 um 11:34 schrieb Hiltjo Posthuma:

On Thu, Apr 27, 2017 at 07:11:56PM +0200, Markus Rosjat wrote:

Hi there,

I was playing arround wit relayd just to get a feeling for it. So I started
with relaying a ssh connection to a machine behind my gateway.

But it seems there is some kind of config value I miss because after like  8
minutes the open ssh connection gets suddenly closed. Running relayd in
foreground shows a splice timeout.

So question is, can I and if so where can I adjust the timeout value.

SSH might be a bad example for relayd use but its the easiest starting point
thought. Better to discover stuff befor a setup gets more complicated.

Regards

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

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

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

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



Hey,

Have you tried "session timeout"?

They can be used for relays and redirections.

See the RELAYS and REDIRECTIONS section in relayd.conf(5).



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

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

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

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

Re: OpenBSD/octeon on EdgeRouter PoE - my experience

[Stefan Sperling]

On Tue, May 02, 2017 at 12:57:23AM +0200, Doggie wrote:
> W dniu 2017-05-02 o 00:01, etie...@magickarpet.org pisze:
> > I also own one of these nice devices, and replaced the usb thumbdrive
> > that was present for my own, to keep the original filesystem intact,
> > just in case. But I had to try a few USB drives, and I suspect the only
> > ones that could be booted on were USB2, and not USB3. That said, I
> > haven't verified this thoroughly.
> > 
> > Cheers,
> 
> I based my USB flashdrive research on this thread:
> 
> https://community.ubnt.com/t5/EdgeMAX/List-of-Compatible-USB-drives/td-p/1185011#
> (especially post #5)
> 
> and eventually chose Transcend JetFlash 710 32GB USB 3.1 Gen 1 - works like
> a charm in my EdgeRouter PoE.

On the ERL, try typing 'usb reset' to the boot loader if it does not detect
a USB stick. That made any USB stick work for me and also improved system
stability in general. 'usb reset' can also be put into the bootcmd variable.
The INSTALL.octeon file mentions this now (in 6.1). See there for details.

Re: iwm0 problems

[Stefan Sperling]

On Sun, Apr 30, 2017 at 06:08:18PM -0400, Steve Throckmorton wrote:
> > I also have this issue with AC 3160. What i did as a workaround was to 
> > switch iwm to 802.11g using
> > ifconfig iwm0 media autoselect mode 11g
> 
> Excellent!  That got my wireless interface working without error messages.  
> As far as I haven’t had a single firmware error in two hours.  I haven’t 
> tested the download speed because I don’t have anything to compare it to, but 
> it’s more than sufficient for my purposes.
> 
> Thank you.

Thank you all for the additional reports! So there is indeed a regression
in 6.1 which is causing this problem.

I will try to find a 3165 device to play with.