Is there any reason why miniroot.kmod and cgdroot.kmod
from release/arch/installation/minitroot aren't documented?
I'm asking because I want to modify cgdroot's rc script to recognise
special gpt labels but I don't want the logic to be hidden deep
inside that rc script.
It already probes for a
Tiago Seco wrote:
> Hello,
>
> I've been trying to install NetBSD on a UEFI system with encrypted root.
> ...
> When booting the system, it is not trying to boot from hd0c, but from hd0a.
If you intend to use cgdroot.kmod, the system should create a md(4)
disk and boot from it rather than hd0.
Greg Troxel wrote:
> ...
> Is this just what happens with when a tor circuit is lost? I would
> expect it to just be closed, without apparent spurious data.
I've been using onions for ssh connections for a couple of years already
and I don't recall anything similar to your problems. I use
Peter Ullberg wrote:
> Do NetBSD have support for the Intel I210-AT Ethernet Controller
>
>http://ark.intel.com/products/64400/Intel-Ethernet-Controller-I210-AT
wm(4)
o Intel I210 Ethernet (Copper, Fiber)
I don't know if -AT variant is supported, though.
--
Alex
Edgar Pettijohn wrote:
> Look at rc.subr. it calls su to start the daemon. Look at the
> manual for rc.subr I think there are some variables you could add
> to the rc.d script to change the behavior.
I only see su -m user -c ... in rc.subr. It's not the same as su -c class user.
$ man su
..
Edgar Pettijohn wrote:
> did you:
> # cap_mkdb /etc/login.conf
Yes, I did run it.
> > My understaning is that the tor process doesn't move to the "tor"
> > login class when switching a user. As a result, I can't restart it
> > when I login as root. I have to set the login class with su -c like
>
Another issue on my server was trying to start a service (tor relay in
my case) with a low limit on open files in the default login class.
The relay starts as root but then it swiches to user "tor". I set the
login class of the tor user to "tor" which further extends a limit of
the daemon class.
TCB (Trusted Computing Base) is a technique for locking down your
system with securelevel and immutable files. It's documented in
secmodel_securelevel(9).
I recently played with it and I had some minor issues which I'd like
to report:
- secmodel_securelevel.9 is in comp.tgz which isn't always
co...@sdf.org wrote:
> But until that, I used USB wifi. the dongle is small enough to not get
> in the way, I don't even have to disconnect it to put ti in a backpack.
If you use edimax, it will eventually break. My two edimax dongles broke
in the same way. They aren't sturdy enough to survive in
Siyar Erdemli wrote:
> Default XEN_DOMU kernel does not include the MODULAR option.
Good to know. I wonder if it's non-modular for a reason on netbsd-7.
> I add this option , options MODULAR , and is now working.
Are you saying that the kernel builds, it loads the modules and they
work as
Hi Siyar,
It's hard to understand from your email what is your config and what
exactly went wrong. I'll try but I may take a completely different
path.
Siyar Erdemli wrote:
> Hi
>
> We use NetBSD 7.0.1 , for virtual router and firewall. ( xen_domU)
>
> I get the following error.
> netbsd7 #
Andrew wrote:
> Hi,
> I've been trying without success for a while to compile and install
> /usr/pkgsrc/net/lua-socket
> It tells me it installed correctly, I've installed it for lua 5.3, I had to
> manually update the LUA_PATH and LUA_CPATH variables, but now when I go to
> load the module
Benny Siegert wrote:
>
> > I learned about this issue from https://github.com/rust-lang/rust/pull/28543
>
> Exciting!
>
> Is anyone working on getting a newer version of Rust into pkgsrc? wip has
> 0.11, and 1.7.0 seems to be the current one.
My friend Ewan created a ticket to track NetBSD
Benny Siegert wrote:
> In benchmarks for Go, the devs found that 64-bit code gives a
> 10-15% speedup across the board. This is because there are more
> processor registers in 64-bit mode, so more variables can be kept
> in registers.
On the other hand, more compact size_t, ptrdiff_t and
31.03.2016, 08:37, "Andreas Gustafsson" <g...@gson.org>:Alexander Nasonov wrote: I don't know if it will help but try building without -pie (--disable-pie).I just tried it, by adding "CONFIGURE_ARGS+= --disable-pie" to the packageMakefile. It didn't help.Keep this opt
Andreas Gustafsson wrote:
> Hi all,
>
> I'm trying to run Firefox under gdb on NetBSD 7.0/amd64, but I'm not
> having much luck.
>
> I found a set of instructions at
>
> https://wiki.netbsd.org/tutorials/pkgsrc/debugging_firefox/
>
> but when I follow them, gdb behaves stranegly, like being
76nem...@gmx.ch wrote:
On Thu, Jul 02, 2015 at 11:08:39AM -0400, Christos Zoulas wrote:
This is not related to PIE; I think you mean -mtune=native, not
-march-native.
christos
Hello,
Yes that true. I have stopped to play with PIE since it does not work
currently.
Do you still
Christos Zoulas wrote:
I think that this problem has been fixed on HEAD, we'll issue a pullup...
Yep, it has been fixed in HEAD. But ran into other compilation issues ...
Alex
Christos Zoulas wrote:
In article 20140914034725.gs22...@arwen.poofy.goof.com,
Aaron J. Grier agr...@poofygoof.com wrote:
On Fri, Jun 20, 2014 at 09:05:33AM +0200, Martin Husemann wrote:
And of course we would accept patches that make [NetBSD] work with yet
another compiler. Are there any
Jimmie Houchin wrote:
tldr;
What are some well supported wireless 802.11n or ac cards for laptops?
I have not been able to get NetBSD to successfully use my wireless card
Intel 1030N.
Hi Jimmie,
I understand your feellings. I'm a developer but I have problems too
with wifi card on my laptop.
20 matches
Mail list logo
