On Tue, 08 Jul 2025 20:18:21 +0100
Martijn Rijkeboer wrote:
> Wants / Requirements:
> - 802.11ax or 802.11be
> - PoE powered
> - Multi AP setup supported
> - No cloud-based controller software
> - No separate controller (preference)
While I haven't used one myself, the OpenWrt One (Banana Pi) se
> I get this:
> cannot open hd0a:/etc/random.seed: No such file or directory
> anyone to help me on this?
ignore it like most of us do?
you don't have a random.seed on install media, and chances are fat that
you don't need it at that time. If you see that error from booting on
your installed di
On Fri, 20 Jun 2025 03:24:38 +
Lloyd wrote:
> obs...@loopw.com wrote:
>
> > not 100% conjecture, but not 100% science either:
> > You have <90MB on /, which is where relinking happens, on a ~30MB
> > kernel. Thats a bit tight. My guess is more than one
> > thing was moving things around in t
On Fri, 20 Jun 2025 12:42:06 +1000
David Diggles wrote:
> This is the disk space available on mine.
>
> Filesystem SizeUsed Avail Capacity Mounted on
> /dev/wd0a 1021M144M826M15%/
> /dev/wd0e 1.9G 22.7M1.8G 2%/home
> /dev/wd0d 2.9G2.4G
On Thu, 19 Jun 2025 18:05:18 +
Lloyd wrote:
> Janne Johansson wrote:
>
> > that doesn't sound 100% like out-of-space.
> >
>
> Disk space was my first guess, but unlikely:
>
> Filesystem SizeUsed Avail Capacity Mounted on
> /dev/sd0a 384M276M 89.6M76%/
> /d
On Thu, 19 Jun 2025 04:32:35 +
Lloyd wrote:
> I encountered a relink failure after the last syspatch as well.
> https://marc.info/?l=openbsd-bugs&m=175023041416172&w=2
> David Diggles wrote:
>
> > Any ideas? I'm thinking maybe the root disk is on a bad CF card.
both David's and Lloyd's err
> ==> hostname.pppoe0 <==
> inet 0.0.0.0 255.255.255.255 0.0.0.1 pppoedev vlan40 authproto pap \
> authname 'user' authkey 'password' \
> up
try:
inet 0.0.0.0 255.255.255.255 NONE mtu 1492 \
pppoedev vlan40 authproto pap \
authname 'user' authkey 'password' \
u
> Just asking the last depict. Surprised, anyone can explain how some
> apps of Xfce like Orage Calendar are fully compatible with Mate
> panels ?
they use a common library - libgtk
> .. |
>+---+ o * . ~ *|
>| my |--> UDP 1337 --> % . pf : . --|--> UDP 31337 --> clouds
>|special| + . magic + |and
>|daemon |<-- UDP 1337 <-- * _ , + <--|--- UDP 31337
> after which was the last time that machine ever POSTed. At the
> time I wrote it off as corrupting EFI somehow but I'm not so sure anymore.
UEFI is stored on disk, it can be wiped. Thats not your POST issue though.
maybe time for a post code reader?
alternatively, wipe the cmos and try aga
> Now only today, I had a mediatek wifi card laying around…
> I even
> unplugged everything and put them back in, and still it's not POSTing
Have you tried slowly troubleshooting by pulling out all the cards, then
putting in one thing at a time and trying to post it each time?
PCI cards are
> The problem is the conflict that occurs naturally when connecting any two
> ISPs.
in my not at all humble opinion, with this topology (each ISP presenting the
same subnet on its client side), I would use multiple firewalls. One per ISP.
Use carp if you want to failover or do fancy tricks
there are situations where stateful firewalls can forget about the wireguard
session, but one or more of the peers remember it. Even wireguard’s keepalive
wont fix that situation.
my “bandwidth is basically free” solution:
I have a script that basically pings an endpoint over a tunnel, and res
pro-tip: for the most part the bsd’s split from each other in the 90s. Thats a
lot of lines of code to this day. You cannot bank on cron being the same in
open as it is in free, net, dragonfly, etc.
> On Nov 13, 2024, at 3:56 AM, Corey wrote:
>
> does BSD's crontab have any difference tha
>> I am a Ph.D. student at Cornell ORIE. I saw a Dell Optiplex 7050 SFF lying
>> around in our department and decided to install OpenBSD on it. The machine
>> does not have WiFi connectivity, but there is an RJ45 Ethernet jack, so I
>> plugged a cable in, and wrote a standard hostname.em0
>>
> In a second server I have upgraded from 7.5 i386 to 7.6 i386 but server sees
> only 4GB of RAM
> Is anybody with similar experiences? Any ideas how to fix RAM?
run 64bit OpenBSD
32bit address space is limited to a max of 4GB**, and some of that is eaten up
by PCI devices, etc.
** PAE can wor
> On May 29, 2024, at 3:48 AM, Radek wrote:
>
> Thank you, that explains everything.
> Does wireguard support replication? Will it work properly in my CARP setup?
wireguard doesn’t have “state” per se. it remembers the last address a key was
associated with. In the event of a failover, if
a bit late, but fwiw -
I have 7.5 on an optiplex 990 on SSD, no issues. Not quite a 980, but I
honestly doubt that this matters given my experience on these boxes. (I also
have 7.5 on a 960, and have had it on a 955 and 945. There’s nothing special
about the Dells that I have and booting MBR
> On Apr 25, 2024, at 10:36 AM, Radek wrote:
>
> Thank you for all your hints.
>
>> match out on egress from $lan_if:network to any nat-to (egress:0)
> This rule doesn't work.
change $lan_if to $int_if, change (egress:0) to $ext_carpif, and it will work
as the rule you say works.
fwiw, th
try using egress and not carp0 on your match out rule. you’re not technically
sending out on the carp interface, only receiving.
make your match rule like this:
match out on egress from $lan_if:network to any nat-to (egress:0)
> On Apr 24, 2024, at 11:05 AM, Radek wrote:
>
> Hi everyone,
> On Feb 20, 2024, at 2:31 AM, Thomas Schmidt wrote:
>
> OP did indeed mean `sysupgrade`,
which makes little sense _unless_ on -current, which will guarantee to break
this every sixth months when -current shifts about.
> but fwiw, `syspatch && reboot` reboots
> your system if a patch as app
> On Feb 15, 2024, at 10:12 AM, b...@fea.st wrote:
> am I the only one using automatic OS updates
> 0 3 * * * root sysupgrade
Maybe you meant “syspatch” there?
Syspatching via cron is questionable - Some of the patches wont be loaded until
the box is rebooted.
I think its bett
They have a pull request open to rename it to “h3sh" -
https://github.com/francoismichel/ssh3/pull/87
Hopefully sanity prevails there. Cutting in the namespace line by stealing the
next version number with zero backwards compatibility isn’t a great move to
gain a user base. Doing so with a se
are you certain that you upgraded your userland packages after upgrading?
wireguard-tools is critical to update in 7.4 (I think due in part to the
wgdescr field being added, which is a sorely missing field imo)
(for what its worth, I ran into the same problem, specifically because I’d
typo’d pk
> On Jun 5, 2022, at 4:24 AM, Sam Lee wrote:
>
> On 2022-06-05 10:58 +0200, Omar Polo wrote:
>> M-x no-tab-mode RET seems to work (as long as you don't want to change
>> the width of the `expandtab'; it always align to the nearest multiple
>> of 8)
>
> How do people write Python programs usin
Sorry to hear about your issues.
Vultr has been top notch in my experience. I have a couple of Vultr hosted
nodes in different data centers that have been running OpenBSD, they’ve been
online for four years now (with downtime for logging into the console to
activate FDE during reboots from sysp
note that ps/2 is not actually designed for hotplug (I fried a keyboard
controller to bring you this knowledge)
> On Nov 1, 2020, at 11:11 AM, Tomasz Rola wrote:
>
> On Sun, Nov 01, 2020 at 01:51:45PM -0500, Brennan Vincent wrote:
>> Is it possible to get hot-plugging of USB mice to work? Can'
My guess is that the nvram of the bios somehow reset its configuration back to
default. (corruption, power loss, etc. - these are cheap parts made with
failure rates after all)
The default of most x86 bioses up until the last few years was to bring up SATA
ports in IDE compatible mode, which w
Use multiple interfaces, one per site to connect with. Overhead isnt really
present, its just routing and hashes at that point.
(I’ve had no issues doing site to sites in this fashion, has been working great
for months)
> On Jul 31, 2020, at 10:43 AM, Sonic wrote:
>
> The need is for site-to
> Is this normal?
Checksum is OPTIONAL in UDP, not required. This is covered in RFC 768.
> On Jun 11, 2020, at 10:31 AM, obs...@loopw.com wrote:
>
> I didn’t have to bisect! Woo! While ipmitool no longer seems to function,
> once I enable ipmi in my running kernel I can successfully reboot a 6.7
> ipmi-of-this-vintage system now - where previously, from 6.6 going back into
> lat
> On May 31, 2020, at 6:07 PM, obs...@loopw.com wrote:
>
> ipmitool is timing out on my system with the kernel driver loaded, where I
> havent seen this in previous releases.
>
> I looked at the changes in 6.7 for ipmi.c, there's a number of them (thank
> you!) My intention is to bisect for
I have a small fleet of protectli firewalls, all of them with em nics. Only
the units i’ve upgraded to 6.7 are showing interface errors, where 6.6 is
definitely not.
> On Jun 8, 2020, at 5:30 PM, Gabri Tofano wrote:
>
> Hi all,
>
> I'm sending this e-mail since I have found other users in t
> What if two systems being used as redundant firewalls had different network
> cards? This would make the names of the interfaces different, resulting in
> rule sets that were not the same, preventing per-rule state timeouts from
> being properly applied.
1) “egress” can be used to referen
ipmitool is timing out on my system with the kernel driver loaded, where I
havent seen this in previous releases.
I looked at the changes in 6.7 for ipmi.c, there's a number of them (thank
you!) My intention is to bisect for which change may have caused this, but if
someone knows what’s going
After upgrade to 6.7 I had to change all of isc_named’s file statements to have
a / before their path. Fwiw, I don’t see this in the Changelog, nor does
sysmerge handle this, so I’m guessing this is some sort of new bug.
Before 6.7, this always worked:
file “master/loopw.com”
after, I have t
You have a 32bit uefi on the MacBook2,1
Its set to only ever address 3GB of ram, even though you can physically have
more in the box.
(this is a 32 bit issue in general - x86 hardware usually defaults to a
3.5GB/512MB split, but the 32bit uefi/64bit cpu macs went with a 3GB/1GB split)
> On Ma
37 matches
Mail list logo
