ddb: machine sysregs for amd64

Hi, This is a short diff to add "machine sysregs" to ddb on amd64 (plus it also prints out gsbase/kgsbase). This command is available on i386 but not amd64. I swear I remember discussing this with mlarkin at some point but I couldn't find a previous patch for it on tech@. If I missed it

Re: ddb: machine sysregs for amd64

On Tue, Aug 31, 2021 at 06:30:40PM +1000, Alex Wilson wrote: > Hi, > > This is a short diff to add "machine sysregs" to ddb on amd64 (plus it also > prints out gsbase/kgsbase). This command is available on i386 but not amd64. > I swear I remember discussing this with mlarkin at some point but I

reduce debug logging from slowcgi

Hi all, On a busy-ish site, I found that slowcgi is doing quite excessive logging: every single environment variable is logged on a separate logline. There's at least 17 variables per hit, but I've seen it go up to 35. If you're writing debug logs from syslog, that adds up rather quickly. Of

teach pf to refragment ipv4 packets

i am in an annoying situation where i peer with a campus network on an ospf link with a 9k mtu, but some corners of that network have layer 2 hops that don't support 9k packets. i sometimes want to tunnel large (1500 byte) packets to hosts in those corners of the network by letting the

rpki-client exclude files from rsync fetch

RPKI repository can only include a few specific files, everything else is just ignored and deleted after every fetch. Since openrsync supports --exclude-file now we can use this to limit what is actually accepted by the client. I used a config file in /etc/rpki instead of using multiple

Re: rpki-client exclude files from rsync fetch

I don't understand -- why would people edit this file? If this list is in argv, it will be difficult to identify targets using ps, because the hostname is way at the end. Job Snijders wrote: > Hi, > > I don't think this should be user configurable. > > If folks remove entries like "+ *.crl"

Re: reduce debug logging from slowcgi

OK florian On 2021-08-31 16:24 +02, Paul de Weerd wrote: > Hi all, > > On a busy-ish site, I found that slowcgi is doing quite excessive > logging: every single environment variable is logged on a separate > logline. There's at least 17 variables per hit, but I've seen it go > up to 35. If

Re: teach pf to refragment ipv4 packets

Hi, This looks like a hack for a problem that should not exist. What is the MTU of the outgoing interface on your pf router? If the layer 2 switches do not support 9k jumbo frames, it must be 1500. Why are the outgoing packets not fragmented to the MTU? Is the dont-fragment flag set? Does pf

Re: rpki-client exclude files from rsync fetch

Hi, I don't think this should be user configurable. If folks remove entries like "+ *.crl" it breaks things. If folks add entries like "+ *.mp3" it wastes network bandwidth. :-) Let's use "--include" and "--exclude" instead. kind regards, Job On Tue, Aug 31, 2021 at 02:23:57PM +0200, Claudio

regress progs

Hi, bsd.prog.mk supports PROGS (with an S) for a while. I think we should have multiple programs in bsd.regress.mk, too. Mainly for consistency, but a few tests could be simplified with this. ok? bluhm Index: bsd.regress.mk ===

Re: rpki-client exclude files from rsync fetch

Theo de Raadt(dera...@openbsd.org) on 2021.08.31 11:09:22 -0600: > I don't understand -- why would people edit this file? > > If this list is in argv, it will be difficult to identify targets using > ps, because the hostname is way at the end. Yes. If we worry about people touching it,

Re: allow KARL with config(8)'d kernels

Thank you for the suggestion Theo, I tried to update those two pages, but a better name did not occur to me. If anything it points at an issue in the boot_config page which currently states: "Changes made can be saved for the next reboot, by using config(8)." Although true, this breaks KARL

diff(1)ing hardlinks

If two files to be compared share the same inode, it should be reasonable to consider them identical. This gives a substantial speedup when comparing directory structures with many hardlinked files, e.g. when using rsnapshot for incremental backup. Comments? OK? /Alexander Index: diffreg.c

uaq(4): aquantia usb ethernet driver

Here's a driver for the Aquantia USB ethernet devices I just added to usbdevs. These are somewhat interesting because they theoretically go up to 5GbE and support jumbo frames (not implemented yet). While working on this I noticed that it doesn't receive 15-25% of the packets it should, even at

acpibtn.4: Mention sleep putton, lid status and machdep.{lid,pwr}action

landry added the sensor back in 2013 and suspend via sleep button also works (at least on ThinkPads). machdep.*action are super useful and I dislike grepping /etc/examples/ for to read about them. acpibtn(4) is the most prominent driver supporting, so documenting them there seems fine and

Re: teach pf to refragment ipv4 packets

On Tue, Aug 31, 2021 at 09:34:19PM +0200, Alexander Bluhm wrote: > Hi, > > This looks like a hack for a problem that should not exist. I should unconditionally refragment reassembled packets? > What is the MTU of the outgoing interface on your pf router? If > the layer 2 switches do not

Re: iked(8): client-side DNS support via resolvd(8)

+ rtm.rtm_priority = RTP_PROPOSAL_STATIC; So my gut reaction is we should have /usr/include/net/route.h:#define RTP_PROPOSAL_TEMPORARY62 I hesitate calling this "VPN", or "road warrior", or making it specific to certain types of proposal offering daemons...

rpki-client add http_proxy support

This diff improves the http code by a) adding an IO timeout and b) implementing http_proxy support. Works for me using tinyproxy as proxy server. -- :wq Claudio Index: encoding.c === RCS file:

iked(8): client-side DNS support via resolvd(8)

IKEv2 allows road warrior servers to announce internal name servers in a configuration payload. iked responders can be configured to send such payloads with the 'config name-server' option. This diff adds support for receiving DNS server configuration payloads as a road warrior client and

Re: [External] : better use the tokeniser in the pfctl parser

On Tue, Aug 31, 2021 at 07:33:40AM +0200, Alexandr Nedvedicky wrote: > Hello, > > On Tue, Aug 31, 2021 at 02:40:57PM +1000, David Gwynne wrote: > > handling the "no" option with a token, and "yes" via a string made my > > eye twitch. > > > > ok? or is the helpful yyerror a nice feature? > > >

Re: iked(8): client-side DNS support via resolvd(8)

Very interesting. Please be very careful that proposal withdrawal actually works, or the experience will be poor.

Re: iked(8): client-side DNS support via resolvd(8)

This diff doesn't set rtm_index (to identify the interface the dns proposal is associated with) I guess that means rtm_index is 0. Inside resolvd, the proposal rtm_index is used to track proposals in the learned[] array. resolvd uses if_indextoname() to annotate the interface name on these

Re: [External] : better use the tokeniser in the pfctl parser

I am really against the idea of the parser inspecting a static buffer from the lex. Also we have a ton of these parsers, and discourage them from deviating. This tiny little "please use the right keyword" change feels so minor; we do not have a generic error-correction-proposing parser, 99% of