00:21:28 ober sweet! I'm working with a 13 intel and I'm running NomadBSD on it. I also got normal 14.1 working too, but Nomad works better running off of the storage exansion card 01:29:05 :D 03:33:55 xit 03:33:57 quit 04:55:32 Can someone please help me recover my FreeBSD server install? 04:56:33 It is a Hyper-V VM that when booted, says $hostname init 574 - - can't exec getty tty' for port /dev/ttyv0: No such file or directory 04:57:05 It is 14.1 RELEASE and I can enter single-user mode and that works, but not multi-user mode for that reason 04:57:42 Seems that a lot of critical system files are missing, but getty and /dev/ttyv0 is there when I check in single-user mode 04:58:48 I was having bootloader/UEFI issues earlier but that seems resolved now, and this is the only thing holding me back. I am using ZFS for the system boot disk on /dev/da0p4 05:31:38 hi, does anyone know of a way to dump the current NAT mappings in ipfw? Like was requested in https://forums.freebsd.org/threads/how-do-i-view-active-mappings-in-libalias-ipfw_nat.31520/ 06:10:16 if not, i am currently doing some ipfw development, maybe i can make patch for this feature too 10:42:48 I've found an essay about how FreeBSD is different from Linux 10:43:25 apparently one major difference is that FreeBSD uses CVS for version control, whereas Linux has not had version control until recently but now it's in BitKeeper 10:44:20 that must have been written a good while ago, then 10:44:24 Also this page is apparently "powered by perl" and "valid HTML 4.01" 10:46:01 I figured after twenty years of Linuxing I should go learn BSD, so I've been casting around for things to do with an old Raspberry PI 2B that I've gotten up and running 10:46:30 if nothing else, not having to deal with a modified ubuntu on a raspberry pi is already a stellar use case 10:47:31 so far I'm getting the feeling that the main things to really go and play with are jails and zfs. Bhyve would be neat, but this hardware wouldn't do it 10:50:33 zip: do make sure you installed the armv7 version, since armv6 support is being dropped in the next release (15.0) 10:51:12 (there are apparently old guides telling people to use armv6 for those older rpis) 10:53:28 I beleive it is 10:53:46 `FreeBSD-14.1-RELEASE-arm-armv7-GENERICSD.img.xz` 11:00:17 also hi ivy we are in far too many channels together 11:02:09 so we are 11:02:53 nerrrrd 11:03:39 there's like 3 people from that other network who are also in every other channel i'm in 11:07:54 oh yes, I think I've spotted one of them already 13:07:51 p9fs update: after markj@'s recent fixes (so 15.0 post October 25) it now seems very stable 13:08:04 i haven't tried it as a root filesystem yet, that's next on the list 15:26:26 ivy: woot 15:56:22 i just with i could reproduce the nd lock crash, the affected system has been stable for days now... i'm wondering if there's a timing issue where running a debug kernel prevents it from happening 15:56:27 s/with/wish 15:58:06 also i think someone broke pf on non-INET kernels again 15:58:36 Oct 18 16:34:00 amaranth kernel: link_elf_obj: symbol vnet_entry_in_loopback_mask undefined 15:58:37 Oct 18 16:34:00 amaranth kernel: KLD file pf.ko - could not finalize loading 16:03:32 https://github.com/freebsd/freebsd-src/commit/27f54be50bbad 16:04:15 can't use IN_LOOPBACK without INET 16:07:29 i don't even know why this is the case, tbh. aside from VNET being a giant pile of technical debt that no one wants to touch 16:11:30 like i think this symbol might only exist to support net.inet.ip.loopback_prefixlen which is a bizarre sysctl that no one should ever change 16:11:36 maybe that should just be removed? 16:57:40 https://github.com/freebsd/freebsd-src/pull/1511 17:01:22 shouldn't you #ifdef INET6 that case AF_INET6 in pf.c? 17:06:24 nimaje: no, reload the page for rationale 17:06:57 i mean, it could be put behind #ifdef INET6, but since the original code wasn't, i see no reason to add it 18:38:39 I'm an FBSD newb and just came across this https://www.freebsd.org/security/advisories/FreeBSD-SA-24:19.fetch.asc and am curious if it's common practice in FBSD to use environment variable for things instead of a config file? 18:39:48 stdout: the problem here isn't that it used environment instead of a config file, the problem is that the API between fetch(1) and libfetch uses an environment variable 18:39:57 which is, frankly, bizarre, and definitely not normal practice 18:40:16 https://github.com/freebsd/freebsd-src/commit/51676e0a3bd3 18:42:58 interesting. thanks for the info ivy 18:46:51 Has anybody managed to use LDAP with FreeBSD 14 stable? 18:47:11 or above stable 12+ 18:47:36 i currently use ldap on 15.0, haven't tried it on 14-stable for ages but i'm sure it used to work. assuming you mean nss ldap 18:50:07 ha.. stable/12 18:50:12 outdated 19:11:02 Do you run it in a Jail or Bare metal? 19:12:21 are you talking about the LDAP NSS client or the server? 19:12:53 i run the NSS client on all my systems, regardless of whether they're jails, VMs, or hardware. the server (openldap) i currently run in a jail 19:13:15 server 19:14:29 I was wondering if I should scrap the installation, Its currently running as a root user but I am encountering lots of issues 19:19:09 I figured I could use the LDAPscripts PKG portz but that won't work either 19:19:42 I wonder if there is some sort of color nflict between the host and the service 19:19:55 I might have to vent it into a Jail 19:20:13 vnet it, into a jail 20:56:03 this all implies that jails don't have the same user accounts, is that right? 20:56:16 zip: by default, correct, a jail has its own /etc/master.passwd 20:56:20 oh neat 20:56:26 you could of course connect a jail to NIS or LDAP or whatever 20:56:26 that sounds like something tinker-able :D 20:56:44 now that I've got a FreeBSD system up and running I've kind of been scratching my head about what to actually _do_ with it 20:56:58 I was thinking ejabberd in a jail would be a good first project 20:57:04 I've been wanting to set up ejabberd for a while anyways 21:03:25 zip: if you want to learn interesting things, make it a vnet jail 21:03:32 they are so much nicer than old-style jails 21:04:03 Ooo 21:04:19 Are these the ones that have a virtual switch 21:04:33 Love the idea of giving each service an ipv6 address tbh 21:04:39 not a switch, just a virtual network cable. you create an epair interface, epair0a is in the host, epair0b is in the jail 21:04:51 if you wanted you could create a switch by adding a bridge interface on the host then putting epair0a into a bridge 21:05:00 or you could just treat it as a pure layer 3 interface and route it 21:06:11 it's very flexible 22:02:24 i've been meaning to boot up a linux container the hard way 22:02:58 full on makign the network namespace and the interfaces, changing cgroups, downloading layers from the interweb and doing the stupid stacked image thing