Switch to DuckDuckGo Search
   September 18, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >

Toggle Join/Part | bottom
[00:24:12] *** merzo <merzo!~merzo@228-53-133-95.pool.ukrtel.net> has quit IRC (Ping timeout: 245 seconds)
[00:31:55] *** sarvet <sarvet!~sarvet@i59F4522F.versanet.de> has quit IRC (Quit: sarvet)
[00:39:53] *** merzo <merzo!~merzo@63-28-133-95.pool.ukrtel.net> has joined #illumos
[00:45:52] <LeftWing> richlowe: I wish I could remember where I put the source for the cstyle PDF
[00:45:58] <LeftWing> We could clarify some things :P
[00:46:38] <richlowe> you were going to manual page it!
[00:46:44] <LeftWing> That sounds like something I would say
[01:00:41] *** andy_js <andy_js!~andy@97e29e78.skybroadband.com> has quit IRC (Quit: andy_js)
[02:01:09] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC (Remote host closed the connection)
[02:02:06] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[02:06:37] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has quit IRC (Ping timeout: 268 seconds)
[02:54:32] <Reinhilde> Does illumos have a tuntap driver?
[02:59:14] <Reinhilde> huh, it's separate
[02:59:38] <richlowe> I think that's largely because nobody has bothered to integrate it
[03:00:24] <Reinhilde> that's an oddity. I'd assume it'd be in the base system
[03:07:53] <Reinhilde> oh, wait, it's GPL, that explains everything.
[03:08:58] *** despair86 <despair86!~despair@exit.i2p.rocks> has joined #illumos
[03:09:52] <Reinhilde> hi despair86
[03:11:10] <despair86> hi
[03:11:34] <Reinhilde> why isn't tuntap integrated into base? oh, wait, it's GPL, that explains everything.
[03:13:08] <despair86> what'd i miss
[03:13:53] <despair86> also, tuntap doesn't export any of the network card kstats
[03:15:16] <despair86> furthermore, Kaizawa's tuntap is BSD or MIT: https://github.com/kaizawa/tuntap/blob/master/tun.c
[03:15:40] <despair86> damn it, i missed a dot
[03:15:50] <despair86> *k. aizawa
[03:16:31] <Reinhilde> henkum
[03:16:51] <Reinhilde> despair86: no, it's GPL too
[03:16:58] * despair86 gasps
[03:17:05] <Reinhilde> furthermore it looks as though it is the same driver
[03:17:14] <Reinhilde> https://github.com/kaizawa/tuntap/blob/master/if_tun.h
[03:17:16] <despair86> oh dAMN IT
[03:17:31] <despair86> serves me right for skimming through licence headers
[03:20:01] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[03:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[03:20:51] * Reinhilde bites despair86
[03:21:19] * despair86 screams
[03:22:15] <Reinhilde> what the hel despair86
[03:24:46] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[03:50:21] *** HotSwap <HotSwap!~hotswap@unaffiliated/hotswap/x-1719459> has quit IRC (Read error: Connection reset by peer)
[03:56:03] *** HotSwap <HotSwap!~hotswap@192.95.27.155> has joined #illumos
[03:56:18] *** HotSwap <HotSwap!~hotswap@192.95.27.155> has quit IRC (Changing host)
[03:56:18] *** HotSwap <HotSwap!~hotswap@unaffiliated/hotswap/x-1719459> has joined #illumos
[04:45:01] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Remote host closed the connection)
[04:45:44] *** baojg <baojg!~baojg@162.243.44.213> has joined #illumos
[04:53:08] *** baojg_ <baojg_!~baojg@162.243.44.213> has joined #illumos
[04:53:30] *** baojg <baojg!~baojg@162.243.44.213> has quit IRC (Ping timeout: 265 seconds)
[05:01:47] *** wl_ <wl_!~wl_@2605:6000:1b0c:4124::87c> has quit IRC (Quit: Leaving)
[05:03:31] *** despair86 <despair86!~despair@exit.i2p.rocks> has quit IRC (Remote host closed the connection)
[06:39:24] *** wl_ <wl_!~wl_@2605:6000:1b0c:4124::87c> has joined #illumos
[07:23:05] *** Yogurt <Yogurt!~Yogurt@c-73-189-45-147.hsd1.ca.comcast.net> has joined #illumos
[07:27:37] *** Yogurt <Yogurt!~Yogurt@c-73-189-45-147.hsd1.ca.comcast.net> has quit IRC (Ping timeout: 258 seconds)
[08:00:51] <tsoome> it is a bit shame we do not have proper GLD enabled tuntap…
[08:28:47] <Reinhilde> tsoome: GLD?
[08:30:05] <Reinhilde> you can contact the original author and the forker and ask them how much it'll cost you for them to do you a CDDL relicensing solid
[08:30:13] <tsoome> https://illumos.org/books/wdd/gld-1.html#gld-1
[08:32:15] <Reinhilde> tsoome: coming from not-kernel not-dev, how easy is it to dive into illumos kernel dev given that i don't run illumos on my home computer
[08:32:16] <tsoome> the current tuntap is hack making tun/tap devices available, but they do not plug into our networking infrastructure.
[08:33:04] <tsoome> get some virtualization software - vbox, vmware workstation/fusion, parallels...
[08:33:22] <Reinhilde> i don't even have 500mb ram free at any given time on my freebsd machne
[08:33:22] <tsoome> qemu:)
[08:33:33] <Agnar> oh, I patched the tun driver for sol9 back then on sparc...what a mess ;(
[08:33:36] <tsoome> freebsd can run bhyve
[08:33:54] <Reinhilde> on a scale from most of the difficult (1) to all of the impossible (255), how easy is it to get into illumos kernel dev once you have a good setup going
[08:33:57] <tsoome> but memory … yes that can be issue
[08:34:21] <tsoome> easy is subjective term
[08:34:29] <Reinhilde> let's see.
[08:34:37] <Reinhilde> I'm not very good at C in userspace
[08:34:44] <tsoome> ah.
[08:34:49] <tsoome> well, then it is hard.
[08:35:11] <Reinhilde> and the only kernel dev I have ever done was on FreeBSD, to make my now dead laptop's touchscreen work.
[08:35:37] <tsoome> not impossible - but it will rquire time and dedication to learn, practice and study.
[08:36:19] <Reinhilde> I have written C in userspace and I know how to read (BSD, not S5R4 Unix) manuals.
[08:37:00] <tsoome> usually the suggestion is to start with small steps:)
[08:38:19] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Quit: tsoome)
[08:38:58] <Reinhilde> also annoying: even if /driver/tuntap plugged into gld, it's in the gpl
[08:39:10] <Reinhilde> which is not cddl compatible.
[08:42:26] *** aszeszo <aszeszo!~aszeszo@unaffiliated/aszeszo> has joined #illumos
[08:42:28] *** andy_js <andy_js!~andy@97e29e78.skybroadband.com> has joined #illumos
[08:54:32] *** awordnot <awordnot!~awordnot@c-73-210-60-203.hsd1.il.comcast.net> has quit IRC (Read error: Connection reset by peer)
[08:55:14] *** awordnot <awordnot!~awordnot@c-73-210-60-203.hsd1.il.comcast.net> has joined #illumos
[08:55:39] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Ping timeout: 240 seconds)
[08:56:46] *** alanc <alanc!~alanc@129.157.69.38> has quit IRC (Remote host closed the connection)
[08:57:13] *** alanc <alanc!~alanc@129.157.69.38> has joined #illumos
[08:58:27] *** merzo <merzo!~merzo@63-28-133-95.pool.ukrtel.net> has quit IRC (Ping timeout: 240 seconds)
[09:29:06] *** merzo <merzo!~merzo@252-127-203-46.pool.ukrtel.net> has joined #illumos
[09:29:47] *** BH23 <BH23!~BH23@94.175.232.126> has joined #illumos
[09:31:39] *** arekinath <arekinath!~arekinath@pdpc/supporter/professional/arekinath> has quit IRC (Ping timeout: 240 seconds)
[09:31:53] *** arekinath <arekinath!~arekinath@pdpc/supporter/professional/arekinath> has joined #illumos
[09:32:03] *** jordan[m]1 <jordan[m]1!~jordantim@104.166.94.34> has quit IRC (Ping timeout: 240 seconds)
[09:32:57] *** jordan[m]1 <jordan[m]1!~jordantim@104.166.94.34> has joined #illumos
[09:41:34] *** merzo <merzo!~merzo@252-127-203-46.pool.ukrtel.net> has quit IRC (Ping timeout: 265 seconds)
[09:59:38] *** Risa <Risa!~dukrat@45.35.251.216> has quit IRC (Ping timeout: 240 seconds)
[09:59:52] *** Risa <Risa!~dukrat@2602:ffc5:30::1:a835> has joined #illumos
[10:25:11] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[10:28:08] *** man_u <man_u!~manu@manu2.gandi.net> has joined #illumos
[10:37:30] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[10:38:40] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:17:05] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:17:40] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:19:19] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[11:34:06] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:34:35] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:45:18] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:45:37] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:53:31] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[11:54:07] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[11:56:11] *** Risa <Risa!~dukrat@2602:ffc5:30::1:a835> has quit IRC (Ping timeout: 276 seconds)
[11:59:26] *** Risa <Risa!~dukrat@45.35.251.216> has joined #illumos
[12:07:50] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[12:08:30] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[12:08:52] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[12:09:30] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[12:14:23] *** Risa <Risa!~dukrat@45.35.251.216> has quit IRC (Ping timeout: 276 seconds)
[12:15:59] *** Risa <Risa!~dukrat@2602:ffc5:30::1:a835> has joined #illumos
[12:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[12:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[12:46:53] *** vila <vila!~vila@2a01:e35:2e63:5f40:ec51:d4c:8983:3af2> has quit IRC (Ping timeout: 276 seconds)
[12:58:40] *** vila <vila!~vila@2a01:e35:2e63:5f40:9463:f55e:f0e7:2e97> has joined #illumos
[13:34:14] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has quit IRC (Quit: Leaving.)
[13:39:51] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[13:40:25] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[13:47:29] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[13:48:23] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has quit IRC (Ping timeout: 268 seconds)
[14:10:27] *** BH23 <BH23!~BH23@94.175.232.126> has quit IRC (Ping timeout: 240 seconds)
[14:10:49] *** Kurlon <Kurlon!~Kurlon@mobile-166-172-61-154.mycingular.net> has joined #illumos
[14:19:38] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[14:20:11] *** jellydonut <jellydonut!~quassel@s91904423.blix.com> has joined #illumos
[14:21:08] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[14:24:00] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[14:24:19] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[14:25:02] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[14:25:18] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[14:28:31] *** BH23 <BH23!~BH23@94.175.232.126> has joined #illumos
[14:28:44] *** gh34 <gh34!~textual@cpe-184-58-181-106.wi.res.rr.com> has joined #illumos
[14:57:08] <jlevon> do we at all support compiling with studio still
[15:00:47] *** Kurlon <Kurlon!~Kurlon@mobile-166-172-61-154.mycingular.net> has quit IRC (Ping timeout: 276 seconds)
[15:09:41] <andyf> I don't know definitively, but I think not
[15:10:25] <andyf> In general, it isn't possible for people to test before RTI at least
[15:11:22] <jlevon> might be nice to assert as such. I just modified an .il file and felt a little silly
[15:20:03] <yuripv> I know Gordon does (for a reason), that should be the end of list
[15:28:06] <sjorge> Studio is the binary only compile from the sun days right?
[15:30:15] <andyf> Yes, Sun/Oracle's commercial compiler
[15:32:20] <andyf> It used to be possible to build gate with it, but only with a version that is not publically available
[15:32:50] <tsoome> was Gordon needing it because of dbx?
[15:32:57] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has quit IRC (Ping timeout: 245 seconds)
[15:34:04] <tsoome> I remember there was some discussion around the subject but can not recall the details…
[15:34:09] *** MilkmanDan <MilkmanDan!~dan@wilug/expat/MilkmanDan> has joined #illumos
[15:35:58] <andyf> I think it was because of dbx, yes
[15:36:01] <yuripv> yes
[15:36:27] <yuripv> he likes the source debugging and gdb doesn't work for him
[15:36:53] <tsoome> I gave up dbx once it did start to crash:D
[15:38:32] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[15:38:55] <gitomat> [illumos-gate] 11684 shadow compilers dislike glib2 2.62 -- Andy Fiddaman <omnios at citrus-it dot co.uk>
[15:39:02] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[15:39:12] <andyf> I don't think I've used it since 1991..
[15:40:20] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[15:40:47] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[15:41:20] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has quit IRC (Ping timeout: 258 seconds)
[15:41:30] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[15:41:47] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[15:44:53] <yuripv> I do dislike glib2/hald/... as well, do we really need them (and consumers) in the gate :D
[15:49:18] *** lblume <lblume!~lblume@greenviolet/laoyijiehe/lblume> has joined #illumos
[15:55:04] <andyf> I don't know. There's a split anyway - hal depends on dbus and dbus is from the distribution and not gate
[16:02:00] *** merzo <merzo!~merzo@185.39.197.205> has joined #illumos
[16:07:16] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 265 seconds)
[16:08:32] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #illumos
[16:11:09] <papertigers> andyf: excited for your netstat email
[16:11:20] <papertigers> maybe we can finally enhance https://github.com/bahamas10/illumos-sockstat and get that into illumos-gate
[16:12:02] <papertigers> it already took a lot of code from netstat
[16:12:15] <andyf> I didn't know about that one - it looks at tcpConnCreationProcess etc?
[16:13:09] <gitomat> [illumos-gate] 11705 adding a swap zvol on a DEBUG build blows assert after 11650 -- Jerry Jelinek <jerry.jelinek at joyent dot com>
[16:13:18] *** wacki <wacki!~wacki@i577B8B03.versanet.de> has joined #illumos
[16:13:59] <andyf> Looks like it would be useful too.
[16:14:02] <papertigers> andyf: I dont remember how bahamas10 implemented it, I just remember he ripped a lot of the code out of netstat to do it.
[16:14:16] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
[16:15:10] <andyf> I wonder if handles TLI/XTI sockets
[16:15:18] <andyf> that's where I lost some more hair..
[16:16:18] <papertigers> I know one of the main issues was accurate pid reporting. It wouldn't follow fork/execs properly iirc
[16:16:35] *** patdk-lap <patdk-lap!~patrickdk@208.95.164.6> has joined #illumos
[16:16:44] <papertigers> but with Mohamed's and your changes maybe we can fix that now
[16:17:06] <andyf> Should be able to, yes (and fd-passing, and...)
[16:18:55] <andyf> I've been working on this off-and-on for over a year, so it's good to get to this stage. I'm expecting there will be comments...
[16:19:47] <papertigers> yeah, I will try and take a look at the code today -- mostly because I am curious about this in general
[16:24:35] <andyf> It will make a nice change from zone-based-nfs and rust :) Thanks
[16:30:37] <jbk> heh
[16:30:39] *** wacki <wacki!~wacki@i577B8B03.versanet.de> has left #illumos
[16:32:30] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[16:33:02] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[16:33:29] <KungFuJesus> https://www.illumos.org/issues/11493 Is this patch in progress?
[16:33:46] *** aszeszo_ <aszeszo_!~aszeszo@unaffiliated/aszeszo> has joined #illumos
[16:34:45] *** jimklimov1 <jimklimov1!~jimklimov@31.7.243.238> has joined #illumos
[16:35:01] *** aszeszo <aszeszo!~aszeszo@unaffiliated/aszeszo> has quit IRC (Ping timeout: 258 seconds)
[16:35:08] *** wacki <wacki!~wacki@i577B8B03.versanet.de> has joined #illumos
[16:37:43] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Ping timeout: 265 seconds)
[16:48:01] <yuripv> KungFuJesus: looks like no one just did review it, RB links don't show any activity
[16:53:11] <KungFuJesus> looks like 3 people reviewed it: https://www.illumos.org/rb/r/2201/diff/1/#
[16:53:26] <KungFuJesus> though, perhaps that's silo'd to joyent
[16:55:34] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[16:56:59] *** jimklimov1 <jimklimov1!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[16:57:29] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[16:58:06] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[17:04:20] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[17:05:04] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[17:05:27] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[17:05:59] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[17:06:24] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[17:07:00] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[17:14:13] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[17:15:04] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[17:15:31] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Read error: Connection reset by peer)
[17:16:19] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has joined #illumos
[17:18:15] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #illumos
[17:26:45] *** jimklimov <jimklimov!~jimklimov@31.7.243.238> has quit IRC (Ping timeout: 258 seconds)
[17:27:59] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has quit IRC (Ping timeout: 265 seconds)
[17:34:51] <gitomat> [illumos-gate] 11600 redirmod: cast between incompatible function types -- Toomas Soome <tsoome at me dot com>
[17:35:00] *** tsoome <tsoome!~tsoome@148-52-235-80.sta.estpak.ee> has joined #illumos
[17:36:17] *** merzo <merzo!~merzo@185.39.197.205> has quit IRC (Ping timeout: 245 seconds)
[17:41:29] *** aszeszo_ <aszeszo_!~aszeszo@unaffiliated/aszeszo> has quit IRC (Quit: aszeszo_)
[17:41:32] <gitomat> [illumos-gate] 11601 telmod: cast between incompatible function types -- Toomas Soome <tsoome at me dot com>
[17:44:14] *** wacki <wacki!~wacki@i577B8B03.versanet.de> has left #illumos
[17:49:57] <gitomat> [illumos-gate] 11663 loader.efi: text mode background color is limited to 3 bits -- Toomas Soome <tsoome at me dot com>
[17:56:42] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has quit IRC (Ping timeout: 245 seconds)
[18:01:00] <KungFuJesus> does anyone know if OpenIndiana has LX branded zones?
[18:01:28] <yuripv> it doesn't, OI uses vanilla illumos-gate
[18:01:44] <KungFuJesus> awww man :(
[18:02:05] <tsoome> need to update gate.
[18:02:57] <yuripv> meaning?
[18:03:16] <tsoome> get lx and bhyve to gate:)
[18:03:33] <jlevon> well it sure would be nice
[18:04:03] <tsoome> otherwise we can start packing.
[18:05:39] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 240 seconds)
[18:07:19] <yuripv> that would require someone with a lot of free time, and desire to work on getting that upstreamed
[18:07:53] <yuripv> (e.g. for me, I have absolutely no use for both, even if I had time and enough knowledge to do it)
[18:10:31] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has joined #illumos
[18:10:45] <KungFuJesus> Someone just asked if we could install docker register directly on the file server. LX branded zones _would_ have been the answer to that
[18:10:56] <KungFuJesus> registry*
[18:11:50] <igork> yuripv: very easy for lx + bhyve: rename illumos-joyent to illumos-gate and use it :)
[18:15:34] <yuripv> yeah, I did propose that actually some time ago, only half joking :D
[18:18:18] <jbk> bhyve might be a bit easier than lx since that's a _bit_ more self contained
[18:19:07] <KungFuJesus> more heavy handed, though. And we don't get to give it a datasets that's easy to just dynamically resize
[18:21:28] <igork> yuripv: i have proposed it too :)
[18:21:50] <dsockwell> Is there a well-regarded introduction or walkthrough for encrypted zpools? I'm trying to plan a transition of my FreeNAS box to SmartOS but coming up short on how FDE will work
[18:23:00] <dsockwell> I'm aware of efforts to integrate zfs encryption into a SmartOS/Triton specific service but I would rather have to transition again than wait
[18:24:07] <bahamat> dsockwell: You're better off waiting until SmartOS implements higher level encryption constructs.
[18:24:45] <bahamat> Whatever you come up with is almost certainly not going to match the pattern that SmartOS eventually uses, and you'll just have to convert or live with your one-off forever.
[18:25:37] <bahamat> We're actively working on it, and there are smaller changes going in all the time because we'll need it later for EDAR.
[18:26:31] <dsockwell> is EDAR the same as 'encrypted data at rest'?
[18:26:35] <bahamat> One example is early admin nic turn up (which also fixed MTUs on links with the admin nic tag)
[18:26:41] <bahamat> Yes, that's what that is.
[18:29:04] <dsockwell> oh right i remember some conundrum about where exactly the config files are stored. i was naively hoping i could come up with a key to unlock my datasets and then bodge that same key into rfd77 or rekey later (however painful that is)
[18:29:11] <KungFuJesus> What's somewhat weird is that the way to bootstrap a docker registry is evidently to spin up a docker container for it
[18:29:28] <dsockwell> and then things would more or less be the same
[18:29:29] <KungFuJesus> which isn't all that weird for the native docker configuration, but for lx branded zones you're double nesting containers
[18:29:45] *** Yogurt <Yogurt!~Yogurt@104-7-67-228.lightspeed.sntcca.sbcglobal.net> has joined #illumos
[18:30:19] <dsockwell> but since i haven't actually tried anything i'm sure there are issues beyond implementing the cryptosystem planned on top of just zfs encryption
[18:30:39] <dsockwell> now that i say that maybe the cryptosystem is the easy part
[18:30:46] <KungFuJesus> For anyone at Joyent - how thorough is your cgroup and namespaces interface?
[18:31:51] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)
[18:32:11] <dsockwell> i'd still like a knee up on zfs encryption if y'all would oblige, it's only a matter of time before i get asked about it professionally
[18:33:52] <bahamat> KungFuJesus: Limited. You can't just treat it like linux.
[18:35:01] <KungFuJesus> I figured. I'm sure triton's docker implementation would kinda sorta work with this if it were SmartOS
[18:35:45] <bahamat> dsockwell: Well, I'd start with the Encryption section in the man page. Until SmartOS has specific support for it, it's no different than illumos in general.
[18:36:03] <KungFuJesus> It's kind of dumb you can't just run a go based daemon to host the images directly. Maybe there's a way to do that, it's just something people try to do, I guess
[18:36:08] *** BH23 <BH23!~BH23@94.175.232.126> has quit IRC (Ping timeout: 245 seconds)
[18:37:00] <KungFuJesus> afterall, it's just hosting tarballs of root file systems, isn't it?
[18:38:58] <rmustacc> LeftWing, andyf: When prototyping a new driver and some other areas, I was using the BITX macro family and wrote some custom versions of it that more properly did type checking.
[18:39:36] <KungFuJesus> hosting the registry itself in its own container is kind of an annoying amount of dogfooding
[18:39:40] <rmustacc> LeftWing, andyf: But those really only work if you're using a design which is giving you bit ranges. If specifications are giving you masks and shifts, it's better to encode those in my opinion.
[18:39:45] <dsockwell> ok thanks bahamat. before i dive in, i'm guessing openzfs and oracle's zfs have wildly divergent encryption implementations?
[18:40:50] <bahamat> dsockwell: Almost certainly. If you've ever used Oracle's zfs encryption, I would suggest you try to just forget all about it when dealing with open zfs.
[18:44:51] <dsockwell> it may even be beneficial for me never to try the oracle way, part of the reason i like illumos/smartos is because i probably won't get asked about it at work. (is that different from hipsterism?)
[18:47:58] *** Kurlon <Kurlon!~Kurlon@cpe-67-253-141-249.rochester.res.rr.com> has joined #illumos
[18:50:52] *** wonko <wonko!~quassel@75.52.174.33> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[18:59:59] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has quit IRC (Ping timeout: 276 seconds)
[19:08:03] *** wonko <wonko!~quassel@75.52.174.33> has joined #illumos
[19:08:34] *** snuff-work <snuff-work!~snuff-wor@202-161-112-134.tpgi.com.au> has joined #illumos
[19:28:59] *** wonko <wonko!~quassel@75.52.174.33> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[19:34:42] *** wonko <wonko!~quassel@75.52.174.33> has joined #illumos
[20:07:00] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Quit: Leaving)
[20:07:35] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[20:08:00] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Client Quit)
[20:08:58] <andyf> dsockwell - jbk had a gist I think but the Encryption section of the zfs man page is pretty good
[20:13:42] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[20:13:42] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has quit IRC (Client Quit)
[20:15:18] <dsockwell> thanks, i just booted smartos in a vm to read the manpage
[20:15:40] <andyf> heh - https://illumos.org/man/zfs
[20:18:32] <dsockwell> that works too
[20:28:14] <gitomat> [illumos-gate] 11660 common.UTF-8.src should be compiled from CLDR data -- Yuri Pankov <yuri.pankov at nexenta dot com>
[20:30:30] *** mnowak_ <mnowak_!~mnowak_@94.142.238.232> has joined #illumos
[20:54:48] *** Teknix <Teknix!~pds@172.56.12.20> has joined #illumos
[21:00:55] *** wonko <wonko!~quassel@75.52.174.33> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[21:08:40] *** wonko <wonko!~quassel@75.52.174.33> has joined #illumos
[21:33:02] *** Teknix <Teknix!~pds@172.56.12.20> has quit IRC (Ping timeout: 265 seconds)
[21:45:30] *** wonko <wonko!~quassel@75.52.174.33> has quit IRC (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
[21:47:00] *** wonko <wonko!~quassel@167.71.188.230> has joined #illumos
[21:47:09] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has quit IRC (Quit: amrfrsh)
[21:52:30] <jbk> dsockwell: you can only set encryption when creating a dataset, and you need to specify the keylocation and keyformat parameters
[21:52:39] <jbk> unrelated
[21:52:43] <jbk> just as a sanity check
[21:52:57] <jbk> va_list ap;
[21:53:06] <jbk> va_start(ap, param);
[21:53:18] <jbk> vprintf(fmt, ap);
[21:53:25] <jbk> vprintf(fmt, ap);
[21:53:34] <jbk> would be wrong (to assume that ap is preserved)
[21:55:50] <jbk> actually it has to be
[21:56:06] <_mjg> i would argue you are trying to do something wrong here
[21:56:17] <jbk> well i'm not, but libcustr is :)
[21:56:21] <_mjg> :)
[21:56:28] <_mjg> argument stadns :-P
[21:57:06] <jbk> illumos#11710
[21:57:28] <jbk> hrm.. guess the bot doesn't paste links to tickets in here..
[21:57:38] *** amrfrsh <amrfrsh!~Thunderbi@134.19.189.92> has joined #illumos
[21:57:50] <richlowe> I don't think the bot is in here at all.
[21:57:57] <jbk> oh maybe just #smartos
[21:57:59] <richlowe> we just use LeftWing directly :)
[21:58:02] <jbk> haha
[21:58:11] <_mjg> https://www.illumos.org/issues/11710
[21:58:47] *** Riastradh <Riastradh!~riastradh@netbsd/developer/riastradh> has joined #illumos
[21:59:28] <_mjg> i think that's a different case than presented, iteration is only done once but it is done by a different function
[21:59:42] <_mjg> oh, it is twice, i did not notice it in the first place
[21:59:53] <jbk> yeah, the first call is to get the # of bytes needed
[21:59:58] <jbk> the second is to do the actual formatting
[21:59:59] <_mjg> ye pretty bad style
[22:00:05] <rmustacc> I believe our internal functions preserve it.
[22:00:23] <jbk> that doesn't seem to be documented
[22:00:38] <_mjg> what does the C standard say about it?
[22:00:46] <_mjg> it does sound quite undefine-y to me
[22:00:56] <richlowe> rmustacc: can we even not?
[22:01:10] <jbk> it causes OS X to crash
[22:01:12] <jbk> FWIW
[22:01:20] <rmustacc> Sure, different systems handle it differently.
[22:01:33] <rmustacc> I believe we always make a copy in the internal printf routines in libc.
[22:01:39] <rmustacc> But I suspect the standard doesn't say.
[22:02:30] <jbk> opengroup says ap is undefined after the call
[22:02:39] <jbk> err unspecified
[22:02:47] <richlowe> sounds like something smatch should check for us!
[22:03:32] <_mjg> hm
[22:03:38] <_mjg> i think this can actually be made to work
[22:03:43] <_mjg> according to c99:
[22:04:08] <_mjg> The object ap may be passed as an argument to another function; if that function invokes the va_arg macro with parameter ap, the value of ap in the calling function is indeterminate and shall be passed to the va_end macro prior to any further reference to ap.
[22:04:26] <_mjg> iow this should work if you push va* stuff inside an start/end across another vprintf
[22:04:45] <_mjg> This is 7.15
[22:05:30] <jbk> I was just going to propose this: https://gist.github.com/jasonbking/9ba5f528f7c08c290692794a51a2ddab
[22:06:19] <yuripv> jbk: http://src.illumos.org/source/xref/illumos-gate/usr/src/common/util/string.c#83 ?
[22:06:20] <rmustacc> _mjg: What's the title of 7.15? I only have C11 sitting around.
[22:06:46] <rmustacc> yuripv: Exactly, that's what I was describing.
[22:06:47] <_mjg> rmustacc: Variable arguments
[22:07:07] <_mjg> rmustacc: just search for va_start
[22:07:17] <rmustacc> Yeah, I found it there, just a single section off.
[22:07:52] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has quit IRC (Read error: Connection reset by peer)
[22:08:00] <yuripv> rmustacc: ah, sorry, didn't read the backscroll
[22:08:13] <rmustacc> No worries, helps to have the code I was describing.
[22:08:47] <rmustacc> What we have works on us, but certainly isn't portable or isn't promissed.
[22:09:06] <rmustacc> But if someone had knowledge of the implementation, it makes sense that they'd end up with what was written.
[22:09:16] *** wiedi <wiedi!~wiedi@ip5b4096a6.dynamic.kabel-deutschland.de> has joined #illumos
[22:09:20] <_mjg> it's an avoidable constraint
[22:09:35] <rmustacc> It's a shame the standard didn't force it itself.
[22:09:41] <rmustacc> For users of it.
[22:09:50] <rmustacc> But how that stuff goes.
[22:10:05] <_mjg> http://dpaste.com/2ATW7D6
[22:10:11] <_mjg> i think this is uglier but more likely to be correct
[22:10:35] <rmustacc> Yeah.
[22:11:00] <rmustacc> You do need a va_end if you va_copy().
[22:12:22] <_mjg> i find it slightly iffy that it's not the target func which is doing the iteration
[22:12:36] <_mjg> which already may be UB
[22:13:02] <_mjg> perhaps this weird wrapping can be rmeoved in the first place
[22:13:55] <_mjg> hm, there is one user elsewhere (libkmf)
[22:15:38] <_mjg> git grep on gate only shows custr_append_vprintf used in tests. should it turn out it is not used by anything it can be folded into custr_append_printf and the entire problem goes away
[22:16:30] <jbk> it's going to be used soon
[22:16:32] <jbk> :)
[22:16:59] * _mjg smells incomming segfaults
[22:17:37] <jbk> heh.. well that's how i found it :)
[22:18:27] <rmustacc> It won't segfault on us today because we always make copies...
[22:18:48] <jbk> i've finished up adding support for the new rust mangling format for illumos, and I guess my arguments for having our own were persuave enough, the guy that wrote the libiberty version is offering to run it through a few GB of test symbols (which necessitated creating a more portable version)
[22:19:21] <_mjg> i would argue the more replaceable your libc is the better
[22:19:33] <rmustacc> _mjg: Not saying we shouldn't change what we're doing.
[22:19:38] <rmustacc> Just that it won't actually segfault today.
[22:19:41] <_mjg> sure
[22:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)
[22:20:02] <rmustacc> I'm not even arguing against the change. Just likely how it got there.
[22:20:08] *** ed209 <ed209!~ed209@165.225.128.67> has joined #illumos
[22:21:51] <_mjg> on that note this one is funny https://reviews.freebsd.org/rS351251
[22:21:58] <_mjg> Remove some compatability with Seventh Edition UNIX realloc().
[22:22:07] <_mjg> you may have similar funzies
[22:22:42] <_mjg> my point was that perhaps it would make for a dcent project to run stuff from -gate on smething else than pure illumos if only to weed out this kind of stuff
[22:23:17] <jbk> heh
[22:23:21] <rmustacc> _mjg: Not disagreeing.
[22:23:49] <rmustacc> I suspect that's how this case was found.
[22:23:52] <_mjg> > In Seventh Edition UNIX, the last pointer passed to free() was guaranteed to not actually have been freed
[22:24:05] <_mjg> incredible
[22:24:34] <_mjg> rmustacc: well ye, i read it as it segfaulted on os x while preeping for someone else to test
[22:25:34] <rmustacc> I thought you were saying we should be prepared for that to crash on illumos, which was where my confusion stemmed from.
[22:26:21] <_mjg> perhaps i was unclear
[22:27:07] <rmustacc> Probably not.
[22:31:01] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has quit IRC (Remote host closed the connection)
[22:34:35] *** BH23 <BH23!~BH23@santoroj.plus.com> has joined #illumos
[22:34:35] *** freakazoid0223 <freakazoid0223!~matt@pool-96-227-98-169.phlapa.fios.verizon.net> has joined #illumos
[22:56:42] *** andy_js <andy_js!~andy@97e29e78.skybroadband.com> has quit IRC (Quit: andy_js)
[23:08:44] *** BH23 <BH23!~BH23@santoroj.plus.com> has quit IRC (Ping timeout: 265 seconds)
[23:17:03] *** clapont <clapont!~clapont@unaffiliated/clapont> has quit IRC (Ping timeout: 246 seconds)
[23:44:00] *** clapont <clapont!~clapont@unaffiliated/clapont> has joined #illumos
top

   September 18, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | >