Switch to DuckDuckGo Search
   January 8, 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 | 31

Toggle Join/Part | bottom
[00:01:45] <Smithx10> it looks like alpine, centos, and the others work.... just ubuntu:latest.... guessing they did something crazy with their userland.
[00:01:55] <Smithx10> probably the way they handle dns.
[00:11:48] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Remote host closed the connection)
[00:23:24] <Smithx10> From what I’ve read it looks like Ubuntu definitely did something
[00:33:37] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Quit: andy_js)
[00:43:02] <adam_vollrath> if nobody else says it, thanks for reporting!
[00:50:28] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[00:51:25] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Client Quit)
[00:56:39] <jbk> could be sendmmsg
[00:56:47] <jbk> IIRC some newer glibcs are using it for dns
[00:56:52] <jbk> (and isn't supported w/ lx yet)
[00:59:35] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[01:03:47] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Ping timeout: 240 seconds)
[01:13:41] *** noahmehl <noahmehl!~noahmehl@dsl-74-83-119-141.fuse.net> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[01:14:05] *** jubal <jubal!~jubal@247-16-237-24.gci.net> has joined #smartos
[01:15:37] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has quit IRC (Ping timeout: 258 seconds)
[01:18:33] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has joined #smartos
[01:18:48] *** jubal <jubal!~jubal@247-16-237-24.gci.net> has quit IRC (Ping timeout: 245 seconds)
[02:14:58] *** strobelight <strobelight!~quassel@c-69-180-7-202.hsd1.ga.comcast.net> has joined #smartos
[02:22:11] <Smithx10> jbk: so .... how hard is sendmmsg to add to lx :P ?
[02:29:45] <jubal_> hi all - i'm scratching my head a bit. I've got machines with onboard intel 82596 quad port nics - and under smartos 20181122 and 20190103 they kind of come up randomly with different igbX device names during boot
[02:30:00] <jubal_> anybody else come across this behaviour per chance?
[02:30:20] <jesse_> it's normal
[02:30:21] *** v_a_b <v_a_b!~volker@p57A27C24.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 252 seconds)
[02:30:46] <jesse_> that's why the normal config system defines the macs to bind to
[02:30:46] <jubal_> oi really? :P Is there any way to make it more deterministic?
[02:30:49] <jbk> Smithx10: i think it's more concerns about potential for abuse -- there's some scheduling work that'd need to happen in conjunction with implementing it (which in itself wouldn't be too bad)
[02:30:56] <jesse_> ...the macs?=)
[02:31:12] <jubal_> So I do give it the correct mac addresses from the cards in /usbkey/config
[02:31:22] <jesse_> yes
[02:31:25] <jubal_> but, still it only works sometimes :P
[02:31:25] <jbk> jubal_: use nictagadm to assign tags to specific MACs and then use that for config
[02:31:33] <jesse_> and then the interface name keeps changing randomly
[02:31:51] <jubal_> jesse_, yes --
[02:31:56] <jubal_> jbk - i'll try that again :)
[02:32:11] <jesse_> which sucks, if you run ipnat on the gz
[02:32:21] <jubal_> ah! i do
[02:32:25] <jubal_> yes
[02:32:44] <jesse_> there's not other fix than re-making your configs at boot, afaik
[02:33:09] <jubal_> got it - maybe i can wire something up in /opt/custom/smf
[02:33:37] <jesse_> sysinfo -p |fgrep <mac> |cut -d _ -f3
[02:33:45] <jesse_> probably will give you the interface name
[02:33:51] <jesse_> (lookign at my notes)
[02:34:13] <jubal_> jesse_, did you bodge something together to make this automatic in your case?
[02:34:24] <jesse_> halfway...ish
[02:34:33] <jubal_> lol, ok thanks
[02:34:37] <jesse_> I always make things a bit more automatic at every reboot
[02:34:56] <jesse_> and I moved to ipnat in a zone after that
[02:36:14] <jubal_> i'm switching my nat function to a dedicated device in a day or two - just kind of puzzled.. do people in the normal case just handle this via vlan tagging or something and not worry about the physical designations?
[02:36:25] <jubal_> like - is bug or feature?
[02:36:46] <jesse_> people just use nictagadm and infinite pool of ipv4 addresses
[02:36:53] <jesse_> (you should see the problem there)
[02:37:31] <jubal_> got it, ok :) i'm super glad i asked, thanks!
[02:37:54] <jesse_> happy to help, too me ....a while years ago=)
[02:37:56] <jbk> nic tags are how things in triton refer to interfaces, so the specific interface name doesn't matter
[02:38:02] <jesse_> *took
[02:38:18] <jesse_> jbk, except for ipnat that uses the actual interface name...
[02:38:40] <jbk> we don't use ipnat in the gz
[02:38:50] <jesse_> yes, you don't
[02:38:53] <jbk> so no one's written anything to make it understand nic tags
[02:39:00] <jbk> (it comes from ipf)
[02:39:29] <jesse_> anyway, smartos is not really ip-starved environments friendly
[02:39:39] <jesse_> but at least it does support ipv6 nowadays
[02:39:51] <jesse_> maybe in 50 years or so it'll be enough!
[02:40:24] <jbk> (I don't think anyone would be opposed to creating something that did, just not something we've needed to date)
[02:41:48] <jubal_> totally clears up my confusion :) switching to vlans when my router shows up, thanks fuys
[02:41:50] <jubal_> *guys
[03:02:07] <jubal_> so, i'm betting that folks with quad nics must have them all plugged in, in the normal case..in order for things to come up reliably and be device name agnostic? /me now rethinking the number of ports i spec'd on my router :P
[03:07:15] <Smithx10> jbk: do you have any idea why they decided to change the way made that change?
[03:07:23] <Smithx10> errrr......
[03:07:31] <Smithx10> Do you have any idea why they decided to make that change*
[03:12:11] <jesse_> jubal_, each port has the same mac address on every reboot
[03:12:33] <jesse_> jubal_, that's why the nic tag is tied to the mac
[03:13:27] <jesse_> jubal_, so on every boot, the system checks the nic tags and through the mac figures out the interface name and sets the system up
[03:14:59] <jesse_> so in a box with bnx0...bnx3, the interface assigned to 'external' or 'admin' tag keeps changing every reboot, but all zones (and gz init) using the tag names have ips on the right interfaces
[03:15:48] <jesse_> (the order depends on the order the interfaces initiaize on the pcie bus scan, which is random. So having all of them connected still randomizes the interface names)
[03:16:53] <jubal_> roger that - i figured it was random init order causing the entropy ---
[03:17:08] <jesse_> well, the root cause is that....
[03:17:25] <jubal_> i thought i could work around it by specifying in /usbkey/config igb3_nic=00:21:28:f8:d7:a3 - but alas, no dice
[03:17:51] <jesse_> it'll give you tag called 'igb3'
[03:18:02] <jesse_> ...which doesn't help with ipnat that expects interfaces=)
[03:18:04] <jubal_> or, more like it's a 4 sided dice roll :P
[03:18:26] <jesse_> yeah, tag igb3 will point to any random interface named igb0..3
[03:18:55] <jesse_> way of most astonishment=)
[03:19:08] <jubal_> even without ipnat though, if i've got igb0 and igb3 plugged in (igb0 as admin, and igb3 as external) and igb1 and two are not plugged in to the switch
[03:19:28] <jubal_> it'll come up and i've got to jack around in the wiring closet, randomly
[03:20:13] <jubal_> which seems....like i should have ordered a 16port managed switch instead of an 8port, fml
[03:21:28] <jubal_> i'm a gambling kind of guy, it's all good :P lol
[03:23:25] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos
[03:24:42] <jesse_> why doesn't it work?
[03:25:07] <jesse_> I mean, your zones specify 'external' as their interface and it will always map correctly
[03:26:08] <jesse_> and your gz will always bind to the interface with 'admin' mac address
[03:26:37] <jubal_> well, so, when i booted a minute ago, i had net3 (i assumed would be igb3) plugged physically into the lan switch, and net0 (assumed to be igb0 (and my admin net tag)) plugged into another switch
[03:26:55] <jubal_> when i booted, i had to move the physical cable from net3 to net1
[03:26:56] <jesse_> think of it this way:
[03:27:04] <jesse_> the name of the interface keeps changing
[03:27:15] <jesse_> but the mac address should be tied to the physial port
[03:27:50] <jesse_> unless that card somehow randomizes the mac addresses
[03:28:00] <jesse_> (that would be the first I've ever heard of)
[03:28:01] <jubal_> i could be sleepy, but this is what i'm seeing:
[03:30:03] <jubal_> with ifconfig igb3 output -- ether 0:21:28:f8:d7:a1
[03:30:15] <jubal_> oh!
[03:31:03] <jubal_> disregard.. i think. i am sleepy :P I'll try a couple of different scenarios after i get some coffee
[03:32:18] <jubal_> no yeah - so:
[03:34:04] <jubal_> no.. bleh, need caffiene. but, you're indicating i shouldn't have to physically switch cables around because the mac's are the mac's are the mac's - which seems rational and good
[03:34:33] <jubal_> thanks for helping me wrap my head around it
[03:44:06] *** jubal <jubal!~jubal@247-16-237-24.gci.net> has joined #smartos
[03:46:27] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has quit IRC (Ping timeout: 268 seconds)
[03:55:16] <Smithx10> w......t.......f........ https://github.com/docker/libnetwork/issues/2068
[04:02:28] <Smithx10> mgerdts: that ubuntu 17.10 thing.... is..... ' wtf is ubuntu doing'
[04:05:33] <Smithx10> I've made notes here: https://github.com/joyent/sdc-docker/issues/136 .
[04:27:07] <Smithx10> if this really is f****ing systemd ruining my life again
[04:27:17] <Smithx10> ./starts to cry.
[04:27:28] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has joined #smartos
[04:47:15] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[05:37:26] *** strobelight <strobelight!~quassel@c-69-180-7-202.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 258 seconds)
[05:56:52] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)
[06:01:24] *** tozhu <tozhu!~tozhu@223.104.170.115> has joined #smartos
[06:38:33] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Remote host closed the connection)
[06:45:23] *** tozhu <tozhu!~tozhu@223.104.170.115> has quit IRC (Quit: tozhu)
[06:50:15] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[07:05:05] *** tozhu <tozhu!~tozhu@223.104.170.115> has joined #smartos
[07:07:15] *** tozhu <tozhu!~tozhu@223.104.170.115> has quit IRC (Client Quit)
[07:21:47] *** tozhu <tozhu!~tozhu@223.104.170.115> has joined #smartos
[07:26:10] *** tozhu <tozhu!~tozhu@223.104.170.115> has quit IRC (Ping timeout: 250 seconds)
[07:28:35] *** tozhu <tozhu!~tozhu@171.34.78.71> has joined #smartos
[07:41:31] *** bdha <bdha!~bdha@165.225.136.186> has quit IRC (Ping timeout: 244 seconds)
[07:42:33] *** nahamu <nahamu!~nahamu@165.225.132.70> has quit IRC (Ping timeout: 245 seconds)
[07:42:33] *** bahamat <bahamat!~bahamat@72.2.113.233> has quit IRC (Ping timeout: 244 seconds)
[07:42:54] *** bahamat <bahamat!~bahamat@72.2.113.233> has joined #smartos
[07:44:28] *** tozhu <tozhu!~tozhu@171.34.78.71> has quit IRC (Quit: tozhu)
[07:45:18] *** fejfighter <fejfighter!~fejfighte@124.171.205.37> has joined #smartos
[07:45:53] *** nahamu <nahamu!~nahamu@165.225.132.70> has joined #smartos
[07:46:08] *** noahmehl <noahmehl!~noahmehl@dsl-74-83-72-196.fuse.net> has joined #smartos
[07:47:33] *** noahmehl <noahmehl!~noahmehl@dsl-74-83-72-196.fuse.net> has quit IRC (Client Quit)
[08:05:10] *** nahamu <nahamu!~nahamu@165.225.132.70> has quit IRC (Ping timeout: 250 seconds)
[08:05:39] *** nahamu <nahamu!~nahamu@165.225.132.70> has joined #smartos
[08:18:10] <sjorge> jperkin yeah, the wiki is stale at best. There is still some useful stuff there but it requires a lot of knowledge to figure that out from the bad
[08:32:10] *** tozhu <tozhu!~tozhu@171.34.78.71> has joined #smartos
[08:36:04] *** JeanParpaillon <JeanParpaillon!~jean@176.164.66.204> has joined #smartos
[08:39:28] *** v_a_b <v_a_b!~volker@p57A27DE9.dip0.t-ipconnect.de> has joined #smartos
[08:47:00] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)
[08:49:57] *** JeanParpaillon <JeanParpaillon!~jean@176.164.66.204> has quit IRC (Ping timeout: 252 seconds)
[09:05:52] *** mariusp <mariusp!~mpana@5.2.156.217> has joined #smartos
[09:05:54] *** DigiDaz <DigiDaz!sid115572@gateway/web/irccloud.com/x-gldswvpmtanvxbus> has quit IRC (Ping timeout: 264 seconds)
[09:06:07] *** mariusp <mariusp!~mpana@5.2.156.217> has quit IRC (Client Quit)
[09:07:06] *** DigiDaz <DigiDaz!sid115572@gateway/web/irccloud.com/x-vqvveydauseegubz> has joined #smartos
[09:09:32] *** mpana <mpana!~mpana@193.226.149.49> has joined #smartos
[09:33:57] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has quit IRC (Ping timeout: 252 seconds)
[09:35:42] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has joined #smartos
[09:41:28] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has joined #smartos
[09:57:25] <andyf> Smithx10: is it this one? https://github.com/omniosorg/illumos-omnios/issues/331
[10:11:59] *** tozhu <tozhu!~tozhu@171.34.78.71> has quit IRC (Quit: tozhu)
[10:13:45] *** man_u <man_u!~manu@manu2.gandi.net> has joined #smartos
[10:43:24] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has quit IRC (Remote host closed the connection)
[10:49:48] *** Latr_work <Latr_work!uid4039@gateway/web/irccloud.com/x-xzqrnvbgqsleksuj> has joined #smartos
[11:10:33] *** insane^ <insane^!~insane@fw.vispiron.de> has joined #smartos
[11:24:45] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has joined #smartos
[13:36:27] <Smithx10> andyf: not sure because this is in docker
[13:37:10] <Smithx10> but it is completely possible that systemd is on the scene there too messing things up :(
[13:40:43] <Smithx10> andyf: do you know if that happens on smartos?
[13:42:22] <andyf> no, sorry. I just saw you mention systemd and remembered that recent problem on OmniOS with ubuntu in an lx zone.
[13:42:27] <andyf> jbk's hunch is probably closer
[13:42:49] <Smithx10> I think its strange that its only ubuntu tho
[13:43:35] <Smithx10> all the other latest dont have an issue :( unless none of them update glibc
[13:49:04] <Smithx10> andyf: I noted that issue in the github issue
[13:49:12] <Smithx10> Thanks for the info
[13:56:31] <tuxillo> hi
[13:57:35] <tuxillo> jperkin: got the "create a zone to build your signed packages" entry at hand?
[13:57:39] <tuxillo> or remember the name
[13:58:11] <jperkin> https://github.com/joyent/pkgsrc/wiki/pkgdev:signing
[13:58:56] <tuxillo> that was fast, thanks :)
[14:00:54] *** jellydonut <jellydonut!~jelly@77.16.70.233.tmi.telenormobil.no> has joined #smartos
[14:01:53] *** neuroserve <neuroserve!~toens@195.71.113.124> has joined #smartos
[14:16:35] *** tozhu <tozhu!~tozhu@223.104.170.115> has joined #smartos
[14:22:59] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos
[14:32:22] *** jellydonut <jellydonut!~jelly@77.16.70.233.tmi.telenormobil.no> has quit IRC (Read error: Connection reset by peer)
[14:41:54] *** jellydonut <jellydonut!~jelly@77.16.70.233.tmi.telenormobil.no> has joined #smartos
[14:44:26] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[15:20:03] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has quit IRC (Ping timeout: 245 seconds)
[15:22:15] *** polishdub <polishdub!~polishdub@ip72-221-23-237.ph.ph.cox.net> has joined #smartos
[15:42:19] *** tozhu <tozhu!~tozhu@223.104.170.115> has quit IRC (Quit: tozhu)
[15:58:35] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has joined #smartos
[15:58:39] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos
[15:59:50] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Quit: andy_js)
[16:03:12] *** andy_js <andy_js!~andy@static-108-31-135-163.washdc.fios.verizon.net> has joined #smartos
[16:08:01] *** andy_js <andy_js!~andy@static-108-31-135-163.washdc.fios.verizon.net> has quit IRC (Ping timeout: 258 seconds)
[16:08:08] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[16:19:26] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos
[16:26:06] *** neuroserve <neuroserve!~toens@195.71.113.124> has quit IRC (Ping timeout: 250 seconds)
[16:50:59] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[16:52:28] *** strobelight <strobelight!~quassel@2001:420:3042:1320:c19e:2b3f:50e6:e4eb> has joined #smartos
[17:07:02] <Smithx10> Is there a reason https://cr.joyent.us/#/c/5290/1/lib/vmapi.js doesn't use assert-plus?
[17:08:21] *** insane^ <insane^!~insane@fw.vispiron.de> has quit IRC (Ping timeout: 260 seconds)
[17:19:46] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Quit: andy_js)
[17:24:07] <jbk> age probably
[17:24:29] <Smithx10> k, was just curious
[17:42:08] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[17:55:34] <rsully> Can I pass through or use an integrated gpu into a zone or VM, or only discrete graphics cards?
[17:56:47] <Smithx10> rsully: I believe you have to pass the entire card into a Bhyve VM at this point
[17:56:52] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Quit: 410 Gone)
[17:57:40] <rsully> Yeah it sounds like discrete cards work, I wasn't sure about integrated graphics on the CPU (e.g. intel p630)
[17:57:56] <rmustacc> rsully: It's at a PCIe level. So those should show up as discrete PCIe devices.
[17:58:06] <rmustacc> However, the OS is usually using the discrete framebuffer.
[17:58:33] <rsully> Can I tell smartos no touchy?
[17:58:44] <rmustacc> You can't really... beacuse of the BIOS, etc.
[17:58:55] <rmustacc> x86 kind of expects a local framebuffer.
[17:59:32] <rsully> Hm I'm sure this is a dumb question but how does it work for CPUs that don't have integrated graphics?
[17:59:54] <rmustacc> Well, for example, server motherboards have a small vga chipset.
[18:00:25] <rmustacc> Most desktops I don't think you could do the BIOS config to get the serial console config to function, tbh.
[18:00:28] <rmustacc> But I don't know off hand.
[18:00:48] <rsully> Sorry could you reword that
[18:01:08] <rmustacc> So servers have a separate vga chipset on the motherboard.
[18:01:21] <rmustacc> That part make sense at least?
[18:01:23] <rsully> Yep
[18:01:31] <rmustacc> For desktops, I honestly don't know how you'd get it to work.
[18:01:41] <rmustacc> Because you usually don't have the BIOS come up over serial by default.
[18:02:22] <rmustacc> But it's been a while since I poked at anything like this.
[18:02:39] <rsully> Hm, so if I can find a server board that can accept a CPU with integrated graphics, then it would be possible to use the vga chipset on the board for bios/smartos, and the CPU's graphics in bhyve?
[18:02:50] <rmustacc> Uh...
[18:02:54] <rsully> or is that not a thing
[18:02:59] <rmustacc> I have no idea.
[18:03:01] <rmustacc> Sorry.
[18:03:03] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #smartos
[18:03:13] <rmustacc> Probably depends on the BIOS.
[18:03:25] <rmustacc> And system board manufacturer.
[18:03:32] <rsully> assuming someone even makes a server board that accepts that socket
[18:03:43] <rmustacc> Again, not really sure.
[18:03:51] <rmustacc> There may be client boards that have a separate gpu chipset on them.
[18:04:06] <rsully> hmm ok good to know
[18:04:15] <rmustacc> I don't know, it's not something I've ever really explored.
[18:04:24] <rsully> so sounds like my best bet is to add a separate board with a gpu, whether it be a PCI graphics card or some type of client board
[18:06:35] <arekinath> most desktop/server boards with integrated graphics I've seen completely disable it when you switch to an external GPU (i.e. it no longer enumerates on the bus at all)
[18:06:36] <rmustacc> Maybe, yeah. Perhaps look if there's a way on Linux/Windows.
[18:06:59] <rmustacc> But I think do the prevalence of using the graphical console...
[18:07:29] <arekinath> or... wait
[18:07:37] <rsully> hmm?
[18:07:48] <arekinath> you want to use like the crappy matrox card on a server board and pass the intel graphics on the cpu through?
[18:08:18] <rsully> yeah if I go the iGPU route I'd want the crappy server board graphics to be used, and pass thru the intel graphics
[18:08:42] <rsully> OR my other option is to go with a server board, and use it's crappy graphics, and then buy a separate GPU board to pass thru (i think)
[18:09:24] <arekinath> yeah I'm pretty sure your firmware will nuke the intel graphics as soon as it starts using the thing on the board during boot
[18:09:35] <arekinath> and the OS won't be able to enumerate the intel graphics at all
[18:09:42] <rsully> so that would mean (1) is no good, but (2) is possible
[18:09:45] <rsully> (?)
[18:09:53] <arekinath> right
[18:10:03] <rsully> ok the way I read it at first sounded like neither would work
[18:10:36] <arekinath> just be sure to instruct the firmware/bios/efi to boot from the server board graphics and not the GPU you added
[18:10:36] <rsully> arekinath for reference which brand boards have you seen do this type of behavior? It doesn't surprise me, but definitely is disappointing
[18:10:53] <rsully> yep good point
[18:11:39] <arekinath> so, I've mostly tried things like this and seen it on workstation class boards by smci and e.g. gigabyte's xeon e3 boards
[18:11:57] <arekinath> I haven't personally tried it on a server class board taht I remember
[18:12:11] <arekinath> but I doubt they would do it differently tbh
[18:12:48] <rsully> yeah I wouldn't see why they would :/
[18:16:28] <rsully> now as far as native/LX zones - can they use PCI or iGPU at all, or am I limited to bhyve hw pass thru?
[18:16:53] <arekinath> native/LX zones are limited to whatever hardware drivers the main OS kernel supports
[18:17:04] *** bdha <bdha!~bdha@165.225.136.186> has joined #smartos
[18:17:04] <arekinath> and we don't currently have a GPU driver for e.g. GPGPU stuff
[18:17:14] <arekinath> and the one we have for graphics is pretty minimal
[18:17:37] <rsully> ahh ok makes sense
[18:17:41] <rsully> bhyve it is
[18:23:06] <Smithx10> arekinath: can I pass more than 1 gpu?
[18:25:27] <arekinath> I believe so
[18:26:04] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[18:32:20] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Remote host closed the connection)
[18:36:47] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has quit IRC (Quit: bens1)
[18:38:09] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)
[18:38:24] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[18:39:51] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has joined #smartos
[18:47:20] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Remote host closed the connection)
[18:49:58] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has joined #smartos
[18:51:25] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[18:52:55] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Remote host closed the connection)
[18:55:48] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[19:40:17] *** jellydonut <jellydonut!~jelly@77.16.70.233.tmi.telenormobil.no> has quit IRC (Read error: Connection reset by peer)
[19:44:08] *** polishdub <polishdub!~polishdub@ip72-221-23-237.ph.ph.cox.net> has quit IRC (Ping timeout: 250 seconds)
[19:45:45] *** blackwood821 <blackwood821!ae3110b0@gateway/web/freenode/ip.174.49.16.176> has joined #smartos
[19:46:20] <blackwood821> hi, I'm trying to add a new vnic to a smartos zone and I can ping the IP I assigned to the NIC from the zone, but I can't ping the gateway of the VNIC
[19:46:23] *** tozhu <tozhu!~tozhu@171.223.92.164> has joined #smartos
[19:46:30] <blackwood821> but I can ping the gateway from the global, any ideas on what I'm doing wrong?
[20:00:46] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[20:04:05] <colstrom> @blackwood821 Maybe something in `nics.*.allowed_ips`?
[20:06:36] *** JeanParpaillon <JeanParpaillon!~jean@176.164.1.146> has joined #smartos
[20:27:02] <LeftWing> blackwood821: Can you gist/pastebin what "vmadm get" tells you for the VM?
[20:39:50] *** JeanParpaillon <JeanParpaillon!~jean@176.164.1.146> has quit IRC (Quit: JeanParpaillon)
[20:47:02] *** jellydonut <jellydonut!~jelly@77.16.78.99.tmi.telenormobil.no> has joined #smartos
[20:52:35] *** axonpoet <axonpoet!~axonpoet@fsf/member/axonpoet> has quit IRC (Ping timeout: 268 seconds)
[21:01:11] *** moatsad <moatsad!~dougt@69.194.139.196> has joined #smartos
[21:07:51] <blackwood821> LeftWing: figured it out, thanks
[21:22:35] <rsully> blackwood821 what was the problem?
[21:26:23] <blackwood821> rsully: I originally created a VNIC on the global and then I tried adding a new NIC to the zone that used the nic tag of the VNIC on the global
[21:26:28] <blackwood821> but it didn't seem to work
[21:26:57] <blackwood821> I was able to get it working by removing the VNIC on the global, adding a nic tag to a physical NIC on the global, then using that nic tag when adding a NIC to the zone
[21:27:41] <blackwood821> not quite sure why it didn't work my mapping the zone NIC to a VNIC but mapping the zone NIC to the nic tag of a physical NIC worked
[21:45:44] *** fejfighter <fejfighter!~fejfighte@124.171.205.37> has quit IRC (Ping timeout: 258 seconds)
[23:03:10] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 258 seconds)
[23:10:28] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos
[23:10:48] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[23:33:46] *** greenline <greenline!~user@unaffiliated/greenline> has joined #smartos
[23:51:09] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has joined #smartos
[23:56:46] *** blackwood821 <blackwood821!ae3110b0@gateway/web/freenode/ip.174.49.16.176> has quit IRC (Quit: Page closed)
top

   January 8, 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 | 31