January 21, 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 | >
January 21, 2019 [00:01:19] <LeftWing> jrg: Are they both KVM VMs?[00:01:39] <LeftWing> You might want to try "vmadm stop -F <uuid>" -- without -F, we try to get the guest to power itself off by pressing the emulated "power button"[00:01:53] <LeftWing> With -F we mostly just shut it off as fast as possible[00:04:43] <jrg> LeftWing: yes they are. one eventually stopped. the other i just killed[00:04:56] <jrg> but i did get zimbra to work with zmconfigd now.. only thing that isn't working is dnscache[00:05:03] <jrg> [1548025440] unbound[66713:0] error: setsockopt(..., IP_MTU_DISCOVER, IP_PMTUDISC_OMIT...) failed: Operation not supported[00:05:03] <jrg> [1548025440] unbound[66713:0] fatal error: could not open ports[00:05:39] <jrg> that looks like some awkward sockets error.. maybe zone related?[00:09:56] <LeftWing> I suspect we don't support the IP_MTU_DISCOVER sockopt in a zone[00:10:06] <LeftWing> It's possible the software isn't configured to skip setting it if it isn't supported[00:11:36] <jrg> i just removed it from the zimbra install[00:11:41] <jrg> i doubt in my case the dnscache is that important[00:12:35] <jrg> but at least i know it can run in a zone minus the dns cache. i had another issue with zmconfigd but i think that was just an issue with the hosts file... which is silly considering i have everything configured on the local dns. but it is what it is.. i'm going to test the install a bit more[00:14:03] <jrg> i guess after changing the hosts file and removing the dnscache ... zimbra is flying now heh[00:14:23] *** leoj3n <leoj3n!~leoj3n@2601:246:100:a5:a94b:495f:8b3:6a35> has quit IRC (Ping timeout: 268 seconds)[00:15:26] <jrg> now all i have to do is figure out what happened to ssh :/[00:27:22] *** aszeszo_ <aszeszo_!~aszeszo@unaffiliated/aszeszo> has quit IRC (Quit: aszeszo_)[00:28:28] <npx> Is it possible to do secure deployments of triton without physically isolating the admin/external/underlay networks as long as zones can't provision untagged vnics?[00:29:38] <LeftWing> npx: I don't think we would ever make any promises about a configuration which doesn't put "admin" and "underlay" on their own VLAN.[00:29:58] <npx> I mean, for single tenant systems I'm not convinced there is even a meaningful distinction between these networks in the first place but just prohibiting untagged traffic from zones[00:30:15] <LeftWing> There absolutely is a distinction, even in single tenant systems.[00:30:59] <npx> didn't finish that sentence, "prohibiting untagged traffic from zones seems like it would prohibit access to the untagged admin network"[00:31:07] <LeftWing> Access to the "admin" network allows you to do a lot of privileged things. If your single tenant workload experiences some kind of compromise, having no isolation from "admin" would make it extremely easy to then have access to other parts of the system.[00:33:24] <npx> That's fair. Just having access to the SSH server on CNs doesn't seem like a particularly huge deal to me, but it'd be hilarious to initiate a reboot and PXE boot a malicious system image.[00:34:08] * danderson pines for secure PXE boot[00:34:15] <danderson> all the infrastructure's there, just nobody uses it :([00:35:32] <LeftWing> I mean, there are some pieces in place[00:35:42] <LeftWing> Actually using it requires a lot of thought, and building extra pieces, really.[00:36:55] <npx> LeftWing, it was you who wrote the data-at-rest RFD? That there is some good stuff.[00:37:38] <LeftWing> I believe you're thinking of arekinath[00:37:49] <npx> ah yeah[00:47:56] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[00:51:29] *** iphy <iphy!uid67735@gateway/web/irccloud.com/x-plbqcwywtyvsxekz> has quit IRC (Quit: Connection closed for inactivity)[00:54:26] <npx> I'll note that the Project FiFO VPN overlay networking is pretty compelling. I'd love to see something similar in Triton. Essentially, it allows containers to have their fabric network be directly addressable across datacenters.[00:56:06] <LeftWing> It would be possible to add that to Triton directly, as well. The underlying transport is just VXLAN, so it can traverse any IP network -- even across routed hops, VPNs, etc.[00:56:28] <LeftWing> The real challenge is doing something useful with respect to discovery of and interaction with the subnets in remote DCs, which are effectively all isolated today.[00:58:36] <npx> I think Consul, RethinkDB, TiDB, maybe CochroachDB and Prometheus and others, are all multi-DC aware. So there is a growing amount of upstack utility there.[01:03:06] <arekinath> I think LeftWing means upstack within triton -- e.g. napi which today is not really aware of other datacentres other than the one it's running in[01:03:22] <arekinath> really only sapi, ufds and binder do any cross-DC stuff today and the way they do it is pretty hacky[01:03:51] <npx> ah, I see your point, I thought the question was "why would any user want that?"[01:03:54] <jbk> and you really need to be careful in how you'd make it know about cross-DC to prevent problems from cascading in even bigger failures[01:04:24] <jbk> because things will fail at some point[01:04:33] <arekinath> right, we want to be very careful not to introduce any cross-DC failure modes beyond the absolute minimum[01:04:41] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[01:04:50] <arekinath> (i.e. we want to preserve the property that if you're not doing something that must cross DCs, it isn't affected by another DC's availability)[01:05:03] <jbk> (and if any software is claiming otherwise, they are lying to you :P)[01:11:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[01:11:15] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[01:37:57] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has joined #smartos[01:49:38] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Quit: andy_js)[01:53:18] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[01:54:12] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[01:54:38] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[01:58:50] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Ping timeout: 244 seconds)[02:04:13] <Smithx10> jbk: .... What Software Company would ever lie?[02:04:25] <Smithx10> :P[02:04:53] <Smithx10> I'm playing around with Jaegor and writting Spans for tracing in Virtual Kubelet, I'm so f'n Fancy.[02:11:09] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos[02:12:27] <npx> Smithx10, planning to try out the virtual kubelet this week, I love that concept[02:12:58] <Smithx10> oOOoo i have a triton provider working, I'm adding some small things to make it a bit better[02:13:02] <Smithx10> but currently its functional[02:13:18] <npx> https://srcco.de/posts/kubernetes-failure-stories.html - I was reading the HN discussion about this and the recent virtualized kubelet article and what I actually want is the K8S API with all of the runtime properties of Triton[02:13:37] <Smithx10> yea[02:13:42] <Smithx10> thats what I did :)[02:13:52] <npx> e.g., no virtual machines although I'm not sure I can actually run K8S + virtkubelet in an LX branded zone[02:14:05] <Smithx10> lolol[02:14:18] <npx> Have you tried it?[02:14:22] <Smithx10> https://github.com/Smithx10/virtual-kubelet/tree/triton[02:14:31] <Smithx10> I Am Writing it :)[02:15:10] <Smithx10> Pay no mind to all the awful ssh code in there, i was testing kubectl exec and logs using ssh[02:15:32] <Smithx10> I think I might just "NoOp" those for the time being[02:15:44] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[02:15:50] <npx> I don't think that the requirement of a VM to run K8S would totally destroy the usefulness of this, I don't even particularly like k8s but lots of people do... and they'll pay me to do stuff[02:15:58] <Smithx10> huh?[02:16:23] <Smithx10> Virtual Kubelet allows you to schedule whatever you want[02:16:48] *** v_a_b <v_a_b!~volker@p4FECCAEC.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 246 seconds)[02:18:03] <npx> Smithx10, the question is, can you run K8S + virtual kubelet in an LX branded zone and use that to schedule containers on Triton? Does K8S still need to be ran in a Linux VM?[02:18:32] <Smithx10> Virtual Kubelet = Go binary.[02:18:42] <Smithx10> So it works inside of LX or SmartOS[02:19:04] <Smithx10> and yes, the provider I am working on will schedule against cloudapi, and dockerapi[02:19:22] <npx> sweet, I'll try it out this week[02:19:30] <Smithx10> https://github.com/Smithx10/virtual-kubelet/blob/triton/providers/triton/provider.go[02:19:40] <Smithx10> See CreatePod[02:44:46] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[02:45:39] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[02:55:47] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[03:10:08] *** bahamas10 <bahamas10!~dave@cpe-98-5-18-139.buffalo.res.rr.com> has quit IRC (Ping timeout: 268 seconds)[03:20:34] <Smithx10> Does triton supported rack affinity?[03:30:40] *** bahamas10 <bahamas10!~dave@cpe-72-231-190-251.nycap.res.rr.com> has joined #smartos[03:35:02] <Smithx10> all the servers have the ability to be tagged with a rack, it would be good to have :)[03:36:50] <npx> there is an RFD about rack-aware networking, I think it's partially implemented[03:37:48] <npx> (but this does bring to mind some interesting possibilities in terms of data locality across datacenters, possibly in conjuncton with the hypothetical VPN overlay stuff I was thinking of earlier)[03:38:52] *** v_a_b <v_a_b!~volker@p57A27C6C.dip0.t-ipconnect.de> has joined #smartos[03:41:21] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:41:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:42:06] <Smithx10> yea, thats fancy and all... but I just want to provision instances in different racks.[03:42:10] <Smithx10> :)[03:43:34] <Smithx10> I believe the goals of that RFD are different to affinity rules.[03:46:16] *** bahamas10 <bahamas10!~dave@cpe-72-231-190-251.nycap.res.rr.com> has quit IRC (Read error: Connection reset by peer)[03:46:26] *** bahamas10 <bahamas10!~dave@cpe-72-231-190-251.nycap.res.rr.com> has joined #smartos[03:52:39] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)[03:56:17] *** bahamas10 <bahamas10!~dave@cpe-72-231-190-251.nycap.res.rr.com> has quit IRC (Read error: Connection reset by peer)[03:57:07] *** bahamas10 <bahamas10!~dave@cpe-72-231-190-251.nycap.res.rr.com> has joined #smartos[04:20:50] *** tozhu <tozhu!~tozhu@101.207.125.86> has joined #smartos[04:51:20] <Smithx10> Does Triton instances not have a "restart policy"[04:55:14] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[05:08:46] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[05:18:23] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Ping timeout: 245 seconds)[05:25:49] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[05:52:09] *** kollross <kollross!~kollross@50-82-219-20.client.mchsi.com> has joined #smartos[06:35:01] *** tozhu <tozhu!~tozhu@101.207.125.86> has quit IRC (Quit: tozhu)[06:55:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[06:56:05] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:01:12] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[07:42:08] *** zaolin <zaolin!~quassel@2001:67c:2050:320::74> has quit IRC (*.net *.split)[07:42:08] *** LeftWing <LeftWing!~jclulow@165.225.151.150> has quit IRC (*.net *.split)[07:42:43] *** zaolin <zaolin!~quassel@2001:67c:2050:320::74> has joined #smartos[07:42:46] *** LeftWing <LeftWing!~jclulow@165.225.151.150> has joined #smartos[07:44:45] *** shaner <shaner!~shaner@rrcs-69-135-23-210.central.biz.rr.com> has quit IRC (Ping timeout: 246 seconds)[07:45:30] *** shaner <shaner!~shaner@rrcs-69-135-23-210.central.biz.rr.com> has joined #smartos[08:52:07] *** JeanParpaillon <JeanParpaillon!~jean@176.164.117.233> has joined #smartos[08:59:12] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[09:06:49] *** aszeszo <aszeszo!~aszeszo@unaffiliated/aszeszo> has joined #smartos[09:18:58] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)[09:28:52] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has joined #smartos[10:15:09] *** neuroserve <neuroserve!~toens@195.71.113.124> has joined #smartos[10:32:45] *** iphy <iphy!uid67735@gateway/web/irccloud.com/x-cipwkqyawouirfsr> has joined #smartos[10:33:15] *** man_u <man_u!~manu@manu2.gandi.net> has joined #smartos[10:45:14] <sjorge> I assume you did not have time to look at https://cr.joyent.us/#/c/4984/ again LeftWing ?[10:48:16] <sjorge> I ran into that one again this weekend[11:06:31] *** tozhu <tozhu!~tozhu@171.221.44.58> has joined #smartos[11:42:50] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[13:05:49] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos[13:14:51] *** aszeszo_ <aszeszo_!~aszeszo@unaffiliated/aszeszo> has joined #smartos[13:15:07] *** aszeszo <aszeszo!~aszeszo@unaffiliated/aszeszo> has quit IRC (Ping timeout: 240 seconds)[13:52:27] *** aszeszo_ <aszeszo_!~aszeszo@unaffiliated/aszeszo> has quit IRC (Read error: Connection reset by peer)[14:07:35] *** strobelight <strobelight!~quassel@2001:420:c0c8:1003::18f> has joined #smartos[14:39:08] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has quit IRC (Remote host closed the connection)[14:43:06] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has joined #smartos[15:11:16] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos[15:13:00] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has joined #smartos[15:13:29] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has quit IRC (Remote host closed the connection)[15:14:04] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[15:17:24] <TyrfingMjolnir> How can I launch a macOS VM in bhyve?[15:17:27] <TyrfingMjolnir> My environment is Joyent SmartOS on MacMini 2018 w/64GB RAM[15:39:20] <shaner> bahamat: I'm logged into the wiki but I don't see where to create a new page/article.[15:45:47] *** npx <npx!~alex@104.222.153.117> has quit IRC (Ping timeout: 240 seconds)[15:52:56] *** JeanParpaillon <JeanParpaillon!~jean@176.164.117.233> has quit IRC (Quit: JeanParpaillon)[15:58:32] *** verin0x <verin0x!~jrt@z6332.pia.fu-berlin.de> has joined #smartos[16:08:27] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has joined #smartos[16:11:18] *** verin0x <verin0x!~jrt@z6332.pia.fu-berlin.de> has quit IRC (Ping timeout: 245 seconds)[16:12:09] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 268 seconds)[16:13:07] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has quit IRC (Ping timeout: 240 seconds)[16:24:46] *** polishdub <polishdub!~polishdub@207.86.38.254> has joined #smartos[16:26:03] *** verin0x <verin0x!~jrt@z6332.pia.fu-berlin.de> has joined #smartos[16:28:42] *** neuroserve <neuroserve!~toens@195.71.113.124> has quit IRC (Ping timeout: 246 seconds)[16:45:40] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[16:49:56] <adam_vollrath> Has anyone used configuration management tools like Ansible on SmartOS?[16:53:18] *** npx <npx!~alex@136.41.160.10> has joined #smartos[17:28:02] <papertigers> npx: did you get your answers?[17:28:43] <papertigers> npx: RUSTFLAGS="-C force-frame-pointers" cargo build --release[17:30:03] <papertigers> and you can set that always in ".cargo" https://doc.rust-lang.org/cargo/reference/config.html[17:30:13] <papertigers> via rustflags[17:32:14] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[17:40:45] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos[17:41:41] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Quit: andy_js)[17:44:21] *** ingenthr <ingenthr!~ingenthr@cpe-108-184-198-4.socal.res.rr.com> has joined #smartos[17:44:52] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[17:54:26] *** verin0x <verin0x!~jrt@z6332.pia.fu-berlin.de> has quit IRC (Ping timeout: 268 seconds)[17:54:38] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 245 seconds)[17:55:17] <npx> papertigers, not really, got sidetracked, but I didn't notice anything special in the Cargo.toml in your repos[17:55:23] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Quit: andy_js)[17:56:32] <npx> ah just noticed, thanks![17:59:04] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[18:02:49] <jrg> im looking at ways to mount smb into lx zones[18:03:05] *** fxhp <fxhp!~fox@138.207.192.33> has quit IRC (Quit: WeeChat 1.9.1)[18:03:12] <jrg> im guessing direct mounting while in the zone is not possible?[18:03:33] <jrg> id have to mount on the smartos side of it and mount it into the zone?[18:04:19] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has quit IRC (Quit: Lost terminal)[18:04:39] <papertigers> jrg: you mean like a lofs mount?[18:04:53] <papertigers> that's how I typically mount stuff into my native and lx zones[18:05:13] <papertigers> since a zone can't see anything in the rest of the system unless you expose it[18:05:41] <papertigers> npx: I also think there is build.rs, that might give more flexibility. Or just a standar Makefile[18:10:58] <npx> I'm kinda feeling the whole Cargo thing, writing gmakefiles kinda sucks[18:11:23] <npx> and writing dmakefiles sucks even more[18:12:11] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #smartos[18:12:40] <npx> Smithx10, are you using upstream K8S binaries? Even the build system assumes you're communicating with the Docker daemon over a Unix domain socket, heh[18:18:36] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 246 seconds)[18:22:34] <npx> (I wanted to build kubernetes natively on smartos and have it only use virtual kubelet, but the same thing in an LX zone doesn't hurt my feels)[18:25:13] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #smartos[18:27:05] *** sjorge_be <sjorge_be!~sjorge@unaffiliated/sjorge> has joined #smartos[18:29:38] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 245 seconds)[18:29:38] *** sjorge_be is now known as sjorge[18:30:22] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has quit IRC (Quit: bens1)[18:33:48] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 245 seconds)[18:35:23] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)[18:42:09] <bahamat> shaner: Add it under here: https://wiki.smartos.org/display/DOC/How+to+create+a+KVM+VM+%28+Hypervisor+virtualized+machine+%29+in+SmartOS[18:42:38] <bahamat> shaner: if you're logged in, there should be an "+ Add" button in the upper right. corner.[18:57:46] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Ping timeout: 250 seconds)[19:03:50] *** fxhp <fxhp!~fox@138.207.192.33> has joined #smartos[19:05:04] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[19:05:18] <jrg> papertigers: lofs is like nullfs?[19:07:06] <papertigers> jrg: not all that familar with nullfs but it looks like a similar concept[19:07:15] <jrg> ah ok.[19:07:45] <papertigers> basically a loopback filesystem[19:08:01] <jrg> I’m a bit confused. the wiki on how to mount smbfs in smartos says to use /etc/vfstab[19:08:05] <jrg> but that wouldn’t be persistent would it?[19:08:34] <papertigers> its probably assuming youre mounting it within a zone[19:09:20] <papertigers> I have a few zones that mount nfs shares as well as their lofs mounts[19:09:26] <jrg> https://help.joyent.com/hc/en-us/articles/226687207-Mounting-an-SMB-share-on-a-SmartOS-Instance[19:10:00] <jrg> papertigers: seems like there would be issues with smb unless you configure winbind in the lx zone[19:10:16] <jrg> which i doubt is a bit problem I’d guess[19:11:19] <papertigers> i personally havent tried smb in lx but i know we did work for nfs to work. And I do have that setup[19:12:24] <jrg> I’ll try it some time next weekend and see if i can get it working[19:12:37] <jrg> it would be nice to migrate off kvm as much as possible. i’m sure the lx zones are a bit more resource friendly[19:16:14] *** waz0wski <waz0wski!waz0wski@hrothgar.distortion.io> has quit IRC (Quit: quit)[19:16:40] <Smithx10> Npx.. what[19:17:12] <npx> Smithx10, hahaha I just always default to trying to build things natively on smartos (even when it seems insane)[19:17:19] <npx> because it's never *actually* insane[19:25:57] <Smithx10> I don’t know what your question is[19:28:24] *** waz0wski <waz0wski!waz0wski@hrothgar.distortion.io> has joined #smartos[19:30:22] <npx> Smithx10, my question is, are you using a Linux distribution's K8S build?[19:31:01] <Smithx10> SmartOS doesnt run a docker daemon.[19:34:14] <Smithx10> I'd suggest reading about the Virtual Kubelet[19:36:27] <Smithx10> OooO nooo mike gerdts isnt in irc :([19:36:30] <Smithx10> ./world ends for me[19:36:40] <npx> I'm aware of the distinction, I have two Triton "DCs" in this room hahah, if K8S didn't assume you were always communicating with Docker over a unix socket then their build system ought to "just work" against a Triton DOCKER_HOST[19:37:16] <pmooney> Smithx10: Holiday - MLK Jr Day[19:37:40] <Smithx10> ahhh true[19:37:43] <Smithx10> its my birthday!!!! tooo[19:38:12] <Smithx10> I was gonna ask him if this was feasible https://github.com/joyent/sdc-docker/issues/137[19:38:44] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has joined #smartos[19:39:18] <pmooney> ah, happy birthday[19:39:34] <Smithx10> Thanks pmooney , you rock![19:39:37] <Smithx10> I went for beers at lunch[19:39:40] <Smithx10> ./feels g00d[19:40:51] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has quit IRC (Ping timeout: 246 seconds)[19:42:05] *** ingenthr <ingenthr!~ingenthr@cpe-108-184-198-4.socal.res.rr.com> has quit IRC (Quit: Leaving.)[19:42:35] *** ingenthr <ingenthr!~ingenthr@cpe-108-184-198-4.socal.res.rr.com> has joined #smartos[19:45:53] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has quit IRC (Quit: use Unix or die.)[19:49:28] *** scarcry <scarcry!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has joined #smartos[19:57:18] <npx> https://github.com/tokio-rs/tokio - this looks to be within tolerable confines of awesomeness[20:26:07] *** Eimann <Eimann!~eimann@courante.etherkiller.de> has joined #smartos[20:26:11] <Eimann> hey folks[20:31:55] *** mmlb <mmlb!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has joined #smartos[20:43:18] <npx> build_live: ERROR: failed to mount "/dev/lofi/2" on "/tmp/build_live-0.76641/a"[20:43:20] <npx> build_live: ERROR: failed to mount "/dev/lofi/2" on "/tmp/build_live-0.76641/a"[20:44:16] <npx> err sorry for double paste there, but any clue what I'm doing wrong here? the smartos build itself works fine in a base-multiarch-lts at 15 dot 4.0 instance[20:44:55] <npx> tmpfs isn't mounted.[20:47:03] <npx> (using mg to build installers that come with various additional images)[21:11:41] *** jperkin <jperkin!~jperkin@trumpton.perkin.org.uk> has joined #smartos[21:11:51] <jperkin> sigh, wondered why it was so quiet..[21:12:32] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has joined #smartos[21:12:33] *** ingenthr <ingenthr!~ingenthr@cpe-108-184-198-4.socal.res.rr.com> has quit IRC (Quit: Leaving.)[21:14:14] <LeftWing> npx: Did you follow the wiki page?[21:14:35] <LeftWing> i.e., https://wiki.smartos.org/display/DOC/Building+SmartOS+on+SmartOS[21:15:58] <npx> I'll try with the 18.3.0 multiarch image[21:18:02] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos[21:23:15] <jperkin> 18.4.0 images are up at https://pkgsrc.joyent.com/images/ if people want them early prior to release[21:26:19] <npx> (I can see how using rust would dramatically reduce the complexity of mg and the build in general; +5)[21:27:26] <LeftWing> Including rust in the OS, though I'm not opposed, will serve only to _increase_ the complexity of the build[21:27:55] <LeftWing> Mountain Gorilla is complicated because it's complicated, not because it's not written in Rust.[21:28:00] <npx> well this is assuming that you just got rid of everything written in node hahah, imagining an all-rust triton[21:28:20] <npx> I realize what a vast and immane project that would be.[21:28:43] <LeftWing> I guarantee rewriting Triton in Rust would create an entire new class of problems, which we won't know about in advance, and I'm sure it will (in that future land) be tempting to rewrite it in something else.[21:29:34] <LeftWing> To be clear: not saying we shouldn't use Rust, just saying that a straight translation won't fix any architectural issues -- but you could also fix architectural issues in Javascript code as well.[21:29:47] <LeftWing> Language choice is basically orthogonal.[21:30:58] <npx> I was specifically thinking about the sorta fractal complexity of builds that rely on specific platform images and versions of Node as compared to (hypothetically) "use the latest multiarch image, do a cargo build, drop the resultant artifact and manifest into a new image, distribute that shit"[21:31:38] <LeftWing> We would likely still require you to use particular versions of Rust & Cargo, for instance[21:32:22] <npx> yeah no free lunch ;( it's just inherently complicated stuff to a large extent.[21:33:14] <LeftWing> Ultimately a lot of that comes down to "are you building and shipping the things we've tested"[21:33:31] <LeftWing> The more flexibility there is in the build process, the less likely that will be true[21:52:14] *** polishdub <polishdub!~polishdub@207.86.38.254> has quit IRC (Remote host closed the connection)[21:52:24] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Read error: No route to host)[21:52:44] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[21:58:21] <npx> PING 1548104300 838271[22:09:30] <sjorge> adam_vollrath: there is some ansible support, also salt works well[22:09:43] <sjorge> Although I am baised in the salt stuff since I wrote most of it haha[22:22:04] <richlowe> LeftWing: code reviews![22:34:28] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)[22:38:58] <shaner> bahamat: I don't see an +Add button, just 'Share' and 'Tools'[22:45:59] <LeftWing> richlowe: I refuse![22:46:03] <LeftWing> Also, sorry :\[22:47:20] <LeftWing> richlowe: It looks like you have 10267 covered. Should I look at 9959 next, or?[22:47:44] *** waz0wski <waz0wski!waz0wski@hrothgar.distortion.io> has quit IRC (Quit: quit)[22:51:22] *** waz0wski <waz0wski!waz0wski@hrothgar.distortion.io> has joined #smartos[22:54:36] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)[22:56:28] *** strobelight <strobelight!~quassel@2001:420:c0c8:1003::18f> has quit IRC (Ping timeout: 264 seconds)[23:04:07] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has joined #smartos[23:07:22] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 250 seconds)[23:08:27] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has quit IRC (Ping timeout: 240 seconds)[23:28:10] *** ingenthr <ingenthr!~ingenthr@rrcs-98-152-51-170.west.biz.rr.com> has joined #smartos[23:35:36] <TyrfingMjolnir> How do I boot a bhyve zone? This line does not work: vmadm boot ZONE_UUID order=cd,once=d cdrom=/FreeBSD-12.0-RC3-amd64-bootonly.iso,ide[23:37:08] *** waz0wski <waz0wski!waz0wski@hrothgar.distortion.io> has quit IRC (Quit: quit)[23:37:44] <jbk> the arguments aren't supported w/ bhyve (yet at least).. I believe you can add a disk of type 'ahci-cd' and give it a path to add a cdrom[23:38:10] *** ingenthr <ingenthr!~ingenthr@rrcs-98-152-51-170.west.biz.rr.com> has quit IRC (Quit: Leaving.)[23:38:11] <jbk> the downside being that you have to go back and remove it from the config when you're done if just wanting to do a one-off thing[23:41:49] *** ingenthr <ingenthr!~ingenthr@rrcs-98-152-51-170.west.biz.rr.com> has joined #smartos[23:49:30] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has joined #smartos[23:49:49] *** ingenthr <ingenthr!~ingenthr@rrcs-98-152-51-170.west.biz.rr.com> has quit IRC (Quit: Leaving.)[23:58:16] *** ingenthr <ingenthr!~ingenthr@rrcs-98-152-51-170.west.biz.rr.com> has joined #smartos