January 2, 2017 
< | 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 2, 2017 [00:04:49] *** mpdnix <mpdnix!~mpdnix@c122-107-112-24.blktn7.nsw.optusnet.com.au> has joined #smartos[00:08:39] *** jelmd <jelmd!~jelmd@p548C0C01.dip0.t-ipconnect.de> has quit IRC (Quit: jelmd)[00:13:29] <richlowe> eripa: nginx service should depend on whatever mounts the delegated dataset in the zone. Which is probably filesystem/local[00:43:45] *** andy_js <andy_js!~andy@2.218.92.54> has quit IRC (Read error: Connection reset by peer)[00:43:54] *** andy_js <andy_js!~andy@2.218.92.54> has joined #smartos[00:54:01] *** mpdnix <mpdnix!~mpdnix@c122-107-112-24.blktn7.nsw.optusnet.com.au> has quit IRC (Ping timeout: 264 seconds)[00:55:27] *** andy_js <andy_js!~andy@2.218.92.54> has quit IRC (Quit: andy_js)[01:00:36] *** gratuxri_ <gratuxri_!~quassel@185.22.173.168> has quit IRC (Ping timeout: 248 seconds)[01:01:11] *** kimc <kimc!~kimc@pdpc/supporter/professional/kimc> has quit IRC (Quit: HydraIRC -> http://www.hydrairc.com <- Organize your IRC)[01:22:11] *** spijet <spijet!~serge@31.47.186.174> has quit IRC (Quit: WeeChat 1.6)[01:23:27] *** spijet <spijet!~serge@31.47.186.174> has joined #smartos[01:26:06] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Quit: ...)[01:29:47] *** leow <leow!~leow@175.137.69.238> has joined #smartos[01:34:32] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 258 seconds)[01:53:24] *** level7_ <level7_!~quassel@31.44.17.250> has quit IRC (Ping timeout: 248 seconds)[02:19:17] *** spijet <spijet!~serge@31.47.186.174> has quit IRC (Quit: WeeChat 1.6)[02:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)[02:20:06] *** ed209 <ed209!~ed209@165.225.128.67> has joined #smartos[02:22:38] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has joined #smartos[02:24:20] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has joined #smartos[02:27:01] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has quit IRC (Ping timeout: 264 seconds)[02:29:32] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[02:58:52] *** v_a_b <v_a_b!~volker@p54BC7D42.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 258 seconds)[03:03:28] *** v_a_b <v_a_b!~volker@p54BC7336.dip0.t-ipconnect.de> has joined #smartos[03:17:39] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has quit IRC (Ping timeout: 252 seconds)[03:18:01] *** leow <leow!~leow@175.137.69.238> has joined #smartos[03:22:25] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 252 seconds)[03:39:59] *** jcea <jcea!~Thunderbi@ks27448.kimsufi.com> has quit IRC (Quit: jcea)[04:07:16] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has joined #smartos[04:12:01] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has quit IRC (Ping timeout: 264 seconds)[04:12:13] *** jhendricks <jhendricks!~jhendrick@8.12.47.4> has joined #smartos[04:12:28] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Ping timeout: 258 seconds)[04:16:12] *** swills <swills!~swills@freebsd/developer/swills> has quit IRC (Quit: Leaving)[04:22:20] *** jelmd <jelmd!~jelmd@port-ip-88-150-13-208.reverse.mdcc-fun.de> has joined #smartos[04:32:45] *** Shaun <Shaun!~quassel@203.161.108.48.static.amnet.net.au> has quit IRC (Remote host closed the connection)[04:33:40] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has joined #smartos[04:34:25] *** Shaun <Shaun!~quassel@203.161.108.48.static.amnet.net.au> has joined #smartos[04:36:09] *** jhendricks <jhendricks!~jhendrick@8.12.47.4> has quit IRC (Quit: Leaving.)[04:45:56] *** jhendricks <jhendricks!~jhendrick@8.12.47.4> has joined #smartos[05:05:06] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[05:06:34] *** leow <leow!~leow@175.137.69.238> has joined #smartos[05:10:49] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 256 seconds)[05:26:44] *** Kurlon <Kurlon!~Kurlon@cpe-104-228-216-170.maine.res.rr.com> has quit IRC (Quit: Leaving...)[05:33:30] *** Kurlon <Kurlon!~Kurlon@cpe-104-228-216-170.maine.res.rr.com> has joined #smartos[05:40:02] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Ping timeout: 245 seconds)[05:41:03] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Remote host closed the connection)[05:41:23] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[05:58:18] *** marsell <marsell!~Adium@115-189-89-162.mobile.spark.co.nz> has quit IRC (Quit: Leaving.)[06:06:34] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Quit: Leaving.)[06:10:20] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Ping timeout: 252 seconds)[06:24:37] *** domag02 <domag02!~domag02@catv-176-63-228-36.catv.broadband.hu> has quit IRC (Ping timeout: 264 seconds)[06:36:36] *** bahamat_ <bahamat_!~bahamat@2001:470:f0f4:a0a:91e0:d29c:537a:7350> has quit IRC (Quit: Leaving.)[06:36:56] *** marsell <marsell!~Adium@115-189-89-162.mobile.spark.co.nz> has joined #smartos[06:40:50] *** jhendricks <jhendricks!~jhendrick@8.12.47.4> has quit IRC (Ping timeout: 258 seconds)[06:54:49] *** leow <leow!~leow@175.137.69.238> has joined #smartos[06:55:09] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has joined #smartos[06:57:01] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has joined #smartos[06:59:03] *** mhicks <mhicks!~Adium@ip72-192-145-123.sd.sd.cox.net> has joined #smartos[06:59:06] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 252 seconds)[06:59:12] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has quit IRC (Ping timeout: 245 seconds)[07:05:27] *** swills <swills!~swills@freebsd/developer/swills> has joined #smartos[07:07:44] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[07:12:20] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[07:23:08] *** mhicks <mhicks!~Adium@ip72-192-145-123.sd.sd.cox.net> has quit IRC (Quit: Leaving.)[08:08:23] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[08:12:36] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[08:13:49] <jzu_> hmm, I should look into how to enable Fabric for single computenode[08:21:27] *** melloc <melloc!~melloc@75-54-22-158.lightspeed.snbuca.sbcglobal.net> has quit IRC (Quit: Leaving.)[08:28:33] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has quit IRC (Ping timeout: 258 seconds)[08:38:17] <jzu_> any official way to delegate a ZFS dataset into LX/Docker Zone through Triton, guess no?[08:39:30] <jperkin> not through the api as far as I'm aware, needs to be done via adminui[08:40:07] <jperkin> I could be wrong but I think there are still some scalability issues which is why we prevent it being done by anyone[08:40:57] *** level7 <level7!~quassel@31.44.17.250> has joined #smartos[08:42:51] <jzu_> ah, right - I'll check the AdminUI[08:42:57] *** leow <leow!~leow@175.137.69.238> has joined #smartos[08:43:20] <jzu_> since well, it would be nice if I could create a instance (let it be through AdminUI to start with) that has delegated dataset, so I could do snapshots inside the LX/Docker[08:44:27] <jzu_> and yeah well.. the gotcha is, you cannot create docker containers through AdminUI :)[08:44:37] <jzu_> but container native Linux you of course can[08:48:01] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 264 seconds)[08:50:17] <jzu_> ah, and need to allow snapshots through CN as it's disabled by default[09:03:03] *** fxhp <fxhp!~fox@172.85.28.36> has quit IRC (Quit: WeeChat 0.4.2)[09:08:21] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[09:34:25] *** leow <leow!~leow@175.137.69.238> has joined #smartos[09:37:09] <wilbury> good morning[09:44:57] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[09:49:51] *** domag02 <domag02!~domag02@catv-176-63-228-36.catv.broadband.hu> has joined #smartos[09:55:53] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[10:05:33] *** mpdnix <mpdnix!~mpdnix@c122-107-112-24.blktn7.nsw.optusnet.com.au> has joined #smartos[10:11:25] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 264 seconds)[10:17:21] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[10:21:40] *** eripa <eripa!~eripa@212.116.78.217> has quit IRC (Ping timeout: 248 seconds)[10:21:48] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 260 seconds)[10:31:07] *** eripa <eripa!~eripa@212.116.78.217> has joined #smartos[10:35:40] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[10:45:44] *** ScottMed <ScottMed!~scottmede@li243-94.members.linode.com> has quit IRC (Ping timeout: 256 seconds)[10:46:13] *** ScottMed <ScottMed!~scottmede@li243-94.members.linode.com> has joined #smartos[10:53:03] *** eripa <eripa!~eripa@212.116.78.217> has quit IRC (Ping timeout: 258 seconds)[10:56:20] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[10:58:12] *** eripa <eripa!~eripa@212.116.78.217> has joined #smartos[11:01:18] *** fling <fling!~fling@fsf/member/fling> has quit IRC (Read error: Connection timed out)[11:02:23] *** fling <fling!~fling@fsf/member/fling> has joined #smartos[11:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)[11:20:06] *** ed209 <ed209!~ed209@165.225.128.67> has joined #smartos[11:20:34] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[11:26:11] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 268 seconds)[11:38:57] *** spijet <spijet!~serge@31.47.186.174> has joined #smartos[11:39:50] *** andy_js <andy_js!~andy@2.218.92.54> has joined #smartos[11:41:17] *** Eimann <Eimann!eimann@courante.etherkiller.de> has quit IRC (Ping timeout: 245 seconds)[11:44:49] *** leow <leow!~leow@175.137.69.238> has quit IRC (Ping timeout: 258 seconds)[11:44:55] *** Eimann <Eimann!eimann@courante.etherkiller.de> has joined #smartos[11:53:05] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[11:57:19] *** man_u <man_u!~manu@fob.gandi.net> has joined #smartos[11:57:40] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[12:02:20] *** man_u <man_u!~manu@fob.gandi.net> has quit IRC (Ping timeout: 252 seconds)[12:02:32] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[12:15:00] *** yomisei <yomisei!~void@ip4d161c9d.dynamic.kabel-deutschland.de> has quit IRC (Quit: How to defend yourself against an attacker armed with a mathematician)[12:16:17] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Ping timeout: 245 seconds)[12:18:12] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[12:26:24] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[12:33:23] *** level7 <level7!~quassel@31.44.17.250> has quit IRC (Ping timeout: 256 seconds)[12:36:04] *** level7 <level7!~quassel@31.44.16.132> has joined #smartos[12:57:15] *** danieli <danieli!~duniel@unaffiliated/danieli> has quit IRC (Remote host closed the connection)[12:57:38] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[12:57:41] *** danieli <danieli!~duniel@resiniferatoxin.duniel.no> has joined #smartos[12:58:17] *** level7 <level7!~quassel@31.44.16.132> has quit IRC (Remote host closed the connection)[12:58:31] *** dragunov11 <dragunov11!b63b444a@gateway/web/freenode/ip.182.59.68.74> has joined #smartos[13:00:53] *** danieli <danieli!~duniel@resiniferatoxin.duniel.no> has quit IRC (Changing host)[13:00:53] *** danieli <danieli!~duniel@unaffiliated/danieli> has joined #smartos[13:01:01] <dragunov11> Hi, i wanted to ask, why Global GZ usage keeps going up due to a process named zpool-zones/250[13:01:26] <jzu_> it's IO in the zpool named zones[13:01:38] <jzu_> check "zpool iostat zones 1" output[13:01:40] <dragunov11> Usage as in Processor Usage. and when I try to check the PID in ptree, it doesn't show up[13:01:57] <jzu_> well it's "software RAID", it uses CPU time :)[13:02:10] <dragunov11> Ok jzu_ , understood now. So nothing to work I think[13:02:27] <jzu_> back in the day the zpool wasn't shown as process, but that's many moons ago[13:03:26] <dragunov11> Cause I wasn't able to find it in ptree too, so was worried what this could be.[13:06:07] <jzu_> http://docs.oracle.com/cd/E19253-01/819-5461/gjhbp/index.html *g*[13:06:19] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[13:06:22] <jzu_> "Oracle Solaris 10 9/10 Release: In this Solaris release, each ZFS storage pool has an associated process, zpool-poolname"[13:06:26] <jzu_> so yeah... back in 2010 :)[13:08:42] <dragunov11> jzu_: ah, ok. Thanks for the info![13:10:20] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Client Quit)[13:10:29] <dragunov11> jzu_: I hear that Oracle is discontinuing Solaris development.[13:10:55] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Ping timeout: 252 seconds)[13:11:12] <dragunov11> I hope Joyent/Samsung continue.[13:15:03] <jzu_> dragunov11: well... that is just a rumour, and Joyent is developing SmartOS which is based on Illumos codebase which is based on OpenSolaris which is based on what became Solaris 11[13:15:23] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[13:16:00] *** dragunov11 <dragunov11!b63b444a@gateway/web/freenode/ip.182.59.68.74> has quit IRC (Ping timeout: 260 seconds)[13:20:31] *** leow <leow!~leow@175.137.71.175> has joined #smartos[13:23:52] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[13:24:44] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Ping timeout: 260 seconds)[13:28:02] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 246 seconds)[13:37:54] *** _mjg_ is now known as _mjg[13:46:35] *** tozhu <tozhu!~Thunderbi@li1067-74.members.linode.com> has joined #smartos[14:02:04] *** tozhu <tozhu!~Thunderbi@li1067-74.members.linode.com> has quit IRC (Ping timeout: 260 seconds)[14:05:48] *** leow <leow!~leow@175.137.71.175> has quit IRC (Remote host closed the connection)[14:17:50] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[14:24:42] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[14:25:26] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[14:27:27] *** leow <leow!~leow@175.137.71.175> has joined #smartos[14:28:42] *** leow_ <leow_!~leow@183.78.44.229> has joined #smartos[14:29:01] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 252 seconds)[14:32:24] *** leow <leow!~leow@175.137.71.175> has quit IRC (Ping timeout: 260 seconds)[14:38:32] <yruss972> Hey all, it might be off topic but does anyone have a comparison of native container support (not necessarily Docker) per operating system, ie. SmartOS/Linux/Windows/MacOS/etc? The thought crossed my mind that Windows container support could theoretically be more feature complete than Linux?[14:44:48] <jzu_> huh, what even is "Windows Native Container" :O[14:47:13] *** mpdnix <mpdnix!~mpdnix@c122-107-112-24.blktn7.nsw.optusnet.com.au> has quit IRC (Ping timeout: 268 seconds)[14:56:24] <modernpa1ifist> jzu_: I believe its just LXC for windows - but it doesn't support anything GUI-ish and even MS say its not a security mechanism[14:57:25] <jzu_> :)[14:59:08] <modernpa1ifist> Win Server containers are basically so you can run Docker on Windows and use the same process/resource isolation tricks. Hyper-V containers are hardware VMs[14:59:33] <modernpa1ifist> It is rather fun though that their base docker image is in th eorder of 1.5GB[14:59:47] <jzu_> :-D[15:04:05] <modernpa1ifist> I think the end goal is to combine that tech with their Linux Subsystem thing to allow win server 2016 to run any image from docker hub (a la LX zone)[15:05:47] <jzu_> yeah I see such as only sane way for windows to do Docker[15:06:14] <jzu_> not like VMware is doing with their Docker implementation .... they provision "lightweight" VM's![15:07:07] <modernpa1ifist> I figure their only option is to provision one hardware VM per container?[15:07:40] <jzu_> I think they use the VM's as Docker hosts, hosting multiple containers in single Photos OS VM[15:07:48] <jzu_> s/Photos/Photon/g[15:08:12] <modernpa1ifist> Oh right their stripped out docker host thing[15:08:22] <jzu_> Photon OS is their own Linux distro for Containers, which has it's own package manager with not so good package selection at the moment :)[15:08:33] <modernpa1ifist> Natrually[15:09:02] <jzu_> they didn't add anything new to the mix really... they should have virtualized the Docker host á la Triton but no[15:09:16] <modernpa1ifist> To quote MS on their container tech: "Windows Server Containers are great for scenarios where the OS trusts the applications that will be hosted on it, and all the applications also trust each other. In other words, the host OS and applications are within the same trust boundary. That’s true for many multi-container applications, applications that make up a shared service of a larger application,[15:09:22] <modernpa1ifist> and sometimes applications from the same organization."[15:09:39] <jzu_> hehe[15:16:13] <modernpa1ifist> But one thing they are doing is blurring hte lines between windows and hyper-v containers - so the user just specifies whether they want process/namespace isolation or kernel isolation (via virtual hardware) just with a flag[15:16:50] <modernpa1ifist> If the user wants a hardware isolation, it just provisions a hardware vm with a single windows container in it[15:26:37] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has joined #smartos[15:28:19] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[15:31:38] *** tozhu <tozhu!~Thunderbi@220.166.250.41> has quit IRC (Quit: tozhu)[15:32:26] *** Kurlon <Kurlon!~Kurlon@cpe-104-228-216-170.maine.res.rr.com> has quit IRC (Read error: Connection reset by peer)[15:32:38] *** Kurlon <Kurlon!~Kurlon@cpe-104-228-216-170.maine.res.rr.com> has joined #smartos[15:33:24] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Quit: Leaving.)[15:42:37] *** anona <anona!~anona@173-28-64-196.client.mchsi.com> has joined #smartos[15:43:53] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[15:44:12] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[15:56:26] *** pmooney <pmooney!~pmooney@50-93-248-179.fttp.usinternet.com> has joined #smartos[16:11:50] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[16:16:53] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[16:18:33] *** kimc <kimc!~kimc@99-9-187-149.lightspeed.livnmi.sbcglobal.net> has joined #smartos[16:18:33] *** kimc <kimc!~kimc@99-9-187-149.lightspeed.livnmi.sbcglobal.net> has quit IRC (Changing host)[16:18:33] *** kimc <kimc!~kimc@pdpc/supporter/professional/kimc> has joined #smartos[16:23:23] *** leow <leow!~leow@175.137.71.175> has joined #smartos[16:23:25] *** leow_ <leow_!~leow@183.78.44.229> has quit IRC (Read error: Connection reset by peer)[16:33:57] *** jakst <jakst!uid201469@gateway/web/irccloud.com/x-jjgbxkzdjxdtaffg> has joined #smartos[16:45:50] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[16:53:37] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Quit: Leaving.)[16:58:59] *** CaTtleyA <CaTtleyA!~CaTtleyA@16.146-14-84.ripe.coltfrance.com> has joined #smartos[17:07:01] <bennabiy> what is the use case for a delegated dataset vs non?[17:10:51] <wilbury> delegated dataset is for situations when you don't want, let's say, an application to eat up all the space within the zone.[17:12:20] <wiedi> with a delegated dataset you delegate administrative tasks to the zone root, so for example snapshots can be taken and rolled back from within the zone[17:13:49] *** leow <leow!~leow@175.137.71.175> has quit IRC (Remote host closed the connection)[17:17:49] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Remote host closed the connection)[17:18:26] *** yruss972 <yruss972!~yruss972@176.56.32.21> has joined #smartos[17:18:48] <bennabiy> wiedi: thank you. So if I do not need those functions to happen from within the zone itself, then do not bother with delegated_dataset. Does delegated dataset work for lx zones as well?[17:20:37] <wiedi> I think so, but you'll probabably need to use the tools from /native[17:23:14] *** yruss972 <yruss972!~yruss972@176.56.32.21> has quit IRC (Ping timeout: 268 seconds)[17:26:51] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[17:28:58] *** CaTtleyA <CaTtleyA!~CaTtleyA@16.146-14-84.ripe.coltfrance.com> has quit IRC (Quit: leaving)[17:30:03] <bennabiy> thank you[17:30:19] <bennabiy> for quota: is the number in GB or MiB?[17:30:41] <bennabiy> like quota 1024 = 1 GB or 1TB?[17:31:21] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 258 seconds)[17:31:54] *** headw01 <headw01!~billhead@209.63.38.162> has joined #smartos[17:40:29] *** gratuxri <gratuxri!~quassel@95.208.21.0> has joined #smartos[17:41:46] <headw01> How do I make the 'mtu' property of a link in a non-global zone permanent (the MTU of the physical NIC was changed after the zone was created)? I have tried "dladm set-linkprop -z ... -p mtu=9000 net0" which changes the MTU of the zone, but rebooting the zone sets the value back to 1500.[17:42:55] <headw01> According to https://wiki.smartos.org/display/DOC/Managing+NICs , the VNIC MTU should match the MTU of the physical nic when it comes up...[17:43:44] *** wizard113 <wizard113!~Adium@astound-66-234-218-247.ca.astound.net> has joined #smartos[17:45:24] <wiedi> headw01: mtu is managed per nictag with "nictagadm"[17:45:38] <wiedi> bennabiy: quota is in gb[17:50:11] <headw01> wiedi: "nictagadm" manages the MTU for the zone vnics? I thought that was for the 'admin' and 'external' tags (which modifies the /usbkey/config file).[17:50:14] *** bahamat <bahamat!~bahamat@2001:470:f0f4:a0a:346f:b3e7:d98f:9e6a> has joined #smartos[17:51:13] *** headw01 <headw01!~billhead@209.63.38.162> has quit IRC (Remote host closed the connection)[17:51:52] *** headw01 <headw01!~billhead@209.63.38.162> has joined #smartos[17:52:37] <wiedi> the vnics have a nic_tag assigned and use the mtu of that[17:53:23] <wiedi> oh, you can actually also set the mtu for each network device in the vmadm json with the "mtu" property[17:54:02] <headw01> Yeah, I know about the vmadm json method, but I wanted it to just default to the "right" value.[17:54:19] <headw01> How do I show the nic_tag of the vnic?[17:56:28] <bahamat> headw01: It's on the nic object.[17:56:41] <bennabiy> can dns_domain take multiple entries?[17:57:08] <bennabiy> and if yes, how would they be delimited?[17:58:31] <headw01> bahamat: OK, so that is 'external' for the 'net0' interface.[17:58:31] <headw01> root@artemis ~]# dladm show-link[17:58:31] <headw01> LINK CLASS MTU STATE BRIDGE OVER[17:58:31] <headw01> igb0 phys 9000 up -- --[17:58:31] <headw01> igb1 phys 9000 up -- --[17:58:32] <headw01> net0 vnic 1500 ? -- igb1[17:59:45] <headw01> According to dladm, the MTU of the physical is 9000. 'external_mtu=9000' in /usbkey/config.[18:00:35] <headw01> When I reboot the zone, the MTU is 1500 for the zone (not 9000 like https://wiki.smartos.org/display/DOC/Managing+NICs#ManagingNICs-ModifyingtheMTU seems to imply).[18:09:33] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has joined #smartos[18:27:45] *** rzezeski <rzezeski!uid151901@gateway/web/irccloud.com/x-yzuplmiwcbmymzzv> has joined #smartos[18:28:49] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has quit IRC (Quit: Leaving.)[18:30:05] *** OkiieDoe <OkiieDoe!~okiiedoe@ool-44c159a2.dyn.optonline.net> has joined #smartos[18:38:51] <bahamat> headw01: This is plain smartos, or triton?[18:39:57] <headw01> plain smartos[18:41:12] <bahamat> Did you use nictagadm to set the MTU, or just the config?[18:41:24] <headw01> bahamat: I punted and did 'vmadm update' with an update_vnics to explicitly set the mtu to 9000. That survives a reboot of the zone.[18:41:47] <headw01> For the external_mtu, I edited the config file and then rebooted the server.[18:43:34] <bahamat> As per that wiki page, the config file only affects the GZ. If you want to control the default value for zones you need to use nictagadm to set it first.[18:49:36] <headw01> bahamat: I am unclear why the DEFAULT for net0 is 9000, but VALUE is 1500 when the 'dladm show-linkprop -p mtu' command is run though.[18:52:07] *** arg <arg!sid180285@gateway/web/irccloud.com/x-jyclgcxylevkkmuj> has quit IRC (Ping timeout: 245 seconds)[18:52:10] <bahamat> Well, let's take a look at the source code and see what it's doing.[18:52:13] *** klynton <klynton!uid20683@gateway/web/irccloud.com/x-tsjdnepbsoucyujg> has quit IRC (Ping timeout: 258 seconds)[18:52:13] *** bbhoss <bbhoss!sid18216@gateway/web/irccloud.com/x-wypqkjfwqiauvzga> has quit IRC (Ping timeout: 258 seconds)[18:52:20] *** jesser <jesser!sid23214@gateway/web/irccloud.com/x-clgtgjuwfhbpripj> has quit IRC (Ping timeout: 260 seconds)[18:52:44] *** klynton <klynton!uid20683@gateway/web/irccloud.com/x-qwxeuwktmcubofku> has joined #smartos[18:52:58] *** deet <deet!sid142786@gateway/web/irccloud.com/x-qwwrjpuktgcvjjqq> has quit IRC (Ping timeout: 245 seconds)[18:53:22] *** yota <yota!sid110921@gateway/web/irccloud.com/x-vanpattbbhzliriw> has quit IRC (Ping timeout: 258 seconds)[18:53:22] *** adulteratedjedi <adulteratedjedi!sid95422@gateway/web/irccloud.com/x-bzkcbithxohhqnoi> has quit IRC (Ping timeout: 258 seconds)[18:53:45] *** pmonson <pmonson!uid55990@gateway/web/irccloud.com/x-dkyvwxqfuhdfzoww> has quit IRC (Ping timeout: 258 seconds)[18:54:13] *** deet <deet!sid142786@gateway/web/irccloud.com/x-bkmexpkmtnihntas> has joined #smartos[18:54:42] <bahamat> So, first of all, if mtu is present in the payload, it's used explicitly: https://github.com/joyent/smartos-live/blob/master/src/vm/node_modules/VM.js#L5364[18:54:53] *** arg <arg!sid180285@gateway/web/irccloud.com/x-bszkajbdhhcpswje> has joined #smartos[18:55:05] *** jesser <jesser!sid23214@gateway/web/irccloud.com/x-vqluzymzogcrxnam> has joined #smartos[18:56:26] *** yota <yota!sid110921@gateway/web/irccloud.com/x-gbxufhcdqugckykc> has joined #smartos[18:56:48] <bahamat> headw01: How many physical nics do you have?[18:58:02] *** adulteratedjedi <adulteratedjedi!sid95422@gateway/web/irccloud.com/x-lmzlvzhqbtjvfggp> has joined #smartos[18:59:37] <headw01> 2 physical nics[19:00:05] *** bbhoss <bbhoss!sid18216@gateway/web/irccloud.com/x-lebteezbzimarnmk> has joined #smartos[19:00:39] <bennabiy> bahamat: can entries like resolvers and dns_domain take multiple entries, and if so, what delimits them?[19:00:58] <bahamat> bennabiy: resolvers can.[19:01:04] <bahamat> resolvers is a list.[19:01:20] <bahamat> dns_domain is a string though.[19:01:20] <headw01> bennabiy: it is a comma separated list.[19:01:29] <headw01> (resolvers that is)[19:01:56] *** headw01 <headw01!~billhead@209.63.38.162> has quit IRC (Remote host closed the connection)[19:02:05] <bahamat> In the json payload it's an array.[19:02:31] *** headw01 <headw01!~billhead@209.63.38.162> has joined #smartos[19:03:01] <headw01> bahamat: back... My client crashed.[19:03:06] <bennabiy> dns_domain gets set at create time, and cannot be changed later, ... is it what is set for 127.0.0.1?[19:03:34] <bennabiy> and is it possible to configure a more populated /etc/hosts?[19:03:41] <bennabiy> at create time[19:04:04] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[19:04:11] <bennabiy> Also, is there something like DRBD for illumos?[19:04:26] <headw01> bennabiy: Are you trying to set these in the /usbkey/config file for the host?[19:04:43] <bennabiy> headw01: no[19:05:22] <headw01> bennabiy: So you are trying to set these values for a non-global zone then?[19:06:18] <bennabiy> yes[19:06:28] <bennabiy> I am configuring a zone creation script for VM[19:07:04] <bennabiy> just wondering how many of these I can have as generic and not have their networking all colliding or being too generic[19:07:53] <headw01> By zone creation script, do you mean a JSON file that has the properties in it? Something that is passed to 'vmadm create -f <file>'?[19:08:28] <bahamat> headw01: Second, vmadm checks the mtu in nictagadm: https://github.com/joyent/smartos-live/blob/master/src/vm/node_modules/VM.js#L1772[19:09:46] <bahamat> headw01: Using the default value of the datalink, looks like a property of the way dladm works.[19:10:41] <bennabiy> headw01: yes, something along those lines[19:10:54] <bennabiy> although a little more flexible[19:11:11] *** OkiieDoe <OkiieDoe!~okiiedoe@ool-44c159a2.dyn.optonline.net> has quit IRC (Read error: Connection reset by peer)[19:17:16] <headw01> bahamat: So it uses the DEFAULT value of the physical link then, not the current VALUE?[19:19:26] <bahamat> headw01: What's the output of `dladm show-linkprop | grep mtu`?[19:19:29] <headw01> bennabiy: The values of those properties can be changed after the fact (by a script in the global zone) using the 'vmadm update' command'[19:20:02] <headw01> bahamat: igb0 mtu rw 9000 1500 60-9000[19:20:02] <headw01> igb1 mtu rw 9000 1500 60-9000[19:20:02] <headw01> net0 mtu rw 9000 9000 60-9000[19:20:15] <bennabiy> headw01: dns_domain says in the vmadm man "Updating this after create has no effect."[19:20:21] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has joined #smartos[19:20:47] <headw01> bahamat: net0 has a current value of 9000 because I updated the nic object with vmadm update.[19:21:02] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has joined #smartos[19:21:47] <headw01> bennabiy: I guess dns_domain would have to be set correctly on create then.[19:22:13] <bennabiy> yes[19:22:19] <bennabiy> which is why I am asking about it[19:22:50] <bahamat> headw01: And what about `nictagadm list`?[19:23:27] <headw01> bennabiy: Actually, I just tried this 'vmadm update <uuid> dns_domain=foo.bar.com', and the value was changed in the <uuid> non-global zone.[19:24:04] <bennabiy> but was it actually change in /etc/hosts in the NGZ?[19:24:16] <bennabiy> changed*[19:24:22] <headw01> bahamat: nictagadm list[19:24:22] <headw01> NAME MACADDRESS LINK TYPE[19:24:22] <headw01> external MAC1 igb1 normal[19:24:22] <headw01> admin MAC2 igb0 normal[19:25:06] <bennabiy> bahamat: does illumos/smartos have support for drbd?[19:25:11] <bennabiy> or some such technology?[19:25:20] *** jhendricks <jhendricks!~jhendrick@ip68-97-70-46.ok.ok.cox.net> has quit IRC (Ping timeout: 268 seconds)[19:25:23] <bennabiy> I need to replicate a block device across wan[19:25:44] <bahamat> headw01: Well, I'm pretty sure that the behavior you're seeing is falling back to the behavior of dladm. Someone like rmustacc or pmooney can confirm that.[19:26:04] <bahamat> bennabiy: No, not quite.[19:26:30] <bennabiy> bahamat: so I would need to do a KVM then if I wanted to do that?[19:26:49] <bahamat> bennabiy: Probably, yes.[19:26:56] <bahamat> What do you want to do with drdb anyway?[19:26:56] <bennabiy> bahamat: ok[19:27:08] <bennabiy> replicate a mail server data store[19:27:13] <bennabiy> for HA failover[19:27:56] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[19:28:52] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[19:28:53] <headw01> bennabiy: the /etc/hosts file was not changed by changing the dns_domain. Do you mean /etc/resolv.conf (the search value)? That didn't change though either[19:30:15] <bennabiy> headw01: the vmadm(1M) says "For OS VMs this specifies the domain vale for /etc/hosts that gets set at create time. Updating this after create will have no effect."[19:30:25] <headw01> bahamat: Thanks for checking.[19:32:36] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 248 seconds)[19:33:02] <headw01> bennabiy: Hmm, well my /etc/hosts file in a brand new zone doesn't have anything like the dns_domain value that I set. I can only see that value in /etc/resolv.conf.[19:33:25] <headw01> Changing the property with vmadm update doesn't affect /etc/resolv.conf either.[19:33:35] <bahamat> dns_domain only sets the domain property of /etc/resolv.conf, and only at create time.[19:33:51] <bennabiy> then the man page needs to be update :)[19:35:12] <bsmithx10> bahamat: ever see the manta nameserver binder svc keep crashing?[19:35:25] <bahamat> bsmithx10: Lots.[19:35:40] <bahamat> bsmithx10: Can show me the stack trace?[19:35:48] *** pmonson <pmonson!uid55990@gateway/web/irccloud.com/x-vpscwpnnhfruxfic> has joined #smartos[19:36:16] <bsmithx10> let me, get it to crash[19:36:47] <bsmithx10> https://paste.ec/paste/xXxaInNw#lX4riHujKR1fLFLIQCucCdqno5g8L-hG0EAPhi/XyE0[19:37:03] <bsmithx10> What command do you want me to run to grab it?[19:37:23] <bsmithx10> guess whatever smf is running?[19:38:28] <bahamat> bsmithx10: Looks like MANTA-3058 I think.[19:38:28] <jinni> https://smartos.org/bugview/MANTA-3058[19:39:43] *** jinni <jinni!~jinni@72.2.113.94> has quit IRC (Remote host closed the connection)[19:40:20] <bsmithx10> bahamat: I have an lbservice running[19:40:34] <bsmithx10> but i assume its .... not finding anyone in order to register?[19:40:40] <bahamat> Can you send me the core then?[19:40:44] <bsmithx10> service discovery[19:40:53] <bsmithx10> this never ran, just forcontext[19:41:04] <bsmithx10> this has no data... my attemp to get it running[19:41:09] <bsmithx10> is there a race?[19:41:18] <bsmithx10> sure, let me host it real quick[19:47:27] <bsmithx10> bahamat: http://tritonhost.com:9001/core.node.42376[19:50:36] <bsmithx10> let me know if that didnt work[19:52:13] <bahamat> bsmithx10: Got it.[19:53:23] *** spijet <spijet!~serge@31.47.186.174> has quit IRC (Quit: WeeChat 1.6)[19:54:08] <bsmithx10> Sweet :)[19:54:20] <bsmithx10> let me know if you want me to run anything on any of those services[19:54:39] *** motu_nb <motu_nb!~motu@p2003006D6C41330085B4FA3C4E426A5C.dip0.t-ipconnect.de> has quit IRC (Quit: Verlassend)[19:56:11] <bahamat> I see what's going on.[19:56:54] *** poige <poige!~poige@2a03:b0c0:3:d0::2e9:5002> has joined #smartos[19:57:09] <poige> Hi[19:58:01] <poige> Trying to get network interface in zone-KVM working as a bridged to global one, still no luck[19:58:36] <bahamat> poige: What exactly are you trying to do?[19:58:47] <poige> I came across allowed-address but it didn't resolve[19:58:52] <poige> well, very simple thing[19:59:36] <poige> I have KVM running and I'm trying to add IPs inside that Linux "box" and get them "connected" from outside[19:59:53] <poige> w/o routing, just on L2 level[20:00:35] <bahamat> poige: I think you're trying to overthink things.[20:00:57] <poige> May be I'm too smart, yep[20:01:05] <poige> But I need it, alas )[20:02:16] <poige> I didn't dig that deep into zones networking so my unanswered yet question is vmadm using "stub" or "overlay"?[20:02:23] <bahamat> Did you follow the instructions for setting up a kvm instance?[20:02:32] <bahamat> poige: Neither.[20:02:38] <bennabiy> How do I go about setting up a KVM with multiple "disks" ?[20:03:06] <bahamat> bennabiy: The disks property is an array. So just have multiple items in the array.[20:03:27] *** mhicks <mhicks!~Adium@ip72-192-145-123.sd.sd.cox.net> has joined #smartos[20:03:39] <poige> bahamat, surely I did follow it[20:03:44] <bennabiy> bahamat: ok...[20:04:31] <bahamat> poige: setting up networking is straightforward. You just need to set the appropriate nic_tag and interface information.[20:04:56] <poige> bahamat, I have working network[20:05:25] <bahamat> poige: Then I still don't understand what you're trying to do.[20:05:26] <poige> bahamat, I just don't have L2 transparency, so if I add another IP in KVM itself it's not reachable from outside[20:05:38] <bsmithx10> ahhh[20:05:57] <bahamat> poige: The allowed_ips property *is* what you want.[20:06:02] <bsmithx10> poige: use dhcp on the interface, and make sure the nic_tag and vlan_id are setproperly[20:06:10] <poige> bahamat, I tried allowed-address[20:06:11] <bsmithx10> and also the mtu, and then your switch :)[20:06:14] <poige> is it different?[20:06:26] <poige> ah[20:06:39] <poige> BTW, I just did change it with zonecfg but didn't reboot zone after the change[20:06:44] <bennabiy> If I make a custom script to be run in the GZ, where do I put it?[20:06:58] <poige> is it not applied immediately, is it? I suppose it's not )[20:07:10] <bahamat> bennabiy: Somewhere under /opt (generally /opt/custom/bin)[20:07:35] <bennabiy> bahamat: and then add /opt/custom/bin to the path?[20:08:44] <bahamat> bennabiy: Yep[20:09:09] <bahamat> poige: It is. You can verify it with dladm show-linkprop[20:09:16] *** yomisei <yomisei!~void@ip4d161e5b.dynamic.kabel-deutschland.de> has joined #smartos[20:09:29] <bennabiy> how do I get the path to be kept between boots?[20:10:13] *** jhendricks1 <jhendricks1!~jhendrick@8.12.47.4> has quit IRC (Ping timeout: 264 seconds)[20:10:22] <tgerczei> I have my rcfiles on a permanent dataset and a custom postboot service links them symbolically under /root[20:10:31] <bahamat> bennabiy: You need a postboot smf service or something like that.[20:11:02] <tgerczei> that means a service manifest under /opt/custom/smf/ and a corresponding script invoked by it under /opt/custom/bin/[20:11:02] <bennabiy> I will have to get that figured out...[20:11:07] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has joined #smartos[20:11:12] <poige> bahamat, thanks, will check now[20:11:51] <bennabiy> tgerczei: would it be possible to have a general postboot smf, which calls a script which can be added to without needing to have individual smf manifests running around?[20:11:52] <poige> Just to make it clear: should I get pings outside working on pure Ethernet, w/o setting up routes to that SmartOS box first and KVM instance in it, second?[20:11:58] <tgerczei> bennabiy: take a look at smartos-guesthandler, that's using the same mechanism to inject an SMF service during boot[20:12:03] <tgerczei> bennabiy: that's exactly what happens[20:12:14] <tgerczei> you need one service and one script, the latter of which you can edit[20:12:26] <bennabiy> tgerczei: thank you[20:12:36] <tgerczei> the manifest merely declares an exec resulting in an invocation of the script[20:12:41] <tgerczei> you're welcome[20:13:01] <bennabiy> tgerczei: I did this once before on my first testbed server, but I could not remember how I did it[20:13:09] <tgerczei> LeftWing had a gist online with his postboot service for the patched KVM module he maintains[20:13:11] <poige> should I remove "mac-nospoof"?[20:13:15] <tgerczei> let me find it[20:13:39] <tgerczei> https://gist.github.com/jclulow/2cb28f48a01b56c01153[20:13:41] <tgerczei> there.[20:13:43] <bennabiy> that would be helpful[20:13:47] *** fxhp <fxhp!~fox@172.85.28.36> has joined #smartos[20:13:59] <tgerczei> I'm using this very same script albeit with alterations of my own applied[20:14:07] <tgerczei> and the same service manifest accompanying it[20:14:20] <tgerczei> you can follow this gist step-by-step, it works as it is[20:14:25] <bahamat> poige: Can you define "outside"?[20:14:37] <tgerczei> but don't download the kvm module and don't try to load it with the script :)[20:14:46] <tgerczei> as that's not what you're after, most likely[20:14:57] <poige> bahamat, on the same Ethernet domain[20:15:25] <poige> bahamat, outside -- not running inside same SmartOS box[20:15:26] <bahamat> poige: Can you show me the vm json?[20:15:47] <poige> I didn't try allowed-ips yet[20:15:57] <bahamat> poige: Just show me what you have right now.[20:16:04] <poige> k[20:16:35] <bennabiy> tgerczei: thank you so much[20:16:56] <bennabiy> Time to get a system fired up to test out drbd[20:17:22] <tgerczei> bennabiy: np, g/l[20:19:20] *** jayschmidt <jayschmidt!~jayschmid@2605:a000:122b:183:5c97:634e:1e97:4cb6> has joined #smartos[20:19:26] <poige> bahamat, it's straightforward, I'd say[20:21:20] <poige> I guess "allow_ip_spoofing" needed[20:22:48] <bennabiy> So if I wanted multiple disks in a KVM, I would do something like this? https://pastebin.mozilla.org/8958006[20:25:18] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[20:26:03] <bennabiy> with other options if I wanted to set them, I know[20:26:09] <bennabiy> is that the general idea?[20:26:55] <bahamat> bennabiy: Yep, exactly that.[20:27:16] <bennabiy> bahamat: and that could be referenced from within the VM as a block device?[20:27:22] <bahamat> Any disk that has an image_uuid property will be cloned from that zvol. If not, it will be an empty zvol.[20:27:43] <bahamat> bennabiy: Yeah, those will show up as block devices in the guest.[20:28:14] <poige> bahamat, so you seen the link?[20:28:54] <bennabiy> is it best practice to use compression on kvm disks?[20:29:08] <bahamat> poige: No, I don't see a link from you.[20:29:40] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Ping timeout: 248 seconds)[20:32:34] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has quit IRC (Quit: Leaving.)[20:35:12] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has joined #smartos[20:35:38] *** headw01 <headw01!~billhead@209.63.38.162> has quit IRC (Ping timeout: 268 seconds)[20:36:07] <bennabiy> can you add disks on the fly to a kvm?[20:36:23] <bennabiy> or would there need to be a reboot for them to be seen?[20:36:29] <bennabiy> and how about external storage?[20:36:37] <bennabiy> USB passthrough type[20:36:54] *** motu_nb <motu_nb!~motu@p2003006D6C41330085B4FA3C4E426A5C.dip0.t-ipconnect.de> has joined #smartos[20:37:56] <poige> for the record -- right guess, "allow_ip_spoofing" did the job[20:38:12] <poige> Many thanks, bahamat[20:38:48] <bahamat> poige: Be warned, that with allow_ip_spoofing, the zone can do nefarious things on your network.[20:39:06] <bahamat> Make sure you're only setting that for zones where you trust the operator not to assault the network.[20:40:05] <poige> Sure[20:47:05] *** jinni <jinni!~jinni@72.2.113.94> has joined #smartos[20:51:20] *** deirdres <deirdres!~tbib@99-90-69-118.lightspeed.sntcca.sbcglobal.net> has joined #smartos[20:52:30] *** yruss972 <yruss972!~yruss972@159.148.40.97> has joined #smartos[20:53:27] <bsmithx10> bahamat: anything blatent in that coredump?[20:53:54] *** deirdres <deirdres!~tbib@99-90-69-118.lightspeed.sntcca.sbcglobal.net> has quit IRC (Client Quit)[20:54:09] <bahamat> bsmithx10: I haven't dug too deeply into it yet, but it is the same issue as MANTA-3058[20:54:10] <jinni> https://smartos.org/bugview/MANTA-3058[20:54:11] <bennabiy> Is there any way to set the disk order when creating a VM?[20:54:23] <poige> and again, for the records, you don't need allow_ip_spoofing if you can use allowed_ips[20:54:29] <bennabiy> any way to control which one shows up as vda or vdb?[20:54:38] <bahamat> bennabiy: I think it's the order of the array.[20:54:39] <poige> it would give you more precise control reducing probable risks[20:54:41] <bahamat> poige: Correct.[20:54:55] <bennabiy> bahamat: ok, will test[20:55:03] <poige> yep. I'm just sharing back what I've learned )[20:55:18] *** ChrisJames02170 <ChrisJames02170!~py@ool-1826eaa1.dyn.optonline.net> has quit IRC (Remote host closed the connection)[20:55:51] <poige> If this channel gets on WEB in a way, there're chances someone would save some time for him and the channel[20:56:12] <tgerczei> poige: echelog has it already[20:56:21] <poige> gr8[20:56:56] <poige> Solaris rocks. ;) SmartOS too )[20:58:27] <poige> I'm really missing some user-space tools in global zone[20:58:41] <bsmithx10> bahamat: anyway to get around that? Can I force the lb to register in zk?[20:58:42] <poige> there's not top, there's no pkgin even to install top[20:59:11] <bahamat> bsmithx10: I don't know much about zk, or binder's use of it. Best to ask arekinath when he's online.[21:01:14] <bahamat> poige: pkg_info -e top[21:01:14] <bahamat> top-3.6.1nb1[21:01:48] <poige> no pkg_info in /usr/bin /usr/sbin /smartdc/bin /opt/smartdc/bin /opt/local/bin /opt/local/sbin /opt/smartdc/agents/bin[21:01:58] <bennabiy> can I pass a UUID to the VM through json which is not a long random collection of bits?[21:02:12] <bennabiy> can I pass, say UUID=bleh0h1[21:02:24] <nbjoerg> poige: there is a special GZ repo somewhere[21:02:25] <bennabiy> as long as it is unique on my systems?[21:02:52] <poige> also, 13M available in / doesn't make me way too happy, not that bold provisioning[21:02:56] <bahamat> bennabiy: The uuid is really the name of a dataset. So it'll look for something that matches.[21:03:15] <nbjoerg> https://pkgsrc.joyent.com/install-on-illumos/ --> GZ rider[21:03:20] <bennabiy> bahamat: if I am creating a VM with an empty dataset...[21:03:23] <poige> nbjoerg, I tried following some instructions for getting dev's tools set but it failed[21:03:27] <bahamat> poige: SmartOS is a read-only live system. You should not be modifying /[21:03:59] <poige> it filled up the root partition first, then when I re-tried with space from /zones it didn't match sha sum[21:04:11] <poige> nbjoerg, so I abandoned the whole idea[21:04:38] <poige> bahamat, surely. But where from are you supposed to have overall look of the system?[21:04:43] <bahamat> poige: cd to /var/tmp before downloading the tar.[21:05:01] <bahamat> poige: prstat[21:05:10] <poige> even mpstat, yeah[21:05:31] <bennabiy> bahamat: according to the man, it seems that the UUID is used to create the dataset, not the other way around, so if I gave it a UUID, no matter how simple, as long as it didn't match anything else on the system it would work, correct?[21:06:02] <bahamat> bennabiy: No.[21:06:22] <bahamat> The name of the dataset for the disk will be /zones/:uuid-diskX[21:06:28] <bahamat> Where X is the array index number.[21:06:33] <bennabiy> yes[21:06:43] <bennabiy> I am saying the UUID of the system, not the disk[21:06:45] <bahamat> If you pass image_uuid then it will clone from the specified image.[21:07:02] <bennabiy> not passing image_uuid[21:07:03] <bahamat> bennabiy: You mean the uuid of the zone?[21:07:06] <bennabiy> yes[21:07:42] <bahamat> You don't need to include the zone uuid in the disks array. It will know by itself.[21:07:48] <bennabiy> what the GZ will see as the UUID of the ngz[21:08:02] <bennabiy> I am not talking about the disks array[21:08:15] <bahamat> You can explicitly set the zone uuid before creation, but I don't recommend doing so.[21:08:18] <bennabiy> I can see where your confusion comes from... I am talking about the general json for creating a VM[21:08:28] <bennabiy> mind if I ask why[21:08:28] <bennabiy> ?[21:09:21] <bahamat> Chiefly because everything we do will assume a full uuid. Using anything shorter may break at any time.[21:13:24] <bennabiy> bahamat: but wouldn't it have to grab a UUID from somewhere? What qualifies as a UUID?[21:14:24] <bahamat> bennabiy: The platform image comes with a number of libraries to generate uuids[21:14:41] <bennabiy> I surrender...[21:15:51] <bahamat> bennabiy: What I'm really saying, is that joyent might (or might have already) created code that makes assumptions on RFC4122 compliant UUID identifiers for zone (or other object) names.[21:16:20] <bennabiy> bahamat: that is what I was wondering.. what specified a "valid" UUID[21:16:24] <bahamat> If you create zones (or other objects) that don't use RFC4122 compliant uuids and it breaks, then you get to keep both pieces.[21:17:14] <bahamat> According to the RFC, hex text with hyphen separators.[21:17:16] <bennabiy> I guess I just wondered what the point of specifying the UUID if it was better to just generate it with VM creation[21:17:39] <bennabiy> is it to support scripting and documenting?[21:18:07] <bahamat> bennabiy: Upstack triton components may generate the UUID elsewhere, so vmadm respects the uuid if present.[21:18:14] <bahamat> If not, it generates one.[21:18:24] *** melloc <melloc!~melloc@2602:304:b361:69e0:6072:6cc5:885d:e543> has joined #smartos[21:18:26] <bennabiy> bahamat: thank you. I understand now[21:19:03] *** jakst <jakst!uid201469@gateway/web/irccloud.com/x-jjgbxkzdjxdtaffg> has quit IRC (Quit: Connection closed for inactivity)[21:19:08] <bahamat> bennabiy: But technically, there's no reason you can't have a zone named "foo". zonecfg will let you create one, afterall.[21:19:29] <bahamat> But there's no expectation that any of the SmartOS tooling will work with it.[21:19:48] <bahamat> It might, and probably will. But that's no guarantee of future compatibility.[21:20:00] *** ed209 <ed209!~ed209@165.225.128.67> has quit IRC (Remote host closed the connection)[21:20:01] <bennabiy> bahamat: so you get what you get and don't expect more if you make your own[21:20:06] *** ed209 <ed209!~ed209@165.225.128.67> has joined #smartos[21:20:32] <jzu_> bahamat: who I should talk to at Joyent regarding Triton networking?[21:20:58] <bahamat> bennabiy: The interface that SmartOS exposes is zones named with UUIDs. Zonecfg is considered an implementation detail.[21:21:17] <bahamat> jzu_: Depends on what area. melloc is working on NAPI.[21:21:30] <melloc> jzu_: What's your question?[21:21:47] <bennabiy> bahamat: I think I can work with that. Thank you[21:22:13] <bahamat> bennabiy: There's nothing preventing you from doing anything you like, I'm just saying to be aware of where the exposed interface boundaries are.[21:22:27] <bennabiy> understood :)[21:22:44] <poige> ergh... it looks like Solaris has kexec[21:22:48] <bennabiy> I am learning smartos, so I am trying to understand :)[21:22:50] <poige> wow[21:22:58] <bennabiy> I printed out the man page for vmadm[21:23:10] <bennabiy> mostly for the properties[21:23:20] <poige> does it really mean I can have smartos rebooting there in tens of seconds instead of minutes?[21:24:00] *** mhicks <mhicks!~Adium@ip72-192-145-123.sd.sd.cox.net> has quit IRC (Quit: Leaving.)[21:24:09] <bennabiy> bahamat: do I need to specify boot on each disk I attach to kvm?[21:24:33] <jzu_> melloc: I would have Internet facing containers which would need to connect to containers on the Inside/Private SDC network[21:24:37] <bennabiy> like if I had 10 disks on it, would I have to have 1 boot: true and the rest specify boot: false?[21:25:17] <jzu_> melloc: should I just have all containers use Fabric and use the NAT instance for routing HTTP/HTTPS to WWW container, or should I assign it to External also along with Fabric?[21:25:37] <jzu_> melloc: I'm kind of thinking "Should the External just be some private range or Internet range"[21:26:03] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[21:26:05] <melloc> jzu_: You should assign it an external NIC along with the fabric NIC.[21:26:17] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has quit IRC (Client Quit)[21:26:35] <jzu_> melloc: Ok, so best decision would be to have the Externel be _real_ External, aka. have Public Internet IP-addresses[21:26:59] <jzu_> melloc: and then just assign the nginx/apache/whatever with interfaces on both External and Fabric network's[21:27:29] <melloc> Yep. You'll want to make sure that the primary NIC is the external NIC, too.[21:27:51] <jzu_> melloc: hehe, and that is because cloudfw is only applied to the primary? :)[21:28:04] <jzu_> melloc: (I've been hit by that with plain SmartOS)[21:28:17] <melloc> Nope, the firewall should be applied to all NICs.[21:28:47] <jzu_> hmm, then friend of mine must be doing something wrongly on plain-SmartOS[21:29:12] <melloc> Were you only listening on the primiary's IP in that case?[21:29:15] *** LiftedKilt <LiftedKilt!kilty@unaffiliated/liftedkilt> has quit IRC (Quit: Kilted Southern)[21:29:30] <bahamat> bennabiy: boot defaults to false.[21:29:38] <melloc> If you bind to a specific address, and not 0.0.0.0 or ::, then you can only access the service over that specific address.[21:29:40] <bennabiy> bahamat: that is what I thought[21:29:41] <jzu_> melloc: will need to verify the details and get back to that[21:29:57] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[21:30:11] <bennabiy> bahamat: if I am going to be firing up a KVM and need to do an install, what is the recommended method (before there is SSH access to it)[21:30:15] <bennabiy> zlogin?[21:30:39] <bahamat> For kvm, use the exposed vnc interface.[21:30:56] <bennabiy> any way to find out what it is before I start the VM?[21:31:19] <bahamat> bennabiy: zlogin connects to a special serial device, and needs cooperation from the guest. So if there's no OS on the guest then you might have some trouble getting it working.[21:31:30] <melloc> jzu_: The reason I recommend making the external NIC the primary is so that the VM will use the external network's gateway, and not pass through the NAT.[21:31:38] <bahamat> bennabiy: No, you have to start it then `vmadm info <uuid>`[21:31:40] <bennabiy> if I set autoboot false, then can I find out the vnc info ?[21:31:43] <bennabiy> ok[21:31:55] <bahamat> bennabiy: info only works on running zones.[21:32:06] <bennabiy> that is right, I remember now[21:32:35] <bennabiy> 6[21:32:38] <jzu_> melloc: ah, yeah thought so :)[21:32:54] <jzu_> melloc: Docker coantainers on Triton also support having multiple NIC's, right?[21:33:41] <melloc> jzu_: Yes, although you'll have to add the extra NICs via CloudAPI after creating it, currently.[21:33:52] <jzu_> melloc: as I'd be running the Internet-facing services as Docker, instead of native Zones[21:34:20] *** LiftedKilt <LiftedKilt!kilty@unaffiliated/liftedkilt> has joined #smartos[21:34:20] <jzu_> melloc: ah, do I need to craft customer JSON payload for that or can it be done through Triton CLI?[21:34:28] <melloc> Docker doesn't support specifying multiple networks at creation currently, and we haven't implemented `network attach'.[21:34:32] <jzu_> melloc: $ docker only supports adding the primary interface?[21:34:33] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 256 seconds)[21:34:38] <jzu_> ok :)[21:34:59] <rsully> Is there any info about Triton + Kubernetes? I've seen it mentioned a couple times a while ago, plus a mention on Joyent's website - but no actual examples or documentation[21:35:07] <melloc> If you provision with the -P flag, you'll get both your fabric and external networks.[21:35:33] <melloc> If you want more than that, you can add with the Triton CLI.[21:35:38] *** ChrisJames02170 <ChrisJames02170!~py@ool-1826eaa1.dyn.optonline.net> has joined #smartos[21:35:56] <jzu_> melloc: ah well that's all I need for most cases[21:36:05] <jzu_> melloc: so I can do it all through Docker then[21:36:55] *** poige <poige!~poige@2a03:b0c0:3:d0::2e9:5002> has quit IRC (Quit: Ex-Chat)[21:41:08] <bennabiy> does the VNC info change every time you start a VM?[21:41:25] <jzu_> yes, the port changes[21:42:51] <bennabiy> what if you just reboot a VM without stopping it?[21:43:32] <MerlinDMC> the port changes if you're not specifying a fixed one - a reboot inside the qemu VM will not restart qemu -> the port will stay[21:43:52] <bennabiy> how do you specify a fixed one?[21:44:35] <MerlinDMC> https://smartos.org/man/1m/vmadm search for "vnc_port"[21:45:19] <bennabiy> MerlinDMC: just foudn it[21:45:26] <bennabiy> I had not made it that far in the man yet[21:46:28] <jzu_> melloc: by the way, how does Joyent handle getting JPC customers VNC access?[21:47:07] <jzu_> melloc: haven't tried creating KVM on JPC so no clue, I was just wondering that because VNC is only on Admin network is it even available for cloud customers?[21:47:54] *** poige <poige!~poige@2a03:b0c0:3:d0::2e9:5002> has joined #smartos[21:48:51] <jzu_> hmm, will need to look into configuring gateway to headnodes usb config for Admin network as I'd need to reach it through a gateway maybe in the future[21:50:48] <melloc> jzu_: VNC access isn't available to JPC customers at the moment.[21:51:31] <jzu_> melloc: ok so that explains how it's handled =)[21:52:10] <jzu_> melloc: wonder how people haven't been nagging about lack of it so much that it would have been done :)[21:52:43] <jzu_> melloc: I wouldn't run prod VM without having recovery access to the console to be honest, but that's just me[21:53:12] <jzu_> melloc: maybe people tend to do HA and have no need to recover, infrastructure as a code :)[22:05:17] *** headw01 <headw01!~billhead@209.63.38.162> has joined #smartos[22:08:27] <jzu_> melloc: but yeah, thanks for the network tips regarding Docker[22:09:29] <jzu_> melloc: by the way, how about multiple External networks, ie. I run out of IP assignment, how that kind situation should be handled?[22:09:43] <jzu_> melloc: do I just add a new nictag 'External2' with new IP range?[22:10:31] <melloc> jzu_: You can create a network pool and place multiple networks on the 'external' NIC tag in it.[22:10:40] <melloc> jzu_: And no problem, glad to help![22:11:02] <jzu_> melloc: ah right os I'll just keep adding new IP ranges to the existing External[22:14:10] <jzu_> melloc: the same with NAT pools? Haven't looked into those yet as I haven't configured Fabrics in my PoC just yet, but I bet I would run out of NAT IP's eventually along with External running out[22:14:59] <jzu_> melloc: the NAT pool will most likely consist of the same IP-range as External, I just kind of split it half and half or such[22:17:52] <jzu_> right the NAT pool is a pool 'Is a pool, so additional networks can be added later.' :)[22:19:22] <melloc> jzu_: Yep, exactly. at the moment, all networks in a pool have to be on the same nic tag, but I've been working on RFD 43, which will change that.[22:19:22] <jinni> https://github.com/joyent/rfd/tree/master/rfd/0043[22:21:03] <jzu_> melloc: great - good to know the limitations at the moment[22:21:10] <jzu_> melloc: I can plan accordingly[22:43:30] *** poige <poige!~poige@2a03:b0c0:3:d0::2e9:5002> has quit IRC (Remote host closed the connection)[22:46:46] *** marsell1 <marsell1!~Adium@122-56-197-210.mobile.spark.co.nz> has joined #smartos[22:47:02] *** marsell <marsell!~Adium@115-189-89-162.mobile.spark.co.nz> has quit IRC (Ping timeout: 265 seconds)[22:47:29] *** headw01 <headw01!~billhead@209.63.38.162> has quit IRC (Quit: Leaving.)[22:51:44] *** forerveranewb <forerveranewb!~Adium@c-67-170-10-122.hsd1.wa.comcast.net> has quit IRC (Quit: Leaving.)[22:53:56] *** yruss972 <yruss972!~yruss972@159.148.40.97> has quit IRC (Remote host closed the connection)[22:54:25] *** yruss972 <yruss972!~yruss972@159.148.40.97> has joined #smartos[22:58:59] *** yruss972 <yruss972!~yruss972@159.148.40.97> has quit IRC (Ping timeout: 256 seconds)[23:13:32] *** yruss972 <yruss972!~yruss972@159.148.40.97> has joined #smartos[23:23:16] *** poige <poige!~poige@2a03:b0c0:3:d0::2e9:5002> has joined #smartos[23:24:33] <poige> BTW, I met interesting problem recently: if you're using Ethernet link aggregation with LACP then you can't have PXE[23:25:32] <poige> If you switch LACP off during boot, then you won't have connectivity when OS booted[23:29:41] <jayschmidt> You need to enable LACP fallback or whatever your switch gear calls it.[23:29:49] <jayschmidt> That's how we boot aggregated servers in the cloud.[23:30:46] <poige> jayschmidt, thanks, didn't hear about that[23:31:09] *** platonic <platonic!~platonic@51.175.98.163> has joined #smartos[23:34:03] *** yruss972 <yruss972!~yruss972@159.148.40.97> has quit IRC (Remote host closed the connection)[23:35:43] *** White_Light <White_Light!~Spacemen_@cpe-108-167-11-28.neb.res.rr.com> has quit IRC (Read error: Connection reset by peer)[23:35:53] *** platonic <platonic!~platonic@51.175.98.163> has quit IRC (Ping timeout: 258 seconds)[23:36:16] *** richardkiene <richardkiene!~kiener@c-73-239-226-222.hsd1.wa.comcast.net> has quit IRC (Quit: leaving)[23:37:21] *** White_Light <White_Light!~Spacemen_@cpe-108-167-11-28.neb.res.rr.com> has joined #smartos[23:49:39] <TyrfingMjolnir> What does this mean? https://bpaste.net/raw/d9fdbb2f4df5[23:51:27] *** ingenthr <ingenthr!~ingenthr@cpe-107-184-13-203.socal.res.rr.com> has joined #smartos[23:51:34] <poige> TyrfingMjolnir, make -- do you have it installed?[23:52:43] <TyrfingMjolnir> make as in? bmake?[23:52:47] <TyrfingMjolnir> build essential?[23:52:54] <TyrfingMjolnir> Or which pkgin in ?!?[23:55:53] *** yruss972 <yruss972!~yruss972@159.148.40.97> has joined #smartos[23:58:08] <poige> TyrfingMjolnir, just type in "make" and see what it gives