January 15, 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 15, 2019 [00:00:02] <Latrina> what a bummer.. Thanks guys[00:06:09] <jperkin> native tun should work fine[00:07:12] *** strobelight <strobelight!quassel@nat/cisco/x-whwhfwrganjlzwom> has quit IRC (Ping timeout: 252 seconds)[00:07:46] <Smithx10> Bahamat do you have a snippet for wiring up the Prometheus metrics for Manatee and Moray applications[00:07:51] <Smithx10> i know they store that in sapi right?[00:09:34] *** greenline <greenline!~user@unaffiliated/greenline> has quit IRC (Remote host closed the connection)[00:10:14] <Latrina> jperkin: says the function is not implemented[00:10:20] <Latrina> when you do modprobe tun[00:10:47] <bahamat> Smithx10: Not really, no.[00:10:50] <Latrina> at least on alpine linux image_uuid 19aa3328-0025-11e7-a19a-c39077bfd4cf[00:11:07] <jperkin> Latrina: I mean using vpn in a native zone[00:11:23] <Latrina> oh I see[00:11:30] <Smithx10> bahamat: okie dokie..... is the monitoring endpoints just registered on the sapi service end point for them?[00:11:36] <bahamat> Smithx10: The best thing to do is poll sapi for a list of instances, then write a json file of targets and configure prometheus to use the file discovery.[00:11:57] <Smithx10> Got it[00:11:58] <bahamat> The monitoring endpoints aren't registered anywhere, you just have to know them.[00:11:58] <Smithx10> thank you :)[00:12:00] <Latrina> didn't honestly try[00:12:27] <Latrina> but I recon openvpn will fail when trying to create a tun device for the very same reason[00:12:31] <Latrina> my guess..[00:12:32] <Smithx10> Are all the ports different?[00:12:36] <bahamat> So you can poll sapi for the instances, get the IP that you want to query, but you need to check the service to see which port the metrics are listening on.[00:12:40] <bahamat> Yeah, it varies.[00:12:42] <Smithx10> k[00:12:50] <Smithx10> ill just ssh in and netstat -an[00:13:04] <bahamat> e.g., muskie has 16 smf instances, each one listens on a different port.[00:13:18] <bahamat> ssh+netstat isn't going to get you there.[00:13:25] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has joined #smartos[00:13:30] <Smithx10> ps aux ?[00:13:31] <Smithx10> :P[00:13:43] <Smithx10> or is there just soooomannnnyyy ports[00:13:54] <bahamat> Well, each service is different.[00:14:56] <bahamat> Eventually this will all be exposed through the cmon front door.[00:15:03] <bahamat> The Triton metrics already are.[00:16:49] <bahamat> Polling sapi for it is really a bad way of doing it.[00:16:57] <Smithx10> What is the best way to message RKeine?[00:18:00] <bahamat> It's actually the manta team that needs make a plugin that will wire it all up.[00:23:25] <Smithx10> I was hoping to maybe put a PR into prometheus for the triton SD for Global Zones[00:23:46] <Smithx10> You have a Prometheus in the joyent github org right?[00:24:22] <bahamat> Yeah, but it's not intended for general use.[00:24:29] <Smithx10> gotcha[00:24:40] <bahamat> That one will be a core triton service in the future, right now it's very nascent.[00:25:03] <Smithx10> I saw you discussing thanos[00:25:10] <Smithx10> Ive been playing aorund with it also[00:25:13] <Smithx10> for our applications[00:25:34] <bahamat> The best thing to do is just build your own prometheus on SmartOS or use their linux binaries.[00:27:23] <Smithx10> yeah, but I added GZ service discovery for cmon[00:27:39] <Smithx10> in the binary[00:27:45] <Smithx10> ill just use lx[00:29:20] <bahamat> Well the triton metrics are already exposed via the instances.[00:29:29] <bahamat> If you query using your admin account.[00:31:17] *** mmalecki <mmalecki!~maciej@144.76.114.210> has quit IRC (Ping timeout: 258 seconds)[00:32:46] <LeftWing> Latrina: For what it's worth, OpenVPN works well in a regular (native) SmartOS zone.[00:33:49] <Latrina> LeftWing: I will consider that in the future, but now I really need to deploy this vpn to tor gw fast[00:34:05] <Latrina> and with the knowledge I already have. Good to know though[00:34:18] <Latrina> what fw does smartos use? pf?[00:35:12] <LeftWing> ipfilter[00:35:35] <LeftWing> (and ipnat)[00:35:51] <Latrina> oh, never used neither of the two.[00:36:15] <Smithx10> LeftWing: is going to replace ipfilter with PF tomorrow[00:36:31] <Smithx10> He has a secret branch out there in the ether[00:36:55] <Smithx10> :P[00:40:09] <Smithx10> but on a more realistic note.... I think The Other Solaris ported over PF....[00:42:00] <Smithx10> Don't know what the ROI is on that is tho[00:44:17] *** yomisei <yomisei!~void@ip4d16bd91.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 246 seconds)[00:45:12] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Quit: andy_js)[00:46:47] <Latrina> LeftWing: please do :D[00:47:05] <Latrina> yeah they did port PF since 11[00:47:40] <Smithx10> Looks like they left a few things ou[00:47:53] <Smithx10> https://blogs.oracle.com/solaris/migrating-from-ipf-to-packet-filter-in-solaris-114[00:48:17] <Smithx10> pfsync and bandwidth management were not ported[00:48:51] <Smithx10> Latrina: haha, I was just kidding.... i dont think anyone is doing anything with that[00:49:07] <Smithx10> the only thing related was the work jbk had been doing with the nat zones[00:49:18] <Latrina> Smithx10: ahaha I know man[00:49:23] <Smithx10> That I am aware of[00:49:44] <Latrina> I just wanted to amplify the joke ;)[00:49:51] <Smithx10> lololo Latrina, gotta make sure.... don't want people digging through logs in the future to think "OMG PF!"?!$?!$ Secret branches[00:49:53] <Latrina> by acknowledging it[00:50:14] <Smithx10> but.... i did hear that illumos is pulling in eBPF[00:50:18] <Smithx10> :P[00:50:56] <bahamat> Smithx10: btw, RFD 150 discusses exposing all of this through the cmon front end.[00:50:57] <jinni> https://github.com/joyent/rfd/tree/master/rfd/0150[00:50:57] <Smithx10> RZ has decided fixing the networking stack was a bad idea and instead is going to implement a packet filter to do it.[00:51:09] <Latrina> well at least they will be entertained enough instead of wasting time on twitter bullshitting saying illumos is dead...[00:51:40] <Smithx10> yea there are some rumours going around[00:52:14] <Smithx10> I think of it as a great thing[00:52:35] <Smithx10> let everyone keep playing with lx kubernetes nonsense.[00:53:22] <Smithx10> Latrina: did you see https://marc.info/?l=linux-kernel&m=154714516832389&w=2[00:53:32] <Latrina> post-up ip link set dev eth2 mtu 9000[00:53:38] <Latrina> wrong paste[00:53:51] <Smithx10> maybe..... your users..... might..... want zfs......[00:54:30] <Latrina> lol what a jerk[00:54:47] <Latrina> if you want to use it make it to work[00:54:54] <Latrina> otherwise don't comply and abandon it[00:54:59] <Latrina> *complain[00:55:45] <Latrina> lol GPL condom[00:55:56] * LeftWing twitches[00:56:03] <Smithx10> this is even more so entertaining[00:56:04] <Smithx10> https://marc.info/?l=linux-kernel&m=154722999728768&w=2[00:56:17] <Latrina> anyways, as far as ZoL is concerned, if I was them I'd rather focus on BtrFS[00:56:20] <LeftWing> Lets not have an LKML party :P[00:56:32] <Smithx10> ok ok ok[00:56:38] <Smithx10> Sorry :([00:56:52] <Smithx10> but Latrina and I were having to much fun joking[00:56:56] <Smithx10> toooo*[00:57:03] <Latrina> jaja[00:57:11] <Smithx10> Latrina: no more poking fun of others.[00:57:26] <Latrina> ;)[00:57:28] <LeftWing> :P[00:57:44] <Smithx10> It's easy to get carried away[01:15:31] *** bahamat <bahamat!~bahamat@72.2.113.233> has quit IRC (Quit: ZNC - http://znc.in)[01:16:19] *** bahamat <bahamat!~bahamat@72.2.113.233> has joined #smartos[01:26:12] <Latrina> is it possible to create a kvm zone and add a raw.img disk as default boot disk?[01:26:36] <Latrina> it is a premade image wich contain kernel, rootfs and everything. No install required[01:28:02] <Latrina> this is what I need https://openwrt.org/docs/guide-user/virtualization/virtualbox-vm[01:28:15] <Latrina> in vbox I convert the .img to .vdi and works[01:28:31] <Latrina> not sure how I would go about doing this in smartos[01:28:34] <LeftWing> Latrina: I'd be inclined to create a blank disk and dd the image file onto the zvol before booting it[01:28:43] <Latrina> oh[01:28:53] <Latrina> I will try that, thanks[01:36:52] *** TyrfingMjolnir <TyrfingMjolnir!~Tyrfing@62.92.82.250> has quit IRC (Ping timeout: 272 seconds)[01:43:32] <Latrina> is there a particular way to umount a zvol after creating the zone?[01:43:34] <Latrina> https://pastebin.com/3KcUsy1X[01:44:07] <Latrina> it suggests the zvol might be mounted, but I can't seem to umount /zones/f82ea1e2-836c-c0b7-e6b7-f6e1b3121c58/cores[01:44:57] <rmustacc> don't you want the rdsk node which is the character device, as opposed to the buffered block device?[01:45:15] <rmustacc> A zvol cannot be mounted.[01:45:23] <rmustacc> At least not in a conventional sense generally.[01:46:15] <Latrina> rmustacc: I have never done it, so I guess I am not doing the right way[01:46:22] <Latrina> will try your way. thanks[01:47:42] <Latrina> so you mean dd the .img to /dev/zvol/rdsk/zones/f82ea1e2-836c-c0b7-e6b7-f6e1b3121c58-disk0 instead?[01:48:31] <rmustacc> Yes.[01:48:45] <rmustacc> /dev/zvol/rdsk are character devices, where as /dev/zvol/dsk are block devices which have rather different semantics.[01:49:04] <rmustacc> Latrina: Also, you're using df, not dd.[01:49:12] <rmustacc> Which is why you're seeing that message.[01:49:46] <Latrina> that was a type my bad[01:50:15] <rmustacc> No problem, it happens.[01:58:06] <Latrina> rmustacc: that worked.. thanks man :)[02:01:56] *** strobelight <strobelight!~quassel@c-69-180-7-202.hsd1.ga.comcast.net> has joined #smartos[02:23:51] *** v_a_b <v_a_b!~volker@p57A27EC0.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 268 seconds)[02:29:25] <Smithx10> What is engbld framework?[02:41:33] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Ping timeout: 246 seconds)[02:49:17] <Smithx10> bahamat: did you add hostname to the meta labels to in cmon?[02:49:23] <Smithx10> I only see server_id[02:49:27] <Smithx10> for GZ*[02:49:53] <bahamat> There's nowhere to get that info from outside of triton[02:51:46] <Smithx10> ? wouldnt the cmon agent for the gz just export a meta_label for the hostname?[02:53:27] <bahamat> It doesn't, currently.[02:55:21] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos[02:56:03] <Smithx10> ok, would that be accepted?[02:56:16] <bahamat> If you submitted a patch?[02:58:28] <Smithx10> Sweet[02:58:38] <Smithx10> Ill take a look :)[03:03:57] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[03:10:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:12:37] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:16:43] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:17:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:19:38] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:20:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:22:01] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:23:39] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:23:58] <rsully> Is smartos fine with 4Kn drives now? The wiki says they're not compatible, but looking at https://docs.joyent.com/private-cloud/hardware shows drives like "HUC101812CS4200 1.2TB SAS 12Gb/s 10K RPM 128M 4kn ISE"[03:24:58] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:25:01] <rmustacc> We've been using 4kn drives for years.[03:25:53] <rsully> ok good to know[03:26:07] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:26:27] <rsully> are SAS drives worth it over their SATA counterparts?[03:26:48] <rsully> looks like one drawback is most of them consume more power[03:27:26] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:28:34] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:30:16] <Smithx10> rmustacc: Are there SATA SSDs that you suggest?[03:30:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:31:24] <rmustacc> Depends on the purpose.[03:32:15] <rsully> that answer likely applies to both of our questions ^^[03:32:17] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:33:10] <Smithx10> I noticed https://docs.joyent.com/private-cloud/hardware#joyent-tenderloin-c[03:33:28] <Smithx10> has sata ssd 9 370-0006-01 Joyent, Inc. N/A SATA SSD, 800 GB, 2.5, w/firmware[03:34:04] <rsully> lol was just going to paste that[03:34:04] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:34:15] <rsully> that's an DC S3700 800 GB SSD[03:35:07] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:36:47] <Smithx10> I'm guessing the question is do you trust a pool of 2 tb sata ssd?[03:37:00] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:37:29] <Smithx10> for general purpose compute without any ridiculous performance demands[03:37:59] *** strobelight_ <strobelight_!~quassel@2001:420:c0c8:1002::3b3> has joined #smartos[03:37:59] *** strobelight is now known as Guest97276[03:38:32] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:38:43] <rsully> Smithx10 is this for slog or do you mean a pool of just ssds?[03:38:53] <Smithx10> pool of just ssd[03:39:00] <rsully> oof. I wouldn't trust that[03:39:02] *** strobelight_ is now known as strobelight[03:39:10] <rsully> just make sure you have backups :)[03:39:13] <Smithx10> lololol[03:39:18] <Smithx10> Well we didnt buy anything[03:39:34] <Smithx10> ASking the hardware gods first.... ./bows to rmustacc[03:39:50] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:39:50] <rsully> I'd go for a pool of mirrored spinning disks, and add a couple DC level SSD for caching[03:39:59] <rsully> that should give enough performance[03:40:11] <rsully> it won't compare to all-SSD obviously, so depends on workload/requirements[03:40:56] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:41:08] <Smithx10> yea, we may be doing Databases soon[03:41:26] <Smithx10> and might just pay the money to save time[03:41:48] *** Guest97276 <Guest97276!~quassel@c-69-180-7-202.hsd1.ga.comcast.net> has quit IRC (Ping timeout: 258 seconds)[03:42:24] <rsully> even with databases, depends on the requirements, depends how its tuned, depends on what type of queries you expect[03:42:39] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:42:54] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:43:08] <Smithx10> yea.... but the time thing[03:43:18] <Smithx10> if you have 1020012040214 iops[03:43:21] <Smithx10> life can be easier[03:43:31] <rsully> hey if money is no object have a blast :)[03:43:36] <Smithx10> it kind of i[03:43:40] <Smithx10> It kind of is*[03:44:06] <Smithx10> I'd still like to know about where sata should fit in to our CN tiers[03:44:07] <rsully> what does the rest of the build look like? how many disks and what configuration are you planning for?[03:44:17] <Smithx10> currentyl we only have 3 nodes, our PoC[03:44:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:44:55] <Smithx10> and they are simple 8 2.2gb SAS into a LSI HBa 12gb with a write intense Log device[03:45:00] <Smithx10> same on all 3[03:45:22] <Smithx10> we will probably put more mirrored pairs in these pools and get to about 14 drives and a hot spare[03:46:13] <Smithx10> but I was curious if the Sata SSDs make sense in a given scenario, but it seems like the general opinion is to run SAS if you have pockets.[03:46:18] <Smithx10> deep *[03:46:26] <Smithx10> Well NVME if you wanna get fannnncccyyyyy[03:47:06] <rsully> from what I've read it sounds like SAS is only a benefit over SATA if you're hitting the 6 Gbps limit (and any SSD in the 400-500MB/s range fits in there nicely)[03:47:27] <Smithx10> https://illumos.topicbox.com/groups/zfs/T8fb5137060ef6844-Ma68dd64ed00322ffb192ac24[03:47:38] <Smithx10> https://www.vikingenterprisesolutions.com/products-2/nds-2244/[03:48:19] <rsully> well yeah NVMe is a whole different ballgame, for that you need sas or pcie[03:48:35] <Smithx10> I just came across that[03:49:15] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:50:34] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:51:37] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[03:52:19] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[03:54:42] *** jellydonut <jellydonut!~jelly@77.16.68.144.tmi.telenormobil.no> has quit IRC (Remote host closed the connection)[04:07:40] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)[04:07:59] *** strobelight <strobelight!~quassel@2001:420:c0c8:1002::3b3> has quit IRC (Ping timeout: 268 seconds)[04:09:27] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[04:12:32] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[04:13:56] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:14:59] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:16:21] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:17:27] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:19:19] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:19:55] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:21:49] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:22:47] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:24:14] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:25:17] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:27:09] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:28:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:33:45] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:35:18] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:40:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[04:45:19] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[04:52:31] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[05:03:42] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Read error: Connection reset by peer)[05:03:45] *** sjorge_be <sjorge_be!~sjorge@unaffiliated/sjorge> has joined #smartos[05:04:09] *** sjorge_be is now known as sjorge[05:09:31] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[05:09:48] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[05:35:16] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[05:45:57] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[05:46:07] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[05:55:13] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)[05:57:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[05:58:16] *** miller7 <miller7!~miller7@unaffiliated/miller7> has joined #smartos[06:00:30] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[06:01:53] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[06:22:46] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[06:24:37] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[06:24:52] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[06:25:16] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[06:26:28] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos[06:30:45] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[06:42:03] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[06:42:22] *** yomisei <yomisei!~void@ip4d16bd91.dynamic.kabel-deutschland.de> has joined #smartos[06:54:34] <jesse_> uh-oh, https://www.theregister.co.uk/2019/01/15/scp_vulnerability/[07:00:27] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:15:21] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[07:15:45] *** TyrfingMjolnir <TyrfingMjolnir!~Tyrfing@62.92.82.250> has joined #smartos[07:18:31] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:18:47] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:20:26] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:22:53] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:23:30] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:23:50] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:24:30] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:27:29] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:27:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:29:14] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:30:15] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:30:30] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:30:46] *** neuroserve <neuroserve!~toens@195.71.113.124> has joined #smartos[07:32:15] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:33:27] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:35:00] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:35:57] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:38:05] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:38:20] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)[07:38:28] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[07:41:51] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[07:46:10] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Ping timeout: 246 seconds)[07:54:22] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)[08:33:14] *** JeanParpaillon <JeanParpaillon!~jean@176.164.92.169> has joined #smartos[09:35:43] *** v_a_b <v_a_b!~volker@p57A27D5A.dip0.t-ipconnect.de> has joined #smartos[09:44:53] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has joined #smartos[09:56:13] *** hotbox_ is now known as hotbox[10:08:38] *** gemelen <gemelen!~gemelen@zooey.gemelen.net> has joined #smartos[10:23:03] *** man_u <man_u!~manu@manu2.gandi.net> has joined #smartos[10:55:36] *** leah2 <leah2!~leah@vuxu.org> has quit IRC (Ping timeout: 260 seconds)[11:00:02] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[12:07:53] *** Latr_work <Latr_work!uid4039@gateway/web/irccloud.com/x-lqvemqlzgkdrjcjj> has joined #smartos[12:08:03] <Latr_work> Good morning peps[12:25:09] <Latr_work> right guys, so yesterday night I created a local ethersub called int_ovpn1 and a VNIC named ovpn1 on top of it.[12:25:43] <Latr_work> However, VNIC name seems to have suddenly changed into net1?[12:26:40] <Latr_work> I have realized it when I started troubleshooting the actual problem, and that is I am unable to ping the assigned VM IP from GZ and from anywhere else really https://pastebin.com/BwxNdUNk[12:28:18] <Latr_work> The entire VM configuration looks like this https://pastebin.com/7SbjJuRC[12:28:26] <Latr_work> not really sure what I am doing wrong[12:39:17] *** leah2 <leah2!~leah@vuxu.org> has joined #smartos[12:49:29] <Latr_work> right, I will add a ipaddr entry see if it makes any difference[13:08:12] <wiedi> Latr_work: the interface property is assigned automatically, I don't think you can give it arbitrary names[13:11:07] <wiedi> which IP are you trying to ping? The one on the admin nic tag or the one of the etherstub?[13:23:22] <Latr_work> wiedi: well apparently it lets create me a vnic with my own name, but it will be renamed netX upon reboot[13:23:25] <Latr_work> who knows[13:23:55] <Latr_work> I am trying to ping the VM IP, which is 10.10.0.1[13:24:08] <Latr_work> etherstub does not have an IP. Do I need to assign it one?[13:27:11] <wiedi> yes, for KVM depending on the image you'll have to set this manually. And also when you want to ping the address from the GZ you'll need a way to reach it - so either have the vm route it via the other interface or plumb the etherstub in the gz and assign it an address[13:27:49] <Latr_work> wiedi: VMs on the same vnic can't ping eachoter either[13:28:25] <Latr_work> so lets recap, for KVM, do I need to assign an IP to both local etherstub as well as VNIC?[13:32:26] <wiedi> I think qemu should give you the address also via dhcp[13:33:13] <Latr_work> [root@homelab ~]# dladm show-vnic[13:33:13] <Latr_work> LINK OVER SPEED MACADDRESS MACADDRTYPE VID ZONE[13:33:13] <Latr_work> eth0 bnx2 0 72:52:6b:86:da:b4 fixed 0 efc6e98c-6aee-c819-9a36-d734620a292c[13:33:13] <Latr_work> net0 bnx2 0 72:bd:20:79:9a:cf fixed 0 4da9efe9-4904-cacd-d4ac-c3a34b119472[13:33:13] <Latr_work> happyvnic1 happyether0 0 8:0:27:c1:50:1a fixed 0 --[13:33:13] <Latr_work> [root@homelab ~]# ipadm create-addr -T static -a 192.10.0.0/24 happyvnic1/static[13:33:13] <Latr_work> ipadm: Could not create address: Can't assign requested address[13:33:21] <wiedi> can you share the "vmadm get" of your vms and the "ifconfig" output from inside the vms?[13:33:38] <Latr_work> sure[13:34:27] <Latr_work> but still I am wondering if the vnic of the local etherstub which I will than pass it to the VM needs to have an entry created in create-addr[13:34:39] <Latr_work> as I was attempting to do above[13:34:58] <wiedi> no[13:35:24] <wiedi> you might want to do something like that if you want to access the etherstub from the gz[13:35:26] <Latr_work> so I don't need to assign an IP to the vnic right?[13:35:34] <Latr_work> I dont need that[13:35:36] <wiedi> not in the gz[13:35:43] <Latr_work> okay[13:35:49] <Latr_work> so I was in the right track[13:36:36] <Latr_work> vmadm get of the affected vm https://pastebin.com/RapKa3Gs[13:37:50] <Latr_work> and ifcoing inside https://pastebin.com/uhgcwu8y[13:38:44] <Latr_work> if I create another kvm vm and I assign it the very same etherstub and VNIC[13:39:07] <Latr_work> the two vms wont reach eachother when they should[13:39:14] <wiedi> they should indeed[13:39:25] <Latr_work> eth1 is configured to act as dhcp server[13:39:28] <wiedi> you've changed the mac and ip addresses? (just making sure)[13:39:33] <Latr_work> but neither static would work[13:39:41] <Latr_work> course lol[13:40:11] <Latr_work> oh well, after the df instead of dd from yesterday night. you have a point to ask lol[13:40:18] <Latr_work> but yes[13:40:30] <Latr_work> I am going nuts[13:40:35] <wiedi> can you share the same dettails (vmadm get + ifconfig) from the second vm?[13:40:43] <Latr_work> sure thing[13:44:08] <Latr_work> I am actually recreating it sorry[13:44:14] <Latr_work> got deleted during the mess[13:56:13] <Latr_work> vmadm get of vm2 https://pastebin.com/0yTqmNzT[13:59:43] <Latr_work> this has to be in screenshot form as I have no ssh access to this side https://imgur.com/zx8lvU6[14:00:03] <Latr_work> it wont get the dhcp ip as vm1 is unreachable[14:00:22] <Latr_work> if I add it manually, it wont be able to ping it due to the very same reason[14:01:36] <wiedi> does vm1 offer a dhcp server?[14:02:26] <drscream> and if so is "allow_dhcp_spoofing": true configured on the interface of vm1 which offers the dhcp server[14:04:45] <wiedi> if vm1 does not run a dhcp server then you'll have to set an ip for vm2 just like you did for vm1[14:12:53] <neuroserve> anyone using check_mk on smartos/triton? do we have libstatgrab available for smartos/pkgsrc?[14:13:46] <jperkin> I don't know what check_mk is (sounds very generic!) but we do have libstatgrab[14:24:40] <neuroserve> jperkin: jup - the agent is just a bash-script - but it says # You need statgrab to be installed. You'll find it[14:24:41] <neuroserve> # here: http://www.i-scream.org/libstatgrab/[14:26:40] <neuroserve> jperkin : is libstatgrab part of pkgsrc or of smartos?[14:26:51] <jperkin> pkgsrc[14:30:22] <Latr_work> back sorry[14:30:38] <Latr_work> yes vm1 has dhcpd configured on eth1[14:31:06] <Latr_work> if I am not mistaken allow_dhcp_spoofing is already specified on vm1 for the vnic[14:31:16] <Latr_work> do I need to add it on vm2 as well ?[14:31:26] <Latr_work> I will double check that though[14:35:15] <drscream> Latr_work: no you need that option only on the VM with the dhcp server[14:56:25] <Latr_work> okay I am adding it[14:56:34] <Latr_work> I had allow_ip_spoofing only[14:56:40] <Latr_work> lets hope that does the trick[15:08:27] <Latr_work> unfortunately that didn't do it either :S[15:11:00] <Latr_work> god know where is the holdup ...[15:11:07] <wiedi> Latr_work: you'll also need to allow_dhcp_spoofing[15:11:21] <wiedi> for net1 on vm1[15:11:21] <Latr_work> it is there now[15:11:29] <Latr_work> but made no difference[15:11:46] <Latr_work> https://www.irccloud.com/pastebin/1I5nZ0GA/[15:11:54] <Latr_work> that is VM1 now[15:14:18] <wiedi> ok, can you verify with tcpdump on vm1 that the dhcp packets go out, and on vm2 that you see them comming in?[15:15:17] <Latr_work> I wonder if allow_ip_spoofing must be also added on vm2 vnic[15:15:33] <wiedi> should not be needed[15:15:59] <Latr_work> the man says[15:16:02] <Latr_work> With this property set to true, this VM will be able to send and[15:16:02] <Latr_work> receive packets over this nic that don't match the IP address[15:16:02] <Latr_work> specified by the ip property.[15:16:26] <Latr_work> I will do this as last attempt before collecting a packet trace[15:16:31] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos[15:16:59] <wiedi> since you have set it to dhcp the dhcp assigned ip should be added to the allowed list[15:17:24] <Latr_work> on vm2 right?[15:17:31] <wiedi> yes[15:17:35] <Latr_work> right[15:20:14] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has quit IRC (Ping timeout: 268 seconds)[15:23:03] <Latr_work> I also see this flag for vm1 could be a potential nics.*.dhcp_server:[15:24:00] <wiedi> hm, not sure what the difference is between that and allow_dhcp_spoofing[15:24:18] <Latr_work> not sure either[15:24:46] <Latr_work> but than again, it doesnt even work through static so I dont see the point in fixing the issue either way[15:25:13] <wiedi> yeah, maybe let's figure out why static doesn't work first[15:25:48] <wiedi> you've created vm1 with static and it got the ip. Did this also work for vm2 at one point?[15:29:03] <mattronix> one of our trition nodes is stuck in a state[15:29:08] <mattronix> where we cant SSH into it[15:29:10] <mattronix> but its online[15:29:16] <mattronix> like a vm is causing some havoc[15:29:29] <mattronix> how can i get to the CLI of that fucker say after rebooting into single usermode[15:29:35] <mattronix> can i pass a boot flag to it somehow?[15:29:39] <mattronix> using trition dashboard[15:30:45] <Latr_work> wiedi: nope static wont work on vm2[15:30:52] <Latr_work> even if I add it in the json template[15:31:03] <Latr_work> btw what does this mean[15:31:04] <Latr_work> Object 10.10.0.2 has no method 'forEach'[15:31:11] <wiedi> Latr_work: can you share vmadm get / ifconfig[15:31:13] <wiedi> uh[15:31:29] <Latr_work> of which vm?[15:31:38] <wiedi> the one that doesn't work, vm2[15:31:46] <Latr_work> "allowed_ips": "10.10.0.2",[15:31:52] <Latr_work> it is not liken it I guess[15:32:10] <wiedi> try "allowed_ips": ["10.10.0.2"],[15:34:20] <Latr_work> worked :)[15:34:37] <Latr_work> firing it up now[15:35:57] <Latr_work> is vmadm get / ifconfig supposed to be a command?[15:36:06] <Latr_work> or the two individually you meant?[15:38:30] <wiedi> two commands, "vmadm get" in the gz and "ifconfig" from inside the vm[15:38:54] <Latr_work> oh okay.. sorry[15:39:27] <Latr_work> for a second I thought I could have extracted the ifconfig of the vm from vmadm[15:41:44] <wiedi> you can do "zlogin <uuid> ifconfig" for native zones, but not for kvm ;)[15:41:52] <Latr_work> blob:https://imgur.com/6f19f3d7-7ec5-40f3-8202-c44b365d9515[15:42:12] <Latr_work> sorry I meant wiedi https://imgur.com/NY3hL2w[15:42:37] <Latr_work> I will move into static address[15:43:08] <Latr_work> it is pointless to insist with the dhcp when I cant get static to work to begin with[15:43:25] <Latr_work> never took me this long to bring up such a simple setup lol[15:43:33] <Latr_work> and will packet trace[15:44:16] <wiedi> i'm sure it's just some tiny config issue, we'll figure it out ;)[15:44:43] <Latr_work> oh I am sure it is. Thanks for your help though. It looks like I have rented you out today[15:52:38] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[15:55:35] <Latr_work> also look this weird thing, this is how vm2 looks like right now on static ip https://pastebin.com/Vsmc9ixS[15:55:48] <Latr_work> it is going through the wrong ethersub ...[15:55:58] <Latr_work> even though I clearly specified net1 as ethersub[15:56:47] <Latr_work> this is its add_nics conf https://pastebin.com/Tz3R6DEv[15:57:03] <Latr_work> I need a walk or I will pass out[15:59:48] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos[16:03:19] *** polishdub <polishdub!~polishdub@207.86.38.254> has joined #smartos[16:12:29] *** psarria <psarria!~phyre___@31.211.191.164> has joined #smartos[16:26:22] <Latr_work> at this point I am starting to wonder if it is possible to feed the same vnic to more than a kvm vm[16:26:48] <Latr_work> in dladm show-vnic when I fire a vm which uses the net1 nictag[16:27:11] <Latr_work> the VNIC will change its MAC according to last fired VM[16:27:46] <Latr_work> sorry I meant interface and not nic_tag[16:28:28] *** neuroserve <neuroserve!~toens@195.71.113.124> has quit IRC (Ping timeout: 250 seconds)[16:30:25] <Latr_work> I am giving it a restart[16:52:51] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[17:03:37] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-rjtvablkxrlhgltc> has quit IRC (Excess Flood)[17:08:31] <mattronix> got a wierd error[17:08:36] <mattronix> kernel panic[17:08:51] <mattronix> ../common/fs/zfs/zfs_znode.c line:820[17:08:55] <mattronix> cant get the os started[17:09:43] <mattronix> https://muc.lists.freebsd.fs.narkive.com/yZqCxwMh/solaris-assert-triggered-after-panic-in-zfs-at-mount-time[17:09:45] <mattronix> related?[17:13:18] *** greenline <greenline!~user@unaffiliated/greenline> has joined #smartos[17:52:34] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos[17:57:11] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[17:59:54] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 250 seconds)[18:00:51] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[18:19:47] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[18:24:15] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos[18:27:18] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 258 seconds)[18:38:35] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Quit: man_u)[18:40:59] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has quit IRC (Quit: bens1)[18:50:46] *** papertigers <papertigers!~papertige@pool-108-17-31-72.bflony.fios.verizon.net> has joined #smartos[18:54:03] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos[19:28:30] <bahamat> mattronix: Can you give me the full stack trace?[19:55:22] <mattronix> nope :([19:55:27] <mattronix> i cant even get it to boot[19:55:34] <mattronix> it pxes[19:55:40] <mattronix> does not show grub[19:55:42] <mattronix> then panics out[19:57:03] <mattronix> it also cant find a dump device[19:58:59] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[20:02:42] <mattronix> bahamat: any ideas how i can stablilize the machine or boot without importing zfs[20:02:52] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[20:04:52] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[20:05:12] <bahamat> mattronix: set noimport=true on the command line[20:05:17] <bahamat> Are you doing this via triton?[20:06:31] <mattronix> yes i am[20:06:43] <mattronix> but dont know how to set it in trition :([20:06:45] <bahamat> So there's a couple of options.[20:07:07] <bahamat> What we'll be doing is modifying the bootparams for just that compute node.[20:07:23] <bahamat> You can either boot noimport then try to examine the zpool manually[20:07:49] <bahamat> Or you can boot with kmdb enabled so that when it crashes you can get a stack trace.[20:08:03] <bahamat> You'll need to have access to the console in order to do this.[20:08:08] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[20:09:59] *** gatekeep <gatekeep!~gatekeep@pe-1-0.wan.rtr1.jmp.cx> has quit IRC (Remote host closed the connection)[20:11:12] <muchomas> ooh, fun[20:11:26] <muchomas> "fun"[20:12:11] <mattronix> ah i do have access to the console :)[20:12:18] <mattronix> but no grub menu appears[20:17:38] <mattronix> somestimes after reboot of headnode moray does not come back[20:17:44] <mattronix> also happeing atm[20:17:57] <mattronix> two unrelated issues but still blocks me from portal :([20:18:04] <mattronix> bahamat: so what steps do i need to do on the console?[20:19:13] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Remote host closed the connection)[20:19:24] <bahamat> mattronix: How do you access the console? keyboard/monitor or serial?[20:19:42] <mattronix> kvm[20:19:46] <mattronix> keybouard and mose[20:19:48] <mattronix> mouse[20:19:56] <mattronix> set the console to vga in trition portal[20:22:33] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos[20:47:29] *** greenline <greenline!~user@unaffiliated/greenline> has quit IRC (Remote host closed the connection)[20:51:43] *** JeanParpaillon <JeanParpaillon!~jean@176.164.92.169> has quit IRC (Quit: JeanParpaillon)[20:55:44] *** Smithx10 <Smithx10!sid243404@gateway/web/irccloud.com/x-voxettqzdfuoirqv> has joined #smartos[20:55:52] <Smithx10> "Cannot delete a VM with no server_uuid" what is the best way to fix this[20:56:00] <Smithx10> I was messing with vmapi and got a machine ina . strange state[20:59:50] <Smithx10> Im also seeing high cnapi.wait_task 5.862s . times[21:00:23] <Smithx10> One took 50.37s :([21:13:52] <black_pete> How did you manage that? server_uuid is computed when listing the properties, so it's not like you can overwrite it[21:16:04] <Smithx10> lolol black_pete I was messing around with adding a delegated dataset based ona . tag[21:16:53] <Smithx10> made a mistake while adding https://paste.ec/paste/WPlWroKU#38tdbmQwIoksLE6FgN5KPS2Lyqj6zvdYJMizXywyO5w[21:17:03] <black_pete> yikes; I'm figuring out delegated datasets right now, too. Hopefully I don't wind up in the same situation[21:17:40] <Smithx10> No no no it was my issue[21:18:48] <black_pete> that pastebin looks like you're automating using node[21:19:13] <Smithx10> its apart of the vmapi contianer[21:19:43] <Smithx10> "/opt/smartdc/vmapi/lib/workflows/provision.js"[21:20:12] <Smithx10> Until mgerdts adds this stuff properly to cloudapi etc.[21:22:05] <bahamat> mattronix: Sorry, I've got a few different things I'm trying to give my attention to.[21:23:01] <bahamat> Smithx10: sdc-vmapi /vms/:uuid?sync=true[21:23:02] <Smithx10> bahamat: why would i see such different times for cnapi.wait_task?[21:23:27] <bahamat> Smithx10: it depends on the task. You can see what task it is in the step immediately preceeding.[21:24:29] <Smithx10> The one I see taking a ton of time in the job is[21:24:29] <Smithx10> cnapi.wait_task[21:24:37] <Smithx10> Where can I see the tasks in there?[21:26:10] <Smithx10> bahamat: this worked https://docs.joyent.com/private-cloud/troubleshooting/stuck-provisions[21:28:00] <Smithx10> Im seeing that wait_task take 5 seconds and sometimes 60 hmmmm[21:29:15] <bahamat> Look at the job. It'll say something like 'cnapi.ensure_image' then 'cnapi.wait_task'[21:29:24] <bahamat> That means it's waiting on ensure_image.[21:29:43] <bahamat> Whenever there's an async step, wait_task is run to wait for that task to finish.[21:29:57] <bahamat> So whatever it did before wait_task is the thing it's waiting for.[21:32:18] <Smithx10> all those steps are 0.341s[21:32:22] <Smithx10> like ensure_image[21:32:46] <Smithx10> The only long step is the wait_task[21:32:49] <Smithx10> Is that normal?[21:37:21] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[21:40:09] <bahamat> ensure_image just dispatches the task.[21:40:17] <bahamat> Then wait_task waits for ensure_image to finish.[21:40:34] <bahamat> wait_task doesn't do anything. It just waits for the previous task to finish.[21:41:41] <bahamat> e.g.:[21:41:56] <bahamat> If your tasks are foo, bar, wait_task, baz, wait_task, bat[21:42:07] <bahamat> the first wait_taks waits for bar to finish[21:42:16] <bahamat> the second wait_task waits for bas to finish.[21:43:18] <Smithx10> Seems like 1f32508c-e6e9-11e6-bc05-8fea9e979940 base-64-lts 16.4.1 active P smartos zone-dataset 2017-01-30 takes 28 seconds longer to deploy ?[21:43:29] <Smithx10> than 7b5981c4-1889-11e7-b4c5-3f3bdfc9b88b ubuntu-16.04 20170403 active P linux lx-dataset 2017-04-03[21:45:30] <Smithx10> To the same node, and the image server should already have the origin[21:48:58] <Smithx10> Yea, very strnage... bhyve images also deploy faster, is there something different with the OS Containers that take longer?[21:49:03] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[21:49:51] <jperkin> I'm pretty sure there's a bunch of suboptimal things in zoneinit, I've just never had the time to go through it - something about an extra reboot prior to it being ready[21:50:19] <Smithx10> jperkin: Thank You, Just trying to confirm Im not experiencing anything goofie[21:51:13] <jperkin> not certain that's what you're seeing, but certain that zoneinit could do with some loving ;)[21:53:03] <Smithx10> does LX use zoneinit?[22:06:11] <plluksie> Hello. I'm wondering if someone could help understand what I'm doing wrong. I'm experimenting with ac99517a-72ac-44c0-90e6-c7ce3d944a0a. If I'm using brand "kvm" - all is fine. If I switch to "bhyve", I have no connectivity to the Internet. The rest of the configuration is exactly the same. The only difference is the brand. Is this expected?[22:24:58] *** psarria <psarria!~phyre___@31.211.191.164> has quit IRC (Ping timeout: 258 seconds)[22:43:48] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)[22:44:30] *** olsner <olsner!~salparot@c193-14-198-108.bredband.comhem.se> has joined #smartos[22:49:50] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[22:50:49] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[22:52:45] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[22:53:02] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[22:53:47] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)[22:54:35] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[22:55:40] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[22:57:09] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has joined #smartos[22:57:35] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[22:57:37] <mgerdts> plluksie that is not expected with that image. On the compute node, there should be a console log for the bhyve instance. Can you provide me with that? You can find it at /zones/$uuid/logs/console.log. Please don't paste the whole thing here. A link to pastebin or similar is fine. If you have sensitive information in your metadata service you should clean that up before making it available to anyone[22:57:37] <mgerdts> you don't trust.[22:59:35] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[22:59:50] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[23:00:37] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 258 seconds)[23:01:07] *** Kurlon_ <Kurlon_!~Kurlon@bidd-pub-04.gwi.net> has quit IRC (Ping timeout: 240 seconds)[23:01:54] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[23:03:20] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[23:04:57] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[23:06:32] <nahamu> Smithx10: os zones initialize themselves and reboot before coming online.[23:06:45] <nahamu> If you set autoboot to false and then boot them, they can come up pretty fast.[23:07:03] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[23:07:04] <nahamu> *set autoboot to false at creation time, then boot them after creation[23:07:21] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[23:07:44] <nahamu> see also the joyent_minimal brand.[23:08:44] <nahamu> *joyent-minimal[23:12:05] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos[23:12:30] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)[23:19:24] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has joined #smartos[23:20:11] <mgerdts> Smithx10 FYI: "we" are looking into the centos 7 resolver issue.[23:22:09] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Remote host closed the connection)[23:30:26] <Latrina> wiedi: so yeah, VM1 doesn't know how to reach VM2 on the same vnic https://imgur.com/dmf0Lz6[23:30:34] <Latrina> the result of the packet trace[23:30:52] <Latrina> at this point I am not really sure how to tell it where to look for.[23:31:56] <Latrina> I wonder if this would help allow_unfiltered_promisc:[23:34:25] <Latrina> this is how the current situation looks[23:34:28] <Latrina> vm1 https://pastebin.com/B4QQD3vF[23:35:16] <Latrina> https://pastebin.com/z6zpAAUK[23:36:31] <Latrina> vm2 https://pastebin.com/6dZhaVqZ[23:37:07] <Latrina> vm2 discover no IP from the dhcpserver, and despite going manual, it just fails to ping anything[23:37:50] <mattronix> bahamat: if you get some time just let me know[23:38:04] <mattronix> bahamat: all i need is the steps to toggale the no import flag the rest i can do myself[23:39:33] <Smithx10> mgerdts: am I in trouble?[23:39:46] <Smithx10> What was the "we" :P[23:39:48] <Smithx10> ./runs![23:40:22] <Smithx10> anytime i see "'s I get scared.[23:43:21] <bahamat> mattronix: From the headnode run: sdc-cnapi /boot/$uuid -X POST -d '{"kernel_args": { "noimport": true } }'[23:47:09] <mgerdts> ha! "we" as in others at joyent are poking at it.[23:48:19] *** ingenthr <ingenthr!~ingenthr@cpe-108-184-198-4.socal.res.rr.com> has joined #smartos[23:48:49] <Smithx10> gotcha, for some reason our network team has 8.8.8.8 blocked......[23:49:02] <Smithx10> :([23:52:53] *** macktronics <macktronics!~mack@olive.macktronics.com> has quit IRC (Remote host closed the connection)