Switch to DuckDuckGo Search
   January 10, 2019  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >

Toggle Join/Part | bottom
[00:00:58] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:02:37] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:04:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:05:03] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:06:24] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:07:55] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:09:38] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:09:54] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:14:19] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[00:14:23] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:15:24] <Smithx10> if im connected over the ipmi ilo
[00:15:31] <Smithx10> what os_console should i set?
[00:15:43] <Smithx10> I thought the default was fine.
[00:19:09] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 244 seconds)
[00:21:15] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)
[00:21:38] <LeftWing> Smithx10: That USB thing is ... sigh.
[00:22:27] <Smithx10> Also seems like it isnt taking keystrokes in the java client when using VGA
[00:22:33] <Smithx10> i think that could be just my make..... but before booting it does
[00:22:40] <Smithx10> im trying ttya with impitool now
[00:22:53] <LeftWing> It's possible the USB thing is related to the keystrokes not working there.
[00:22:55] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:23:03] <Smithx10> ahhhh perhaps ilo is pwning me
[00:23:10] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:23:17] <Smithx10> i was able to boot smartos.iso and get a tty in impitool
[00:23:21] <LeftWing> Nice
[00:23:22] <richlowe> I'm still not convinced the SOF thing is actually necessary.
[00:23:25] <LeftWing> richlowe: Right
[00:23:36] <Smithx10> it dropped me into the triton prompt
[00:23:45] <LeftWing> I mean, I think it would be strictly no worse to attach anyway even if it's not going to work sometimes.
[00:23:52] <Smithx10> but now with the triton install bundle im trying ttya and tried ttyb but didnt see anything
[00:24:07] <Smithx10> Is there something magically i need to do.... or is it just not drawing because of the cool asci boxes?
[00:24:28] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:25:35] <LeftWing> err
[00:25:56] <Smithx10> just confirmed its supposed to be ttyb right?
[00:26:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:26:06] <LeftWing> Well it depends on the machine, really.
[00:26:24] <LeftWing> The BIOS/firmware config stuff might let you pick
[00:26:46] <Smithx10> smartos vanilla booted into ttyb just fine
[00:26:52] <LeftWing> The console is ttyb on my ProLiant DL580 G7
[00:26:58] <Smithx10> IN the grub menu i used ttyb
[00:27:00] <LeftWing> But that's pretty old, and I'm not sure if I configured it to get it there.
[00:27:02] <LeftWing> OK
[00:27:04] <Smithx10> and it worked over ipmitool
[00:27:09] <LeftWing> That's good
[00:27:25] <LeftWing> So is the problem that the Triton media installer isn't working?
[00:27:31] <Smithx10> im trying again
[00:27:33] <LeftWing> OK
[00:27:42] <Smithx10> going to wait for it
[00:32:47] <jbk> fyi, it doesn't appear (at least from a brief glance) that fbsd is doing a similar wait for uhci
[00:33:13] <jbk> it enables interrupts then checks for some status bits (vs. waiting for an interrupt)
[00:42:23] <Smithx10> running VSP from the ilo results in Virtual Serial Port Active: COM1
[00:42:24] <Smithx10> lolol
[00:42:36] <LeftWing> https://www.illumos.org/issues/1857
[00:42:36] <Smithx10> wonder why VGA keyboard doesnt work
[00:47:44] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:51:42] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Ping timeout: 244 seconds)
[00:53:01] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)
[00:53:06] <Smithx10> wtf......
[00:53:09] <Smithx10> i hate computers.
[00:53:37] <Smithx10> just tried html 5 and the keyboards work in the grub menu
[00:53:43] <Smithx10> but not in VGA after they boot
[00:54:02] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:55:54] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Client Quit)
[00:57:21] <LeftWing> The after boot thing is probably the USB bug preventing us (in error) from getting to the keyboard
[00:58:19] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[00:59:01] <Smithx10> yea
[01:00:40] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[01:00:54] <Smithx10> and then reboot
[01:00:57] <jemershaw> Smithx10: way to go :p
[01:00:58] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Read error: Connection reset by peer)
[01:01:10] <Smithx10> So.... It seems.... ipmitool worked but was hung and required a power cycle
[01:01:13] <Smithx10> "INSANE"
[01:01:15] <Smithx10> AHHHHH
[01:01:20] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[01:05:38] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 250 seconds)
[01:07:42] <Smithx10> How do I set a vlan on a interface in the gz
[01:07:56] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has quit IRC (Quit: andy_js)
[01:14:08] <LeftWing> So you can't have a VLAN tag on the "admin" interface, but for other networks you can add a NIC to the CN -- e.g., using AdminUI
[01:14:30] <LeftWing> An IP will be assigned, and on a reboot the platform will bring an IP interface up on that network using the VLAN tag etc
[01:16:24] <rsully> that requires a reboot? o_O
[01:18:52] <LeftWing> In theory it needn't, but it's generally an uncommon operation (in our experience) so we haven't worked to make it a dynamic change
[01:19:03] <LeftWing> Most NICs are added to _zones_ not to the GZ
[01:21:03] <jesse_> LeftWing, I _just_ set up a nictag and a zone with a vlan_id on smartos, does sdc make it somehow harder?
[01:21:37] <bahamat> No, it should be easier on Triton.
[01:22:02] <jesse_> tried a nic tag with vlan_id but it didn't work
[01:22:13] <bahamat> In Triton you just define the vlan id on the network when it's created and anything on that network automatically is configured properly.
[01:22:20] <LeftWing> Smithx10 was specifically asking about a VNIC in the global zone, as far as I could tell
[01:22:29] <jesse_> which is a bit annoying, having to remember the vlan on zone create instead of nic tag add
[01:22:32] <LeftWing> Which is not the usual thing people want -- which is VNICs inside zones
[01:22:51] <LeftWing> The NIC tag abstraction is meant to identify equivalent NIC ports basically
[01:22:53] <LeftWing> like
[01:22:56] <jesse_> aaaah
[01:23:01] <LeftWing> "NICs with tag A can see VLAN 100, 101, 102"
[01:23:08] <LeftWing> NICs with tag B can see 200, 201, 202
[01:23:28] <jesse_> that doesn't work at all? the smartos way is to have interface per vlan?
[01:23:31] <LeftWing> You might put both A & B on the same NIC -- or maybe you do on some machines where the switch is configured that way, and not on some others where they get separate ports
[01:23:33] <Smithx10> Yea if IM in the GZ, can i do it with ifconfig?
[01:23:40] <Smithx10> I see i can plumb a vnic on that physical device
[01:23:46] <Smithx10> but i wanted to just apply it to the device
[01:24:11] <LeftWing> Smithx10: If you want to do a temporary thing, you could... dladm create-vnic -l $NIC -v $VLAN_ID -t whatever0
[01:24:13] <LeftWing> I think
[01:24:20] <Smithx10> I did
[01:24:22] <Smithx10> to test
[01:24:29] <Smithx10> but I was curious if I could just put ixgbe0 on vlan 5
[01:24:37] <LeftWing> Is ixgbe0 the admin NIC?
[01:24:40] <Smithx10> no
[01:24:48] <Smithx10> this is pre install
[01:24:51] <LeftWing> Ah OK
[01:25:01] <Smithx10> I was trying to test to make sure my networking team actually did their job
[01:25:04] <LeftWing> So if you want interfaces on other networks, with (or without) a VLAN ID
[01:25:04] <Smithx10> which of course..... they didnt
[01:25:15] <LeftWing> You would configure that through NAPI, AdminUI, etc
[01:25:49] <LeftWing> e.g., if you do it for "external" we'll create an "external0" VNIC in the GZ for you if you configure it to do so
[01:37:46] <jesse_> is there a way to limit a vnic to some bandwidth?
[01:48:38] <bahamat> jesse_: You can use flowadm for that
[01:53:37] <jesse_> related to dladm show-linkprop -z $UUID -p maxbw net0 ?
[01:55:00] <rzezeski> If setting maxbw it's worth keeping these two issues in mind: OS-6205 and OS-6194
[01:55:02] <jinni> https://smartos.org/bugview/OS-6194
[01:55:02] <jinni> https://smartos.org/bugview/OS-6205
[01:55:25] <rzezeski> I've had fixes for both of them for something like two years now but got put on the back burner when I tried writing programmatic tests for them
[01:56:07] <jesse_> I feel OS-6194 is somewhat confused, full duplex 300M means 300M both ways to me
[01:56:33] <rzezeski> basically, IIRC, you may allow more BW than your bargained for, and if you set it too low you will kill the link
[01:56:34] <jesse_> it never makes sense to add up up/down
[01:56:37] <rzezeski> so might want to test it
[01:57:33] <rzezeski> fair enough re full duplex
[01:57:54] <jesse_> but OS-6205 is really a problem
[01:58:01] <rzezeski> yes
[01:58:09] <jesse_> would need some fancy queueing and scheduling to make it stay under the limit
[01:58:11] <rzezeski> well, it's a problem that can be worked around
[01:58:20] <rzezeski> nothing too fancy, I have the fix
[01:58:27] <rzezeski> tested in manually
[01:58:36] <jesse_> allowance-based?
[01:58:39] <rzezeski> but we wanted to try to get some automated testing in, and that's where it got sidetracked two years ago
[02:01:46] <rzezeski> IIRC, I made sure the min setting is at least one MTU packet-per-second and as for max enforcement I convert the user's bits-per-second into bytes-per-tick and then with each tick add to the limit that can be sent that second, as long as the current usage counter + the size of the next packet is below that limit it can send, otherwise it needs to wait for the next tick to roll around, each second the counters are reset
[02:02:00] <rzezeski> whatever you would call that...it's a bit fuzzy since I did it two years ago ;)
[02:02:58] <Smithx10> T
[02:03:18] <Smithx10> rzezeski: what’s new?
[02:03:31] <rzezeski> Smithx10: nothing, i40e owns me
[02:03:38] <Smithx10> Saw you pushed those 2 i40e patches
[02:03:42] <Smithx10> :)
[02:03:46] <Smithx10> More coming ???
[02:03:53] <rzezeski> though I think I'm on the last freeze: OS-7492
[02:03:55] <jinni> https://smartos.org/bugview/OS-7492
[02:04:11] <rzezeski> have a patch that seems to work
[02:05:36] *** jellydonut <jellydonut!~jelly@77.18.49.241.tmi.telenormobil.no> has quit IRC (Read error: Connection reset by peer)
[02:05:40] <jesse_> rzezeski, how about a counter that gets bits per timeunit (check the time and accumulated bits when receiving/somethings send queue) (not sure when send would be triggered) and only send/receive if packet has less bits than accumulated
[02:06:09] <jesse_> your kind of sounds the same...
[02:06:16] <rzezeski> I think we just said the same thing
[02:06:26] <rzezeski> And I think "allowance based" is a decent way to describe what I did
[02:06:36] <rzezeski> You basically get bytes added to your allowance every tick
[02:06:42] <rzezeski> and this system resets each seond
[02:06:47] <jesse_> yeah, that's what I was after
[02:06:57] <rzezeski> sorry it's been a long time since this was in my brain
[02:07:02] <rzezeski> and I've got other things taking up space
[02:07:16] <jesse_> not sure about the reset... It might just have to have a maximum you can accumulate?
[02:08:13] <jesse_> I'd need clear head and couple of hours with a calculator to make sense of it
[02:08:20] <rzezeski> so the allowance makes sure you never exceed your limit per second (since that's the interface we give the user: per second) and the per-tick makes sure it's not bursted in one go but smoothed out. I guess you could argue the user should be able to burst it all at once. I dunno. But I know the patch held the bandwidth as it should.
[02:08:33] <jesse_> your solution sounds like it'll do it anyway, though=)
[02:08:36] <rzezeski> well the interface is per-second, so that's why I did it that way
[02:09:05] <jesse_> that makes sense
[02:09:08] <rzezeski> like if you only used half your allowance one second it doesn't carry over, like sick time :)
[02:09:14] <jesse_> just thinking latency-side
[02:09:16] <rzezeski> use it or lose it
[02:09:29] <jesse_> you send 1500 bytes 1ns after the clear
[02:09:49] <jesse_> (before clear the allowance was maxed)
[02:09:52] <rzezeski> right, so that's why I think I didn't give the allowance all up front, but add a portion with each tick
[02:10:21] <rzezeski> which maybe could be argued against if I put a code review up
[02:10:29] <rzezeski> but there may have been other reasons I had to do it that wqay
[02:12:03] <rzezeski> I guess it's six-in-one-half-a-dozen-in-another: if you give everything up front then you end up waiting the rest of the second, but if you give a slice per tick then you wait all the way to the end...in either case you have to wait because that's your allowance
[02:12:36] <jesse_> per-tick allowance accumulation is good
[02:13:02] <jesse_> but if it accumulates up to mtu you can never send data with quota from past
[02:13:07] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[02:13:16] <rzezeski> I think it probably is better at preventing unnecessary retrans
[02:13:53] <jesse_> (well, from 'more than mtu accumulation past'
[02:14:22] <jesse_> gives low latency for rarely sending parties and caps the bandwidth for bulk transfers
[02:15:24] <jesse_> anyway... I just wanted to limit a vnic down a bit=)
[02:17:58] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 245 seconds)
[02:18:18] <rzezeski> yea, the current implementation has sharp edges but can probably be used to keep things under control
[02:25:26] <jesse_> uh, maxbw=100 really takes a hit when you meant to type maxbw=100M
[02:26:08] <jesse_> uh, it should have been the same according to the manpage, interesting
[02:26:37] <_Tenchi_> haha
[02:26:51] *** andy_js <andy_js!~andy@50-206-180-2-static.hfc.comcastbusiness.net> has joined #smartos
[02:27:17] <jesse_> ...there may be a bug there=)
[02:27:48] *** v_a_b <v_a_b!~volker@p57A27D60.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 244 seconds)
[02:32:51] <Smithx10> rzezeski: The 700-series parts have a Tx descriptor limit of 8 for a single packet (or segment of an TSO packet).
[02:33:03] <Smithx10> Does x520 not have this hard limit ?
[02:37:22] <Smithx10> I see they handle it via a pull-up... but does the txd limit have any implications ?
[02:37:44] <Smithx10> I’m wondering if x520 is faster than x710 at the same speed
[02:38:49] <rzezeski> Smithx10: x520 also has a limit, but it's 32 I believe
[02:39:19] <rzezeski> this is the reason behind all that goofy logic in ixgbe_ring_tx
[02:39:53] <Smithx10> Are they almost at parity ?
[02:39:59] <Smithx10> On SmartOS ?
[02:40:12] <rzezeski> I think the txd limit shouldn't really play into throughput, no
[02:40:39] <rzezeski> x520 has more support, e.g. it has the hardware VLAN filtering
[02:40:48] <rzezeski> i40e still doesn't have that...I have incomplete patches
[02:40:52] <rzezeski> my queue is long
[02:40:56] <Smithx10> Hahahaha
[02:41:02] <Smithx10> Keep grinding :)
[02:41:09] <Smithx10> <3 it
[02:41:25] <rzezeski> For now, ixgbe is the best to stick with...but i40e isn't too far behind at this point
[02:41:33] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)
[02:41:34] <Smithx10> Cool
[02:41:50] <rzezeski> Though I vastly prefer debugging ixgbe
[02:41:51] <Smithx10> We have no crushing need for more than 10 go for compute atm
[02:41:59] <Smithx10> Also with ppt
[02:42:30] <rzezeski> yea and I've only used X710/X722 parts, no XXV/XL parts
[02:42:41] <Smithx10> I can just give the crazy guys a mellanox card when they request that
[02:43:07] <rzezeski> as for 25/100G, I'm most interested in hopping over to chelsio for a bit
[02:43:14] <rzezeski> but we'll see
[02:43:22] <Smithx10> X700 is 40 right ?
[02:43:29] <rzezeski> XL iss
[02:43:34] <rzezeski> the letters denote the speed
[02:43:39] <rzezeski> roman numerals
[02:43:49] <Smithx10> Ahhh sorry
[02:43:54] <Smithx10> I’m stupid
[02:44:27] <rzezeski> I just figured that out this summer
[02:44:33] <Smithx10> Hahahaha
[02:44:55] <Smithx10> Make LeftWing help
[02:45:20] <Smithx10> I hear he just sits in zfs meetings and proclaims his hatred for Linux :P
[02:46:34] <rzezeski> alright I gotta go eat
[02:47:32] <Smithx10> Later, have a good night
[02:51:00] *** andy_js <andy_js!~andy@50-206-180-2-static.hfc.comcastbusiness.net> has quit IRC (Quit: andy_js)
[02:51:57] <Smithx10> LeftWing: j/k you reeled them in :)
[02:53:22] *** BrownBear <BrownBear!~BrownBear@107.175.247.247> has joined #smartos
[02:55:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Ping timeout: 268 seconds)
[02:58:48] <Smithx10> jemershaw: thanks for helping me learn how to use iLo
[02:58:54] <Smithx10> I feel like a real boy now
[03:10:11] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has quit IRC (Remote host closed the connection)
[03:32:29] *** vaxsquid <vaxsquid!~vaxsquid@104.161.79.35> has joined #smartos
[03:53:43] *** theandro1edan <theandro1edan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has joined #smartos
[03:55:56] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has quit IRC (Ping timeout: 250 seconds)
[04:14:09] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[04:18:53] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 244 seconds)
[04:22:56] *** papertigers <papertigers!~papertige@pool-108-17-31-72.bflony.fios.verizon.net> has quit IRC (Remote host closed the connection)
[04:49:09] *** jubal_ <jubal_!~jubal@64-16-237-24.gci.net> has joined #smartos
[04:52:50] *** jubal <jubal!~jubal@247-16-237-24.gci.net> has quit IRC (Ping timeout: 272 seconds)
[04:53:50] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has joined #smartos
[05:54:25] <Smithx10> Any Special trick to running lldpd on a interface besides lldpd -i interface?
[05:56:46] <Smithx10> can you run lldpd on a vnic?
[05:59:12] <jbk> you can, but i'm not sure what good it'd do
[06:00:33] <jbk> well at least in theory -- dlpi_open should work on a vnic
[06:00:35] <Smithx10> Im trying to just see if anything else on a vlan tag
[06:01:04] <jbk> what are you trying to determine?
[06:01:17] <Smithx10> its been awhile since ive used cdp / lldp
[06:01:28] <Smithx10> but i thought it just advertised the neighbors on your switch
[06:01:48] <jbk> well the mac address it uses is not supposed to be forwarded beyond a bridge
[06:01:50] <jbk> IIRC
[06:02:26] <Smithx10> yea, the networking guys have just totally botched the config
[06:02:31] <Smithx10> ill just stop poking it
[06:04:02] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)
[06:10:44] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has quit IRC (Remote host closed the connection)
[06:11:23] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos
[06:15:18] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[06:17:59] *** jubal_ <jubal_!~jubal@64-16-237-24.gci.net> has quit IRC (Ping timeout: 268 seconds)
[06:20:14] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 250 seconds)
[06:21:33] *** jubal <jubal!~jubal@64-16-237-24.gci.net> has joined #smartos
[06:32:23] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has joined #smartos
[06:37:01] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[06:39:52] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)
[06:40:57] *** Gathis <Gathis!~TheBlack@unaffiliated/gathis> has quit IRC (Remote host closed the connection)
[06:41:34] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 246 seconds)
[06:52:18] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[06:52:32] *** fejfighter <fejfighter!~fejfighte@124-171-205-37.dyn.iinet.net.au> has joined #smartos
[07:11:47] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[07:15:42] *** olsner <olsner!~salparot@c83-249-185-75.bredband.comhem.se> has quit IRC (Ping timeout: 250 seconds)
[07:40:19] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[07:44:47] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 240 seconds)
[07:55:04] *** neuroserve <neuroserve!~toens@195.71.113.124> has joined #smartos
[07:57:23] *** _Tenchi_ <_Tenchi_!~phil@207-255-80-203-dhcp.aoo.pa.atlanticbb.net> has quit IRC (Ping timeout: 246 seconds)
[08:04:48] *** _Tenchi_ <_Tenchi_!~phil@207-255-80-203-dhcp.aoo.pa.atlanticbb.net> has joined #smartos
[08:05:13] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[08:09:50] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 244 seconds)
[08:36:49] *** JeanParpaillon <JeanParpaillon!~jean@176.164.125.94> has joined #smartos
[08:51:24] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[08:54:48] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)
[08:56:07] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 240 seconds)
[09:06:26] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has joined #smartos
[09:09:14] *** jubal <jubal!~jubal@64-16-237-24.gci.net> has quit IRC (Ping timeout: 250 seconds)
[09:19:06] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has quit IRC (Quit: Leaving.)
[09:21:42] *** olsner <olsner!~salparot@c83-249-185-75.bredband.comhem.se> has joined #smartos
[09:23:43] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has quit IRC (Quit: black_pete)
[09:27:35] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has joined #smartos
[09:27:45] *** v_a_b <v_a_b!~volker@p57a27d4f.dip0.t-ipconnect.de> has joined #smartos
[09:31:49] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has joined #smartos
[09:32:26] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[09:34:38] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 245 seconds)
[09:35:30] *** jellydonut <jellydonut!~jelly@77.16.74.235.tmi.telenormobil.no> has joined #smartos
[09:37:03] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 252 seconds)
[09:43:51] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[09:48:41] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 246 seconds)
[10:28:06] *** man_u <man_u!~manu@manu2.gandi.net> has joined #smartos
[10:28:57] *** insane^ <insane^!~insane@fw.vispiron.de> has joined #smartos
[11:17:49] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[11:22:27] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 240 seconds)
[11:31:28] *** jellydonut <jellydonut!~jelly@77.16.74.235.tmi.telenormobil.no> has quit IRC (Read error: Connection reset by peer)
[11:54:47] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[11:59:43] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 252 seconds)
[12:02:09] *** strobelight <strobelight!~quassel@2001:420:c0c8:1004::a5> has quit IRC (Read error: Connection reset by peer)
[12:12:56] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[12:16:30] *** jellydonut <jellydonut!~jelly@77.16.74.235.tmi.telenormobil.no> has joined #smartos
[12:17:47] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 246 seconds)
[12:40:18] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[12:44:52] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 246 seconds)
[12:58:11] *** jellydonut <jellydonut!~jelly@77.16.74.235.tmi.telenormobil.no> has quit IRC (Read error: Connection reset by peer)
[13:00:20] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[13:05:34] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 272 seconds)
[13:07:39] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[13:12:20] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 250 seconds)
[13:22:52] *** strobelight <strobelight!quassel@nat/cisco/x-kadhsdodhnxkcnay> has joined #smartos
[13:36:17] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[13:40:52] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 246 seconds)
[14:06:59] <neuroserve> do we have cdpr for smartos?
[14:27:56] <Smithx10> cluster zfs.... ? https://twitter.com/_thnet/status/1083252321885265920
[14:31:27] <insane^> https://blogs.oracle.com/solaris/cluster-file-system-with-zfs-introduction-and-configuration
[14:31:33] <insane^> support for zfs as a failover...
[14:31:40] <insane^> oh
[14:31:44] <insane^> no its multihost
[14:31:59] <insane^> It must be noted that zpool for globally mounted ZFS file systems does not actually mean a global ZFS pool, instead there is a Cluster File System layer that is present on top of ZFS that makes the file systems of the ZFS pool globally accessible.
[14:32:08] <insane^> and layered fobar
[14:39:40] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos
[14:58:08] <Smithx10> ahhhh so magic on top of zfs
[14:58:13] <Smithx10> not zfs magic
[15:01:42] <neuroserve> trying to remember what the last clusterfs was, I had to deal with ...
[15:02:30] *** jellydonut <jellydonut!~jelly@77.16.74.235.tmi.telenormobil.no> has joined #smartos
[15:20:40] *** andy_js <andy_js!~andy@static-108-31-135-162.washdc.fios.verizon.net> has joined #smartos
[15:37:01] *** miller7 <miller7!~miller7@unaffiliated/miller7> has quit IRC (Read error: Connection reset by peer)
[15:37:21] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[15:37:37] *** ingenthr <ingenthr!~ingenthr@47.150.244.29> has joined #smartos
[15:43:28] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has quit IRC (Ping timeout: 244 seconds)
[15:49:04] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos
[15:50:56] *** lystra <lystra!~lystra@d53-64-11-169.nap.wideopenwest.com> has quit IRC (Remote host closed the connection)
[15:53:50] <Smithx10> neuroserve: just alias it with tirefire
[16:04:18] *** kkantor <kkantor!~kkantor@c-24-118-59-107.hsd1.mn.comcast.net> has joined #smartos
[16:06:04] *** mzf <mzf!~fakhraee@unaffiliated/mzf> has quit IRC (Ping timeout: 272 seconds)
[16:13:57] *** Teknix <Teknix!~pds@2601:801:101:f429:f538:f4d7:c565:a3b3> has quit IRC (Ping timeout: 250 seconds)
[16:23:45] *** black_pete <black_pete!~black_pet@gateway.peterguy.com> has joined #smartos
[16:23:48] *** insane^ <insane^!~insane@fw.vispiron.de> has quit IRC (Ping timeout: 272 seconds)
[16:30:25] *** Teknix <Teknix!~pds@2601:801:101:f429:69b8:e67d:5d12:894c> has joined #smartos
[16:36:37] *** polishdub <polishdub!~polishdub@207.86.38.254> has joined #smartos
[16:40:14] *** theandro1edan <theandro1edan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has quit IRC (Read error: Connection reset by peer)
[16:43:33] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has joined #smartos
[16:48:14] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has joined #smartos
[16:55:00] *** neuroserve <neuroserve!~toens@195.71.113.124> has quit IRC (Ping timeout: 268 seconds)
[16:56:36] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)
[17:20:04] *** JeanParpaillon <JeanParpaillon!~jean@176.164.125.94> has quit IRC (Ping timeout: 252 seconds)
[17:41:07] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has quit IRC (Remote host closed the connection)
[17:41:45] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has joined #smartos
[18:02:56] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos
[18:04:44] <greenline> I've got a rather perplexing situation. My triton headnode is not booting unless I add console=vga to the kernel parameters at boot time. If I don't, the kernel starts, but then prints a warning, and never proceeds with startup (no USB file copy, no SMF init).
[18:04:51] <greenline> WARNING: /pci@0,0/pci1028,235@1d,7/storage@1/disk@0,0 (sd0):
[18:04:51] <greenline> Command failed to complete...Device is gone
[18:05:28] <greenline> Although I believe that warning is not related.
[18:06:29] <greenline> actually, maybe the VGA thing is a spurious correlation
[18:06:52] <greenline> Are there good debug options for start-up?
[18:07:39] <jlevon> you're definitely on the right serial console?
[18:08:04] <greenline> yes, I can see it print the SunOS banner, etc, and when it boots up normally, I get the ttyb login prompt
[18:08:18] <greenline> the system is not starting, not responding to ping or anything in this condition
[18:09:26] <jlevon> you can get a lot more detail with -m verbose -v -B prom_debug=true
[18:10:34] <greenline> kk
[18:17:09] <greenline> ah, this is starting to look like the USB key is failing.
[18:19:23] <greenline> if that were the case, is there a procedure for replacing the USB drive?
[18:21:26] <jlevon> https://www.mail-archive.com/smartos-discuss at lists dot smartos.org/msg05977.html
[18:22:04] <jlevon> BTW you're not intentionally named after the project codename for SMF are you :P
[18:25:02] <greenline> no, never heard of that.
[18:25:07] <greenline> funny coincidence
[18:25:17] <greenline> Was SMF like systemd in terms of stirring controversy? ;-)
[18:26:14] <jlevon> not quite that much
[18:26:26] <jlevon> but it was rough for sure
[18:27:17] <jbk> though it seems to have shown at least a bit more restraint than systemd
[18:29:24] <jlevon> oh sure, putting an SQL DB right in essential system boot is very restrained :P
[18:29:52] <jbk> relatively speaking, yes :)
[18:30:13] <jbk> smf hasn't evolved it's own NTP client, dns resolver, custom binary log format, ...
[18:30:32] <jbk> not to say there aren't things that couldn't have been done better
[18:30:34] <greenline> jlevon: I see the link, but isn't there some data about my env stored on the USB drive? Like NTP, DNS, etc. etc.?
[18:30:55] <jbk> but in comparsion, i'd take it every day of the week and twice on sundays over systemd
[18:33:06] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[18:34:29] *** HBorman <HBorman!~oleg@87.99.79.115> has quit IRC (Ping timeout: 244 seconds)
[18:36:48] *** man_u <man_u!~manu@manu2.gandi.net> has quit IRC (Ping timeout: 272 seconds)
[18:38:31] *** bens1 <bens1!~bens1@161.122.2.81.in-addr.arpa> has quit IRC (Quit: bens1)
[18:47:05] *** greenline <greenline!~user@unaffiliated/greenline> has quit IRC (Remote host closed the connection)
[18:51:16] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has joined #smartos
[18:52:54] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has quit IRC (Quit: My MacBook has gone to sleep. ZZZzzz…)
[18:54:14] *** Kurlon <Kurlon!~Kurlon@bidd-pub-03.gwi.net> has quit IRC (Ping timeout: 250 seconds)
[18:59:08] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has joined #smartos
[19:05:37] *** greenline <greenline!~user@unaffiliated/greenline> has joined #smartos
[19:06:29] <Smithx10> to confirm
[19:06:39] <Smithx10> f8:f2:1e:3b:b:35 that is 0b correct ?
[19:08:18] <pmooney> believe so
[19:09:26] <rmustacc> Yes
[19:11:01] <Smithx10> Thanks :)
[19:11:51] <Smithx10> rmustacc: to confirm.... the onboard nics on the dl380 g10 looked like they were discovered
[19:12:18] <Smithx10> https://gist.github.com/Smithx10/b9b5d0aa0bbcc57ceb4d132665c61465
[19:12:29] <Smithx10> they are using the bge driver
[19:14:27] *** chorrell <chorrell!~chorrell@unaffiliated/chorrell> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[19:15:16] <rmustacc> I'd expect them to probably work.
[19:34:18] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has quit IRC (Quit: Leaving)
[19:37:15] *** mno-hime <mno-hime!~mno-hime@94.142.238.232> has joined #smartos
[20:07:02] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has joined #smartos
[20:26:27] *** axonpoet <axonpoet!~axonpoet@fsf/member/axonpoet> has joined #smartos
[20:33:24] *** JeanParpaillon <JeanParpaillon!~jean@176.164.107.140> has joined #smartos
[20:36:26] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has quit IRC (Read error: Connection reset by peer)
[20:36:53] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Remote host closed the connection)
[20:37:53] *** lgtaube <lgtaube!~lgt@46.165.251.115> has joined #smartos
[20:38:00] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos
[20:38:19] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has joined #smartos
[20:48:07] <Smithx10> During SmartOS setup.... the CLI i drop to
[20:48:15] <Smithx10> it should be full fledged smartos and be able to route traffic etc correct?
[20:48:29] <Smithx10> lololol, sorry my networking guy is trying to set something up
[20:48:36] <Smithx10> and im just 0240204204 checking
[20:48:41] <rmustacc> Well, no networking is configured...
[21:10:53] <LeftWing> But you can configure things
[21:11:04] <LeftWing> Using the manual OS tools and such
[21:11:20] *** blackwood821 <blackwood821!ae3110b0@gateway/web/freenode/ip.174.49.16.176> has joined #smartos
[21:11:31] <blackwood821> is there a way to delete a route from a zone via `vmadm update`?
[21:12:25] <LeftWing> blackwood821: I would look for "remove_routes" in the vmadm(1M) manual page.
[21:12:29] <rmustacc> blackwood821: If you look at the vmadm manual page search for 'remove_routes'
[21:12:36] <LeftWing> JYNX
[21:12:40] <rmustacc> d'oh
[21:13:46] <blackwood821> ok thanks
[21:14:33] *** JeanParpaillon <JeanParpaillon!~jean@176.164.107.140> has quit IRC (Quit: JeanParpaillon)
[21:17:46] *** moatsad <moatsad!~dougt@69.194.139.196> has quit IRC (Quit: Leaving.)
[21:27:04] <Smithx10> anyone a screen ninja?
[21:27:09] <Smithx10> when I do the screen verticle split
[21:27:13] <Smithx10> I didnt get a shell
[21:27:22] *** JeanParpaillon <JeanParpaillon!~jean@176.164.107.140> has joined #smartos
[21:27:24] <Smithx10> http://aperiodic.net/screen/quick_reference
[21:27:32] <Smithx10> trying to ipmi in, and snoop + ping etc..
[21:28:00] *** JeanParpaillon <JeanParpaillon!~jean@176.164.107.140> has quit IRC (Client Quit)
[21:28:29] <LeftWing> Smithx10: I think you might need to start another ... whatever they call the commands that run in windows
[21:28:38] <LeftWing> (I've been using tmux for too long)
[21:28:41] <Smithx10> hahaha me too
[21:28:52] <LeftWing> I seem to remember pressing ^A and then "c" might do something
[21:28:54] <Smithx10> I did split display vertically C-a | or C-a V (for the vanilla vertical screen patch
[21:29:00] <Smithx10> and then C-a tab
[21:29:08] <rsully> Do I need to worry about compatibility of a 10Gb nic built into a supermicro motherboard with smartos?
[21:29:45] <rmustacc> rsully: In general, no.
[21:30:06] <rmustacc> rsully: Most of them are Intel nics which we should support reasonably well.
[21:30:33] <rsully> gotcha, in this case it says "Dual LAN with 10GBase-T with Intel® X722 + X557"
[21:30:37] <rsully> is there a list of compatible parts?
[21:30:43] <rmustacc> The X722 is supported.
[21:30:59] <rmustacc> There have been one or two driver bugs we've seen on that that we're still wrapping up though.
[21:31:27] <rsully> edge cases or normal use?
[21:32:17] <rmustacc> I think one last edge case that rpz is working on
[21:33:39] <rsully> if it matters I'm only connecting this to 1Gb switches
[21:35:21] <rsully> are those fixes expected to be done sometime in the next month or so?
[21:38:06] *** scarcry_ <scarcry_!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has joined #smartos
[21:41:57] <Smithx10> Does the GZ have any firewalling for ICMP ?
[21:42:06] <Smithx10> that I have to turn off for testing .... just double checking I believe no
[21:42:27] <Smithx10> fwadm list shows empty so
[21:42:41] <Smithx10> should be good...... I dont have root to these switches.... 2.5 hrs into a call.... OMG
[21:42:45] <Smithx10> ./starts to cry.
[21:43:11] *** scarcry_ is now known as scarcy
[21:43:25] *** scarcy <scarcy!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has quit IRC (Quit: use Unix or die.)
[21:44:37] <melloc1> The GZ firewall is independent of the fwadm. You can see whether there are any rules by running `ipfstat -nio` in the GZ.
[21:45:56] *** scarcry_ <scarcry_!~scarcry@2001:980:93d7:1:80c9:7fff:fe0f:aaf8> has joined #smartos
[21:52:24] *** scarcry is now known as Guest14587
[21:52:24] *** scarcry_ is now known as scarcry
[21:52:49] *** lystra <lystra!~lystra@d53-64-11-169.nap.wideopenwest.com> has joined #smartos
[21:57:56] <greenline> ok, so I made a new USB drive, put the existing config onboard, booted up, all is well. ran all upgrades again for good measure
[21:58:14] <greenline> now the only concern is that the default option at GRUB on the USB drive was PXE for a compute node, but this is the head node.
[21:59:31] <greenline> oh, is it supposed to boot from hdd on the headnode?
[22:17:14] <bahamat> No, it's always supposed to boot from the usb.
[22:17:41] *** npx <npx!~alex@136.41.160.10> has joined #smartos
[22:17:56] <npx> Are there any integrations between UFDS and Windows?
[22:18:11] <bahamat> npx: Not intentionally.
[22:18:21] <npx> I can only imagine what sort of evil chicanery would be entailed by that
[22:18:33] <npx> but it seems like the sort of thing you might pay a grown up to do
[22:19:26] <bahamat> UFDS doesn't have support for any external back ends.
[22:19:48] <bahamat> You can't just point Triton at AD.
[22:20:18] <npx> I was imagining what it'd be like if each account in triton had an implicit "activedirectory domain" and windows instances automatically joined it
[22:20:24] *** blackwood821 <blackwood821!ae3110b0@gateway/web/freenode/ip.174.49.16.176> has quit IRC (Quit: Page closed)
[22:20:40] <greenline> So I found in the USB drive I can edit the grub menu.lst and menu.lst.tmpl, but is there some more correct way to get it to boot the platform image by default?
[22:20:40] <npx> Basically, I'm trying to imagine how to do Windows on Triton with a minimal amount of suck but there is just a high level of implicit suck
[22:21:04] <bahamat> greenline: Set `default 1`, if it's not already.
[22:21:25] <greenline> bahamat: as in, just edit the files?
[22:21:30] <greenline> or is there an sdcadm command for that?
[22:21:35] <bahamat> npx: What do you mean by "windows on triton"? You can run windows images.
[22:21:46] <bahamat> greenline: Just edit the file. There's no sdcadm command for it.
[22:22:13] <npx> Sure, but I meant managing some arbitrary number of windows images in a way that would be pleasing to a windows admin (and I admittedly am not sure what that might be)
[22:22:53] <bahamat> You can create new images from running instances.
[22:23:03] <greenline> bahamat: thanks
[22:23:17] <bahamat> So boot up a base instance, prepare it as you like, then CreateImagefromMachine
[22:23:20] <npx> Like, they tend to have windows hosts join an AD domain and a bunch of policies etc are set via that mechanism
[22:24:14] <bahamat> npx: I think maybe you're looking for the Triton guest tools? https://github.com/joyent/sdc-vmtools-windows
[22:24:57] <bahamat> Triton's UFDS is not intended to be exposed to tenant instances.
[22:25:13] <npx> that's useful, I can create images but I was thinking more about managing an aggregation of running instances in a cohesive way much like what you'd do with windows on physical hardware but I see your point
[22:25:25] <npx> (I should really use bhyve though)
[22:26:13] <bahamat> I don't see how that would be any different on Triton than any other type of virtualization infrastructure.
[22:26:23] <npx> I guess you'd just use actual activedirectory, yeah
[22:26:53] <bahamat> Yeah. Again, tenant instances should not have access to UFDS under any circumstances.
[22:27:08] <bahamat> That pretty much precludes UFDS being used as an AD back end.
[22:27:33] <npx> Gotcha, even if it worked I can see how it'd be marginal
[22:27:51] <bahamat> It would be better to set up your AD controller instances and then craft a custom image that will automatically join that domain.
[22:28:09] <npx> roger that
[22:28:50] <bahamat> For example, we use LDAP for user accounts, but we have ldap server instances that are completely separate from UFDS.
[22:38:52] <npx> it would admittedly be interesting if a Triton headnode could also PXE boot Windows Server 2019 and automatically provide AD domain manager services to actual physical nodes, it's kinda dystopian to imagine actually doing it but interesting
[22:47:06] *** polishdub <polishdub!~polishdub@207.86.38.254> has quit IRC (Quit: leaving)
[23:02:02] *** rsully <rsully!~rsully@unaffiliated/rsully> has quit IRC (Quit: rsully)
[23:02:32] *** npx <npx!~alex@136.41.160.10> has quit IRC (Ping timeout: 250 seconds)
[23:05:09] *** gh34 <gh34!~textual@rrcs-162-155-144-114.central.biz.rr.com> has quit IRC (Quit: Textual IRC Client: www.textualapp.com)
[23:18:39] <greenline> I have found that some triton service VMs are not following the order of nameservers in /etc/resolv.conf
[23:19:02] <greenline> First entry is up and working (the binder0) entry, but queries go to the second server listed by default
[23:32:43] *** strobelight <strobelight!quassel@nat/cisco/x-kadhsdodhnxkcnay> has quit IRC (Ping timeout: 246 seconds)
[23:36:19] <rmustacc> greenline: For requesting what names?
[23:36:44] <rmustacc> greenline: The binder resolvers explicitly issue a servfail if they find a name that they don't know for split DNS purposes
[23:38:50] <greenline> rmustacc: requesting internal service names, like moray.datacenter.example.com
[23:39:50] <greenline> rmustacc: the binder0 is the first nameserver in resolv.conf, but nameserver queries are going to the second nameserver (ours) by default. this is in the adminui vm, which is multi-homed
[23:40:31] <greenline> If I direct queries directly to the binder0 instance with nslookup, it works as expected. but unqualified nslookups, host, etc, go to the 2nd nameserver by default
[23:59:07] *** Kurlon_ <Kurlon_!~Kurlon@98.13.72.207> has quit IRC (Remote host closed the connection)
top

   January 10, 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 | >