January 26, 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 26, 2019 [00:00:17] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[00:00:56] *** jubal_ <jubal_!~jubal@122-4-237-24.gci.net> has joined #smartos[00:02:53] <rmustacc> sjorge: There was a change upstream here. I thought a heads up had been sent out, but I'm having trouble finding it. If you rerun ./configure, you'll get sorted.[00:03:10] *** blackwood821 <blackwood821!ae3110b0@gateway/web/freenode/ip.174.49.16.176> has quit IRC (Quit: Page closed)[00:07:06] <sjorge> Ah :) I'll do hat and try again[00:07:33] <sjorge> I didn't know it was safe to run ./configure multiple times[00:07:43] <sjorge> I usually delete the entire zone and start again[00:08:16] *** jubal_ <jubal_!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[00:11:38] <rmustacc> sjorge: Oh, it certainly is.[00:11:48] <rmustacc> sjorge: Only real gotcha with ./configure is it will reset non-default branches.[00:12:20] <rmustacc> But my general, rule of thumb for no-fail, is gmake clobber; gmake update; ./configure, that should hopefully get someone past almost any flag day.[00:12:44] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has joined #smartos[00:15:58] *** Kurlon <Kurlon!~Kurlon@98.13.72.207> has joined #smartos[00:25:12] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[00:26:28] *** jubal_ <jubal_!~jubal@122-4-237-24.gci.net> has joined #smartos[00:42:05] <andyf> Are all of the fixes in https://cr.joyent.us/#/c/5425/ related, or just linked to base-64+gcc7?[00:42:17] <andyf> Just wondering as it's unusual to see different issues being conflated together into one commit[00:44:15] <rmustacc> andyf: They were all work that was found and tested together.[00:44:45] <rmustacc> The non-tools pieces are all related to gcc7.[00:44:57] <rmustacc> I'll be getting that upstream once we wrap up review.[00:45:14] <andyf> I just read through OS-7511... wow[00:45:17] <jinni> https://smartos.org/bugview/OS-7511[00:45:49] <rmustacc> Bryan went through and did a pretty big analysis of the issue to see what other code was impacted by this.[00:47:59] <andyf> I suppose if your code base is big and complicated enough, you're going to trigger edge cases[00:48:30] <andyf> but one does not expect a compiler to start removing loops[00:48:46] <jlevon> well that's not quite true.[00:48:52] <andyf> (well, not ones that are necessary)[00:48:59] <jlevon> but doing so based on these without even warning. that sucks[00:49:13] <jlevon> I don't hold out much hope for https://gcc.gnu.org/bugzilla/show_bug.cgi?id=88867 but one has to try[00:49:19] <Smithx10> Lololololololol[00:50:41] <andyf> not the most promising response so far..[00:50:58] <Smithx10> I guess the answer to all problems is[00:51:02] <jlevon> indeed[00:51:12] <Smithx10> Rewrite it gpl, or use Linux[00:51:14] <Smithx10> Hahahahahaha[00:51:30] <Smithx10> What a nightmare[00:51:52] <Smithx10> ./walks away slowly[00:54:03] <andyf> rmustacc - thanks, I can see how most of them are related now.. nice catches by Bryan[00:57:58] <rmustacc> andyf: There's really no great way on all of it, but I basically just work with what bmc gives me.[00:58:58] *** andy_js <andy_js!~andy@94.3.60.133> has quit IRC (Quit: andy_js)[01:08:19] <akole> rmustacc: disable-bnxe worked wonderfully and all the rest of the hardware seemed to detect correctly... except that horse-neighing G10 Smart Array. I'd imagine it's no use waiting for a driver and better to just start looking for LSI replacement[01:12:41] *** thebug <thebug!~thebug@unaffiliated/thebug> has quit IRC (Quit: leaving)[01:26:39] *** JtH <JtH!~5CE0A708@unaffiliated/cmdlnkid> has joined #smartos[01:34:15] <rmustacc> akole: Unfortnately, yes.[01:34:34] <rmustacc> akole: If we had a platform that should fix the bnxe issues, would that be helpful?[01:37:03] <akole> absolutely! and I'm happy to run more tests on the hardware if required[01:37:42] <rmustacc> What would the best format for that be?[01:37:53] <rmustacc> An updated platform tgz, a smartos USB image, something else?[01:38:17] <akole> let's say that our university is very interested on deploying a triton-based cloud so I'm willing to do a lot of work in getting this stuff working. Sadly due contract reasons we are now stuck with this HP hardware...[01:38:46] <akole> mm... I think if it'd be possible to get USB image it'd be the easiest to deploy[01:39:06] <rmustacc> OK, gotcha.[01:39:17] <rmustacc> Well, certainly we'll happy to help with issues as they come up.[01:40:07] <akole> wonderful :)[01:51:02] *** nfg243 <nfg243!~lounge-us@159.65.234.236> has quit IRC (Quit: The Lounge - https://thelounge.github.io)[02:04:11] *** jubal_ <jubal_!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[02:04:58] *** gotwf <gotwf!~gotz@unaffiliated/gotwf> has joined #smartos[02:11:28] <Smithx10> I found a bug in the public key code automation for adding to instances[02:12:12] <Smithx10> if the public key contains "" it seems to fail to create a authorized keys file[02:12:13] <Smithx10> https://paste.ec/paste/kKDHIe0l#HAD6N8kkWQ4bEz21Sbkdtq1bsPUJ+kbsag3bWOHl5ak[02:12:25] <Smithx10> example of a public key that makes the image provisioning not work[02:12:35] <Smithx10> jbk: that was the key that brought me to my knees :([02:12:55] <bahamat> Because the key comment is quoted?[02:13:18] <bahamat> Smithx10: Have you tried this on multiple zone brands?[02:13:30] <Smithx10> Just the latest CentOS VM image[02:14:25] <Smithx10> I don't know if the authorized keys are done through cloud-init now[02:14:28] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has joined #smartos[02:14:32] <Smithx10> or use /lib/smartdc[02:15:06] <jperkin> it's pretty unusual for it to contain quotes, you have to go out of your way to do that[02:15:32] <Smithx10> jperkin: yea, I agree, I told the user to stop being weird[02:15:41] <Smithx10> buttttt......... if he can do it..... others will :P[02:15:48] <jperkin> not saying we shouldn't handle quoting correctly, but it's more of a fuzzing thing than something people will run into on a daily basis[02:16:17] <bahamat> Smithx10: So what happens? It just fails to populate the authorized_keys file, or something worse?[02:16:19] <Smithx10> for sure, not something I'd go to war over[02:16:26] <Smithx10> just fails to populate[02:16:36] <Smithx10> i zlogin -C'd in and checked[02:16:42] <Smithx10> there was no authorized_keys[02:16:58] <Smithx10> strange thing tho....... i executed the /lib/smartdc/root-keys or whatever it is called[02:17:03] <Smithx10> and it populated the authorized keys[02:17:19] <Smithx10> which makes me think that for CentOS we are using cloud-init[02:18:27] <Smithx10> bahamat: do you know if docker resize is going to be fixed for the next release ?[02:18:39] <bahamat> I don't[02:18:44] <Smithx10> kk[02:19:11] <Smithx10> Guessing that isn't in the support chnanel?[02:19:37] <jesse_> |sed s/"/\\"/ ?=)[02:20:48] <Smithx10> lolololol[02:21:03] <Smithx10> like a bosssssss[02:21:20] <bahamat> Smithx10: Is the key properly formatted if you do `triton key get`?[02:22:12] <jesse_> Smithx10, if it fails its most likely because of json, so....=)[02:22:14] <Smithx10> I'll have to check on monday, I'm on my phone in the back seat of a car :)[02:22:36] <Smithx10> driving off the edge of flat earth :P[02:22:56] <jesse_> that's an excellent place to be[02:26:09] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[02:29:21] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has joined #smartos[02:37:29] <Smithx10> im bonding with frank castle[02:38:50] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[02:43:45] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has joined #smartos[02:53:20] <Smithx10> bahamat: looks like it didnt work[02:53:26] <Smithx10> i tried a few attempts[02:54:20] <Smithx10> actually, im stupid.... it looks proper in triton key get[02:54:37] <Smithx10> didnt realize that arg1 had to be the fingerprint[02:57:55] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has joined #smartos[02:58:05] *** v_a_b <v_a_b!~volker@p50905AC2.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 268 seconds)[02:59:00] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has quit IRC (Ping timeout: 246 seconds)[03:03:28] *** JtH <JtH!~5CE0A708@unaffiliated/cmdlnkid> has quit IRC ()[03:20:00] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has quit IRC (Quit: jcea)[03:26:05] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has joined #smartos[03:27:16] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has quit IRC (Read error: Connection reset by peer)[03:40:02] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has joined #smartos[04:35:39] *** jubal <jubal!~jubal@122-4-237-24.gci.net> has joined #smartos[04:37:07] *** jubal_ <jubal_!~jubal@247-16-237-24.gci.net> has quit IRC (Ping timeout: 240 seconds)[05:49:43] <NginUS> Can anyone recommend a zfs tutorial? Are some better than others, or is the top of a google search just as well as any?[06:09:52] *** axonpoet <axonpoet!~axonpoet@fsf/member/axonpoet> has quit IRC (Ping timeout: 268 seconds)[06:10:53] *** BrownBear <BrownBear!~BrownBear@107.175.247.247> has quit IRC (Ping timeout: 245 seconds)[06:12:20] *** fxhp <fxhp!~fox@138.207.192.33> has quit IRC (Quit: WeeChat 1.9.1)[06:16:40] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has quit IRC (Ping timeout: 250 seconds)[06:33:12] *** miller7 <miller7!~miller7@unaffiliated/miller7> has joined #smartos[06:40:33] <bahamat> NginUS: It's a bit old, but still relevant and a good primer: https://youtu.be/tPsV_8k-aVU and https://youtu.be/wy6cJRVHiYU[06:41:40] <NginUS> bahamat: thanks![06:42:18] <bahamat> NginUS: Beware, once you fall into that rabbit hole, you'll never want to come out.[06:44:21] <NginUS> bahamat: My only experience thus far is NexentaStor's Web UI. Aside from RAM-eating, I hear only good things.[06:45:18] <bahamat> Well, all filesystems of recent vintage will use all available ram for disk cache.[06:52:30] <bahamat> Smithx10: That public key you showed us before doesn't have quotes in it. It has smart quotes.[06:53:14] <bahamat> Smithx10: Python reads the string as`\u201ccomment_here\u201d`[06:53:43] <bahamat> Smithx10: Using actual quotation marks works fine.[06:56:46] <Smithx10> you provisioned an instance with that public key? d6e56c0e-1e51-11e9-8ff1-2fecf38a5566 was the image[06:57:15] <bahamat> Smithx10: Yes[06:59:04] <bahamat> I also did 62ade7b4-a35c-11e8-a083-73a0170b4eda, which is debian-8, uses cloud-init, and exhibits the same problem; and f7d5c462-9ace-11e6-9e1e-83b238b5d758 which is debian-7, does not use cloud-init, and does not exhibit the problem.[06:59:23] <bahamat> So this is very clearly a cloud-init bug.[06:59:29] <Smithx10> ahhhh so you reproduced it ?[06:59:33] <bahamat> This would happen anywhere.[06:59:40] <bahamat> Yeah, I can easily reproduce it.[07:01:08] <bahamat> Anything with cloud-init does the same thing. It provisions, but /root/.ssh/authorized_keys doesn't exist.[07:01:36] <Smithx10> gotcha, so cloud-init has a bug[07:01:53] <Smithx10> w00t[07:02:06] <Smithx10> ./winners!!!!! :)[07:03:11] <bahamat> cloud-init also does not set the root password according to the internal_metadata.root_pw param, so that's two bugs in one hunt.[07:04:17] <bahamat> Smithx10: I'll pm you a small script that will open the cloud-init pickel and let you see what python thinks of the key.[07:04:25] <bahamat> Because it *does* make it into cloud-init.[07:04:33] <bahamat> But something breaks on the way out.[07:04:40] <bahamat> I didn't bother to look into what or why.[07:07:27] *** kayront <kayront!~kayront@unaffiliated/kayront> has quit IRC (Quit: ZNC 1.7.1 - https://znc.in)[07:51:54] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has joined #smartos[08:02:50] *** miller7 <miller7!~miller7@unaffiliated/miller7> has quit IRC (Ping timeout: 244 seconds)[09:26:58] *** fejfighter_ <fejfighter_!~fejfighte@58.84.227.51> has joined #smartos[09:29:30] *** fejfighter <fejfighter!~fejfighte@58.84.227.51> has quit IRC (Ping timeout: 250 seconds)[10:00:42] <jubal> gross/stupid question: any chance virtualbox will run on smartos - say in a base-64 zone?[10:02:41] <jubal> (relocating a few vb instances from solaris - would love to kick the can down the road and move them into a smartos zone running vbox instead of moving them to bhyve)[10:35:10] *** fejfighter_ <fejfighter_!~fejfighte@58.84.227.51> has quit IRC (Ping timeout: 246 seconds)[10:42:52] <jubal> i'll move them to bhyve.. gah, lazy[11:26:25] *** andy_js <andy_js!~andy@94.3.60.133> has joined #smartos[11:28:29] <jrg> is there a reason why images from imgadm have a snapshot?[11:28:44] <jrg> like zvol snapshots are in zfs list -t snapshot as uuid@final[11:31:11] <wiedi> the snapshot is needed so the image can be cloned (to create a new vm)[11:45:56] <jrg> ah ok[11:46:09] <jrg> thanks[11:46:44] <jrg> i'm trying to migrate a zimbra install from a kvm to a lx zone[11:47:06] *** v_a_b <v_a_b!~volker@p4FECCA15.dip0.t-ipconnect.de> has joined #smartos[11:47:08] <jrg> i've been testing it for a couple of days with another lx zone and it seems to be working rather well[11:47:22] <jrg> so i'm going to migrate and kill the kvm and use the lx zone for it on another one[11:53:15] <jrg> Total: 184 processes, 1438 lwps, load averages: 6.46, 5.48, 4.74[11:53:23] <jrg> well then heh[11:59:38] <jrg> so with regard to zones. would increasing the quota increase the reported disk space in the zone?[12:24:37] <jubal> jrg - you can vmadm update $thevmguid quota=256 to enable it to see 256G of usable space :)[12:24:50] <jubal> it works live[12:25:18] <jubal> (in native and lx zones)[12:56:26] <jrg> nice thanks[12:57:15] <mattronix> bahamat: what do you think is best for me to do with the server?[12:57:21] <mattronix> bahamat: reset it?[13:01:28] *** olsner <olsner!~salparot@c193-14-198-108.bredband.comhem.se> has quit IRC (Ping timeout: 246 seconds)[13:09:07] *** olsner <olsner!~salparot@c193-14-198-108.bredband.comhem.se> has joined #smartos[13:38:19] <jrg> awesome. migrated a zimbra install from a smartos kvm to a smartos lx zone and it’s running without a problem. :)[13:38:26] * jrg takes his ram back[14:14:12] *** jcea <jcea!~Thunderbi@2001:41d0:1:8a82:7670:6e00:7670:6e00> has joined #smartos[16:21:36] *** vila <vila!~vila@lec67-4-82-230-53-244.fbx.proxad.net> has quit IRC (Ping timeout: 268 seconds)[16:23:28] *** vila <vila!~vila@lec67-4-82-230-53-244.fbx.proxad.net> has joined #smartos[17:08:12] <Smithx10> Lololol[17:08:22] <Smithx10> <3 ram[17:32:12] *** v_a_b <v_a_b!~volker@p4FECCA15.dip0.t-ipconnect.de> has quit IRC (Ping timeout: 244 seconds)[17:49:03] <jrg> Smithx10: ram is nice[17:49:12] <jrg> if lx zones ever support smb.. that would make the lx zones perfect heh[17:49:31] <jrg> i would rather avoid mounting on the smartos side[18:03:48] *** mmlb0 <mmlb0!~mmlb@76-248-148-178.lightspeed.miamfl.sbcglobal.net> has quit IRC (Ping timeout: 245 seconds)[18:27:44] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has quit IRC (Remote host closed the connection)[18:30:12] *** tuxillo <tuxillo!~antonioh@89.128.97.235> has joined #smartos[18:42:23] *** rsully <rsully!~rsully@unaffiliated/rsully> has joined #smartos[18:44:14] *** taylorjonl <taylorjonl!c7c0634b@gateway/web/freenode/ip.199.192.99.75> has joined #smartos[19:03:42] *** v_a_b <v_a_b!~volker@p57A27243.dip0.t-ipconnect.de> has joined #smartos[19:29:16] *** morgib <morgib!uid342182@gateway/web/irccloud.com/x-azhumailgiyjhosb> has joined #smartos[19:43:13] <jrg> # AUTOMATIC ZONE CONFIG[19:43:23] <jrg> where do i cahnge the resolv.conf for debian9 lx zones?[19:43:27] *** axonpoet <axonpoet!~axonpoet@fsf/member/axonpoet> has joined #smartos[19:52:56] *** Asgaroth <Asgaroth!~Asgaroth@212.2.172.228> has quit IRC (Ping timeout: 240 seconds)[20:15:06] *** fejfighter_ <fejfighter_!~fejfighte@58.84.227.51> has joined #smartos[20:24:39] *** yomi <yomi!~void@ip4d16bd91.dynamic.kabel-deutschland.de> has joined #smartos[20:25:03] *** yomi is now known as Guest86086[20:28:10] *** Guest22418 <Guest22418!~void@77.22.189.145> has quit IRC (Ping timeout: 250 seconds)[20:31:26] *** Guest86086 <Guest86086!~void@ip4d16bd91.dynamic.kabel-deutschland.de> has quit IRC (Ping timeout: 240 seconds)[20:32:26] *** axonpoet <axonpoet!~axonpoet@fsf/member/axonpoet> has quit IRC (Ping timeout: 240 seconds)[20:32:33] *** Guest86086 <Guest86086!~void@ip4d16bd91.dynamic.kabel-deutschland.de> has joined #smartos[20:45:56] <LeftWing> jrg: I believe the values get populated at boot via: https://github.com/joyent/illumos-joyent/blob/5d687f5a37b08ffb3af39f1fc475eba83078c761/usr/src/lib/brand/lx/zone/lx_boot_zone_debian.ksh#L39-L62[20:46:38] <LeftWing> The values come from zonecfg(1M), which is where vmadm(1M) stores the values for the VM-level "resolvers" and "dns_domain" properties.[21:10:03] <sjorge> rmustacc: what does clobber and update do? I also have my non default ‘repo’ set in the config so that is all fine.[21:12:58] <jrg> LeftWing: is there a way to change the value from the default with vmadm?[21:16:06] <jrg> oh[21:16:09] <jrg> nm i found it[21:26:54] <Latrina> good evening guys[21:26:59] <Latrina> hope you are all well[22:08:39] *** morgib <morgib!uid342182@gateway/web/irccloud.com/x-azhumailgiyjhosb> has quit IRC (Quit: Connection closed for inactivity)[22:14:18] <sjorge> The JTD playlist is agreat![22:14:30] <sjorge> Just saw a few new ones got added![22:15:03] <jrg> winbind.service: Failed to set invocation ID on control group /system.slice/winbind.service, ignoring: Operation not supported[22:15:06] <jrg> hm[22:16:35] <jrg> is that a lx zone thing?[22:34:40] <jrg> ah nm. seems like i had to join the domain[22:34:52] <jrg> then restart it. hm. most instructions don't have it happening that way[22:35:01] <jrg> whatever. it's actually working in the lx zone. i can authenticate against the AD[22:40:18] *** notmatt <notmatt!uid342873@gateway/web/irccloud.com/x-ykbainekyhvsfavy> has joined #smartos[22:40:56] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has quit IRC (Ping timeout: 240 seconds)[22:43:42] *** sjorge <sjorge!~sjorge@unaffiliated/sjorge> has joined #smartos[22:48:14] <jrg> awesome. i'm using irssi in the lx zone now and everything works. (other than smb of course)[22:48:40] <jrg> loving lx zones[22:49:08] <jrg> should have used these first. but i really needed smb support for linux. i put in a request though so hopefully someone looks into getting smb mounting to work in the lx zone like nfs.[23:48:12] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has quit IRC (Quit: leaving)[23:54:41] *** theandromedan <theandromedan!~pfranz@c-174-59-156-209.hsd1.pa.comcast.net> has joined #smartos