Switch to DuckDuckGo Search
   March 25, 2013  
< | 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 | >

bottom
[00:04:24] <ira> "Yes."
[00:04:59] <ira> Literally, the error is as it says.
[00:05:22] <ira> If you use the imgapi it works, the dsapi is broken.
[00:06:18] <Licenser> hmmmm?
[00:07:04] <ira> I have an imgapi server I just wrote, it seems fine.
[00:10:57] *** xinkeT has joined #smartos
[00:10:57] *** LeftWing has quit IRC
[00:12:23] *** id__ has joined #smartos
[00:12:31] *** scarcry has quit IRC
[00:12:44] *** scarcry_ has joined #smartos
[00:13:15] <ira> https://github.com/joyent/smartos-live/issues/190
[00:13:21] *** hungryblank has quit IRC
[00:13:25] *** fortytwo_ has quit IRC
[00:14:01] *** dizko_ has quit IRC
[00:14:07] *** nhubbard_ has joined #smartos
[00:14:29] *** Melinko has quit IRC
[00:14:29] *** olafm_ has joined #smartos
[00:14:33] *** Melinko has joined #smartos
[00:14:51] *** nhubbard has quit IRC
[00:14:56] *** nhubbard_ is now known as nhubbard
[00:14:58] *** arekinath has quit IRC
[00:15:01] *** dizko_ has joined #smartos
[00:15:09] *** Tekni has quit IRC
[00:15:12] *** apsonge has joined #smartos
[00:15:26] *** Zigara_ has joined #smartos
[00:16:12] *** wolstena has joined #smartos
[00:18:17] *** jesse___ has joined #smartos
[00:18:17] *** amgrecu has joined #smartos
[00:18:18] *** Hazel|ar1emis has joined #smartos
[00:19:26] *** LeftWing has joined #smartos
[00:19:26] *** arekinath has joined #smartos
[00:19:26] *** arekinath has joined #smartos
[00:23:38] *** mAniAk-_- has quit IRC
[00:23:39] *** JT-EC has quit IRC
[00:23:41] *** olafm has quit IRC
[00:23:43] *** Zigara has quit IRC
[00:23:44] *** id_ has quit IRC
[00:23:46] *** asonge has quit IRC
[00:23:48] *** aszeszo has quit IRC
[00:23:53] *** porkbelt_ has joined #smartos
[00:23:54] *** szaydel has quit IRC
[00:23:56] *** mAniAk-_- has joined #smartos
[00:23:58] *** bahamas101 has joined #smartos
[00:23:59] <nahamu> Licenser: I added one line to the json of manifest I wanted to import and it worked.
[00:24:01] <nahamu> I had to add "state":"active"
[00:24:01] <nahamu> which I believe is technically part of imgapi, not dsapi
[00:24:08] *** hij1nx has quit IRC
[00:24:12] *** phus has quit IRC
[00:24:13] *** Hazel|artemis has quit IRC
[00:24:13] *** jbergstroem has quit IRC
[00:24:14] *** rbrown2 has quit IRC
[00:24:15] *** jesse_ has quit IRC
[00:24:15] *** samu has quit IRC
[00:24:21] *** samu has joined #smartos
[00:24:22] <Licenser> nahamut thanks :)
[00:24:23] *** porkbelt has quit IRC
[00:24:24] *** rbrown__ has quit IRC
[00:24:24] *** opeth__ has quit IRC
[00:24:25] *** porkbelt_ is now known as porkbelt
[00:24:25] *** samu has quit IRC
[00:24:26] *** hij1nx has joined #smartos
[00:24:28] *** bahamas10 has quit IRC
[00:24:29] *** Niamkik has quit IRC
[00:24:29] *** Niamkik has joined #smartos
[00:24:30] *** jbergstroem has joined #smartos
[00:24:52] *** rbrown__ has joined #smartos
[00:24:53] *** phus has joined #smartos
[00:25:50] *** szaydel has joined #smartos
[00:27:02] *** olafm_ has quit IRC
[00:27:27] <ira> When talking over /datasets ?
[00:27:35] <ira> That's busted… :(
[00:28:20] <nahamu> ira: yeah, my guess is that the imgadm client is looking for that "state":"active"
[00:28:35] <nahamu> which is part of the imgapi but not of the dsapi
[00:28:40] <ira> yeah… that's part of the imgapi spec...
[00:28:57] *** JT-EC has joined #smartos
[00:28:57] *** aszeszo has joined #smartos
[00:28:57] <nahamu> but adding that to a dsapi manifest is enough to trick it into cooperating.
[00:29:03] *** 45PAACJ8Z has joined #smartos
[00:29:03] *** opeth__ has joined #smartos
[00:29:12] <ira> Please file a bug.
[00:29:24] <nahamu> I think technically it's the bug you already filed.
[00:30:10] <ira> I think there's other changes between the two api's.
[00:30:30] <ira> then follow up to mine if it isn't clear :)
[00:30:47] <ira> (Because I don't document that issue...)
[00:31:06] <nahamu> I'll add a comment to that issue.
[00:31:06] <ira> And having a imgapi server now… X_x. It is less of an issue.
[00:31:26] <ira> Thanks.
[00:31:26] <nahamu> any chance of you posting your code for others to use?
[00:33:36] <ira> As soon as I talk to my boss.
[00:34:48] <nahamu> cool
[00:35:32] *** arekinath has quit IRC
[00:36:23] *** samu has joined #smartos
[00:36:30] *** 45PAACJ8Z has quit IRC
[00:36:31] *** fortytwo_ has joined #smartos
[00:36:51] *** arekinath has joined #smartos
[00:37:03] *** samu is now known as Guest6423
[00:39:44] *** Guest6423 has quit IRC
[00:40:01] *** Guest6423 has joined #smartos
[00:40:13] *** dalethion has joined #smartos
[00:43:49] *** opeth___ has joined #smartos
[00:43:50] *** fortytwo1 has joined #smartos
[00:44:01] *** arekinath has quit IRC
[00:44:21] *** arekinath has joined #smartos
[00:44:21] *** arekinath has joined #smartos
[00:44:59] *** fortytwo_ has quit IRC
[00:45:00] *** opeth__ has quit IRC
[00:47:32] *** noeyesan1icw has joined #smartos
[00:48:01] *** amgrecu has quit IRC
[00:48:02] *** Xenith has quit IRC
[00:49:18] *** Xenith has joined #smartos
[00:49:25] *** wolstena1 has joined #smartos
[00:54:05] *** echelog-1 has joined #smartos
[00:54:56] *** badboy_ has joined #smartos
[00:56:18] *** dizko_ has joined #smartos
[00:57:52] *** apsonge has quit IRC
[00:57:52] *** kfr_ has quit IRC
[00:57:53] *** ryao has quit IRC
[00:58:01] *** halldorh has quit IRC
[00:58:02] *** EMH_Mark3 has quit IRC
[00:58:06] *** bahamas101 has quit IRC
[00:58:08] *** wolstena has quit IRC
[00:58:09] *** id__ has quit IRC
[00:58:11] *** elijah-mbp has quit IRC
[00:58:17] *** wesolows has quit IRC
[00:58:19] *** windows has quit IRC
[00:58:20] *** jeffpc has quit IRC
[00:58:21] *** joshie has quit IRC
[00:58:22] *** scanf has quit IRC
[00:59:02] *** benbangert has quit IRC
[00:59:02] *** konobi has quit IRC
[01:00:34] *** nhubbard has quit IRC
[01:01:48] *** denizr has left #smartos
[01:02:21] *** tardisx has joined #smartos
[01:02:44] *** tardisx` has quit IRC
[01:02:54] *** apsonge has joined #smartos
[01:02:54] *** kfr_ has joined #smartos
[01:02:54] *** ryao has joined #smartos
[01:02:54] *** halldorh has joined #smartos
[01:02:55] *** EMH_Mark3 has joined #smartos
[01:03:43] *** ajcdotme has joined #smartos
[01:03:44] *** chris--- has joined #smartos
[01:04:06] *** bahamas101 has joined #smartos
[01:04:06] *** elijah-mbp has joined #smartos
[01:04:07] *** wesolows has joined #smartos
[01:04:07] *** windows has joined #smartos
[01:04:07] *** jeffpc has joined #smartos
[01:04:07] *** joshie has joined #smartos
[01:04:07] *** scanf has joined #smartos
[01:04:11] *** benbangert has joined #smartos
[01:04:12] *** nhubbard has joined #smartos
[01:05:35] *** nikolam has quit IRC
[01:06:28] *** konobi has joined #smartos
[01:06:56] *** Zigara_ is now known as Zigara
[01:07:43] *** joshie_ has joined #smartos
[01:08:15] *** jeffpc has quit IRC
[01:08:53] *** joshie has quit IRC
[01:09:33] *** jeffpc has joined #smartos
[01:10:04] *** Zigara has quit IRC
[01:10:05] *** Zigara has joined #smartos
[01:10:11] *** Melinko has quit IRC
[01:10:38] *** Melinko has joined #smartos
[01:12:07] *** cmang has quit IRC
[01:12:13] *** cmang has joined #smartos
[01:14:00] *** olafm has joined #smartos
[01:14:56] *** rbrown_ has quit IRC
[01:15:08] *** rbrown_ has joined #smartos
[01:15:14] *** wesolows has quit IRC
[01:15:15] *** scanf has quit IRC
[01:15:29] *** wesolows has joined #smartos
[01:22:22] *** scanf has joined #smartos
[01:22:58] *** ajcdotme has quit IRC
[01:22:59] *** chris--- has quit IRC
[01:23:30] *** chris--- has joined #smartos
[01:23:36] *** ajcdotme has joined #smartos
[01:24:23] *** quasi has quit IRC
[01:24:23] *** bbarker` has quit IRC
[01:24:23] *** ice799 has quit IRC
[01:24:26] *** bdha has quit IRC
[01:24:27] *** robinsmidsrod has quit IRC
[01:24:28] *** prasm has quit IRC
[01:25:02] *** ryancnelson has joined #smartos
[01:28:32] *** nhubbard has quit IRC
[01:28:40] *** dumfries has quit IRC
[01:28:41] *** yofuh has quit IRC
[01:29:14] *** yofuh has joined #smartos
[01:29:25] *** bdha has joined #smartos
[01:29:50] *** quasi has joined #smartos
[01:29:50] *** quasi has joined #smartos
[01:30:12] *** jeffpc_ has joined #smartos
[01:30:21] *** bixu has joined #smartos
[01:30:34] *** cmang_ has joined #smartos
[01:30:37] *** nhubbard has joined #smartos
[01:31:01] *** olafm_ has joined #smartos
[01:31:27] *** dumfries has joined #smartos
[01:32:36] *** prasm has joined #smartos
[01:32:54] *** xenol has quit IRC
[01:33:08] *** sheppard has quit IRC
[01:33:23] *** xenol has joined #smartos
[01:33:47] *** elijah-mbp has quit IRC
[01:34:02] *** ismell has joined #smartos
[01:34:23] *** elijah-mbp has joined #smartos
[01:34:43] *** ajcdotme has quit IRC
[01:34:44] *** chris--- has quit IRC
[01:34:47] *** rbrown_ has quit IRC
[01:34:48] *** olafm has quit IRC
[01:34:49] *** cmang has quit IRC
[01:34:50] *** jeffpc has quit IRC
[01:34:53] *** joshie_ has quit IRC
[01:34:58] *** id_ has quit IRC
[01:34:59] *** wolstena1 has quit IRC
[01:35:02] *** ismell_ has quit IRC
[01:35:10] *** bixu has quit IRC
[01:35:23] *** rbrown_ has joined #smartos
[01:35:30] *** robinsmidsrod has joined #smartos
[01:36:46] *** windows has quit IRC
[01:36:53] *** richlowe has quit IRC
[01:38:57] *** sheppard has joined #smartos
[01:40:22] *** prasm has quit IRC
[01:40:52] *** prasm has joined #smartos
[01:41:05] *** denizr has joined #smartos
[01:41:22] *** wolstena has joined #smartos
[01:41:48] *** richlowe has joined #smartos
[01:42:35] *** id_ has joined #smartos
[01:43:54] *** windows has joined #smartos
[01:45:26] *** joshie has joined #smartos
[01:45:26] *** ajcdotme has joined #smartos
[01:45:26] *** chris--- has joined #smartos
[01:46:06] *** joshie is now known as Guest16753
[01:48:11] *** wolstena1 has joined #smartos
[01:48:22] *** trentster_ has joined #smartos
[01:48:34] *** Trixboxer has quit IRC
[01:48:36] *** trentster_ has left #smartos
[01:49:15] *** tardisx` has joined #smartos
[01:49:31] *** xmerlin_ has joined #smartos
[01:50:30] *** wolstena1 is now known as 65MAAQJCX
[01:50:33] *** wolstena1 has joined #smartos
[01:50:37] *** wolstena has quit IRC
[01:50:41] *** trentster has quit IRC
[01:50:44] *** tardisx has quit IRC
[01:50:54] *** wolstena1 has quit IRC
[01:51:11] *** tallship has quit IRC
[01:51:11] *** xmerlin has quit IRC
[01:51:39] *** tallship has joined #smartos
[01:52:54] *** trentster has joined #smartos
[02:03:55] *** szaydel has quit IRC
[02:05:16] *** szaydel has joined #smartos
[02:07:06] *** richlowe` has joined #smartos
[02:08:03] *** richlowe has quit IRC
[02:08:12] *** aszeszo has quit IRC
[02:08:13] *** JT-EC has quit IRC
[02:08:14] *** aszeszo has joined #smartos
[02:08:14] *** JT-EC has joined #smartos
[02:08:17] *** Guest6423 has quit IRC
[02:08:19] *** samu has joined #smartos
[02:09:57] *** Xenith has quit IRC
[02:12:26] *** kfr_ has quit IRC
[02:13:15] *** Xenith has joined #smartos
[02:15:35] *** hungryblank has joined #smartos
[02:17:25] *** richlowe` is now known as richlowe
[02:18:57] *** wolstena has joined #smartos
[02:19:30] *** samu has quit IRC
[02:20:00] *** ryao has quit IRC
[02:20:15] *** ryao has joined #smartos
[02:20:55] *** JT-EC has quit IRC
[02:20:55] *** aszeszo has quit IRC
[02:20:55] *** aszeszo has joined #smartos
[02:22:25] *** samu has joined #smartos
[02:24:42] *** Cpt-Oblivious has quit IRC
[02:28:02] *** szaydel has quit IRC
[02:28:04] *** xmerlin_ has quit IRC
[02:28:05] *** 65MAAQJCX has quit IRC
[02:28:05] *** trentster has quit IRC
[02:28:06] *** bahamas101 has quit IRC
[02:28:49] *** EMH_Mark3 has quit IRC
[02:28:50] *** szaydel has joined #smartos
[02:29:56] *** trentster has joined #smartos
[02:30:50] *** apsonge has quit IRC
[02:30:51] *** halldorh has quit IRC
[02:30:51] *** ryao has quit IRC
[02:31:09] *** mikeal has joined #smartos
[02:31:34] *** Guest16753 has quit IRC
[02:31:39] *** samu has quit IRC
[02:31:39] *** samu has joined #smartos
[02:31:40] *** samu has joined #smartos
[02:31:48] *** joshie_ has joined #smartos
[02:31:50] *** asonge has joined #smartos
[02:32:14] *** EMH_Mark4 has joined #smartos
[02:33:10] *** aszeszo has quit IRC
[02:33:45] *** Xenith has quit IRC
[02:34:07] *** xmerlin_ has joined #smartos
[02:38:39] *** halldorh has joined #smartos
[02:38:53] *** Xenith has joined #smartos
[02:39:10] *** Xenith has quit IRC
[02:40:52] *** asonge has quit IRC
[02:41:04] *** richlowe has quit IRC
[02:41:07] *** richlowe has joined #smartos
[02:41:55] *** aszeszo has joined #smartos
[02:43:50] *** ajcdotme_ has joined #smartos
[02:44:25] *** trentster has quit IRC
[02:45:19] *** ryao has joined #smartos
[02:45:38] *** mikeal has left #smartos
[02:45:52] *** Xenith has joined #smartos
[02:46:42] *** JT-EC has joined #smartos
[02:47:48] *** trentster has joined #smartos
[02:48:16] *** asonge has joined #smartos
[02:50:11] *** bahamas101 has joined #smartos
[02:50:26] *** joshie has joined #smartos
[02:50:43] *** samu has quit IRC
[02:56:44] *** joshie__ has joined #smartos
[03:00:54] *** wolstena1 has joined #smartos
[03:02:04] *** JT-EC has quit IRC
[03:02:05] *** aszeszo has quit IRC
[03:02:41] *** Xenith has quit IRC
[03:09:39] *** joshie_ has quit IRC
[03:09:39] *** wolstena has quit IRC
[03:09:46] *** ajcdotme has quit IRC
[03:09:46] *** ajcdotme_ is now known as ajcdotme
[03:10:29] *** axonpoet has joined #smartos
[03:10:43] *** Zigara has quit IRC
[03:11:06] *** aszeszo has joined #smartos
[03:11:07] *** Zigara has joined #smartos
[03:11:27] *** Zigara has quit IRC
[03:11:27] *** Zigara has joined #smartos
[03:11:55] *** ira has quit IRC
[03:11:56] *** joshie has quit IRC
[03:12:02] *** szaydel has quit IRC
[03:12:03] *** chris--- has quit IRC
[03:14:01] *** Xenith has joined #smartos
[03:14:50] *** samu has joined #smartos
[03:15:42] *** JT-EC has joined #smartos
[03:19:37] *** tallship has quit IRC
[03:20:54] *** Zigara has quit IRC
[03:24:33] *** samu has quit IRC
[03:31:28] *** bixu has joined #smartos
[03:33:20] *** trentster has quit IRC
[03:33:47] *** wolstena1 has quit IRC
[03:33:56] *** trentster has joined #smartos
[03:34:06] *** trentster has left #smartos
[03:34:18] *** richlowe has quit IRC
[03:34:25] *** wolstena has joined #smartos
[03:34:47] *** JT-EC has quit IRC
[03:34:48] *** aszeszo has quit IRC
[03:34:48] *** aszeszo has joined #smartos
[03:34:48] *** szaydel has joined #smartos
[03:35:19] *** bixu has quit IRC
[03:35:29] *** ice799 has joined #smartos
[03:35:31] *** Xenith has quit IRC
[03:35:55] *** Xenith has joined #smartos
[03:36:14] *** szaydel has quit IRC
[03:38:26] *** denizr has quit IRC
[03:39:01] *** samu has joined #smartos
[03:41:58] *** jaimef has quit IRC
[03:41:59] *** blu has quit IRC
[03:42:29] *** richlowe has joined #smartos
[03:42:32] *** Zigara has joined #smartos
[03:42:42] *** fortytwo1 has quit IRC
[03:42:43] *** ChrisPartridge has quit IRC
[03:42:43] *** leecallen has quit IRC
[03:42:49] *** CPartridge has joined #smartos
[03:43:40] *** blu has joined #smartos
[03:44:54] *** leecallen has joined #smartos
[03:45:58] *** ice799 has quit IRC
[03:46:36] *** wolstena1 has joined #smartos
[03:46:42] *** ice799 has joined #smartos
[03:46:43] *** jaimef has joined #smartos
[03:47:22] *** aszeszo has quit IRC
[03:47:23] *** aszeszo has joined #smartos
[03:47:36] *** JT-EC has joined #smartos
[03:48:32] *** wolstena has quit IRC
[03:49:51] *** fortytwo_ has joined #smartos
[03:51:59] *** szaydel has joined #smartos
[03:52:56] *** Zigara has quit IRC
[03:52:56] *** Zigara has joined #smartos
[04:02:46] *** fortytwo1 has joined #smartos
[04:05:08] *** fortytwo_ has quit IRC
[04:05:38] *** wolstena has joined #smartos
[04:05:50] *** ice799 has quit IRC
[04:05:59] *** joe1 has joined #smartos
[04:06:50] *** JT-EC has quit IRC
[04:06:50] *** aszeszo has quit IRC
[04:06:51] *** aszeszo has joined #smartos
[04:09:19] *** wolstena1 has quit IRC
[04:10:42] *** JT-EC has joined #smartos
[04:13:45] *** denizr has joined #smartos
[04:17:40] *** ipalreadytaken has joined #smartos
[04:19:57] *** jwk404 has quit IRC
[04:20:09] *** jwk404 has joined #smartos
[04:26:42] *** ajcdotme has quit IRC
[04:34:35] *** xmerlin_ has quit IRC
[04:50:34] *** denizr1 has joined #smartos
[04:51:56] *** samu has quit IRC
[04:57:28] *** samu has joined #smartos
[04:58:14] *** denizr has quit IRC
[04:58:14] *** JT-EC has quit IRC
[04:58:16] *** aszeszo has quit IRC
[05:01:50] *** aszeszo has joined #smartos
[05:02:15] *** JT-EC has joined #smartos
[05:05:05] *** JT-EC has quit IRC
[05:07:15] *** JT-EC has joined #smartos
[05:17:35] *** trentster has joined #smartos
[05:28:18] *** enmand has quit IRC
[05:31:12] *** bixu has joined #smartos
[05:35:25] *** bixu has quit IRC
[05:36:20] *** szaydel has quit IRC
[05:53:59] *** sachinsharma has joined #smartos
[05:59:30] <trentster> Whats the best way to see cpu usage per vm? both zone and kvm?
[06:02:01] *** ChrisPartridge has joined #smartos
[06:03:31] *** CPartridge has quit IRC
[06:05:22] <LeftWing> trentster: prstat -Z ?
[06:05:42] *** jgrafton has joined #smartos
[06:06:07] <trentster> LeftWing: umm yeah but thats an interactive command , was looking at soemthing to pull metrics for graphing and monitoring purposes..
[06:06:20] <trentster> I guess I probably should have said that ;-)
[06:06:42] *** ryancnelson has quit IRC
[06:07:23] <rmustacc> The zone cpu kstats.
[06:07:40] *** ryancnelson has joined #smartos
[06:07:42] <rmustacc> But keep in mind for kvm the user/sys time isn't very useful.
[06:07:54] <trentster> rmustacc: which ones specifically and can you give me an e.g. that works for both kvm and zones?
[06:08:02] <trentster> would this be cpu cap and cpu cap used ?
[06:08:06] <rmustacc> No.
[06:08:14] <rmustacc> Depends on what you're actually trying to see.
[06:08:31] <rmustacc> I can't tell you what the guest thinks its cpu usage is.
[06:08:43] <trentster> trying to see how each vm, is using their cpu allocation
[06:09:14] <rmustacc> You can look at the zone_misc stat
[06:09:21] <rmustacc> Well, that's different entirely then.
[06:13:21] *** leecallen35 has joined #smartos
[06:13:21] *** leecallen has quit IRC
[06:13:39] *** nefilim has quit IRC
[06:14:22] <trentster> rmustacc: what kstat would you use to graph cpu usage per vm, for the reason of seeing if the vm is CPU bound?
[06:16:01] <rmustacc> Do you want to graph when it's bursting?
[06:16:09] <rmustacc> I mean there are several things to graph.
[06:16:17] <rmustacc> You could start with the zone user/sys time.
[06:16:48] <rmustacc> You could graph the number of seconds its bursting over time.
[06:17:54] <trentster> well lets say for e.g.I wanted to use it to make a pretty report that tells a user that his cpu/s were working at 90% capacity 80% of the time.
[06:18:14] <trentster> if in fact they were, which combination of kstats would tell me that
[06:18:28] <rmustacc> I don't know off hand.
[06:18:42] <rmustacc> Have you looked at the cpucaps?
[06:18:50] <rmustacc> The specific kstats.
[06:19:09] <trentster> yup, been looking at it "kstat -p caps::cpucaps_zone*"
[06:22:55] <trentster> rmustacc: thanks
[06:38:19] *** amgrecu has joined #smartos
[06:40:25] *** wolfeidau has quit IRC
[06:40:42] *** wolfeidau has joined #smartos
[06:41:54] *** rmustacc_ has joined #smartos
[06:42:34] *** |woody_| has joined #smartos
[06:43:44] *** jesse___ is now known as jesse_
[06:44:26] *** sjorge_be has joined #smartos
[06:44:57] *** ryao_ has joined #smartos
[06:44:58] *** xenol_ has joined #smartos
[06:45:04] *** wesolows_ has joined #smartos
[06:45:38] *** Lumb_ has joined #smartos
[06:46:20] *** dalethion has quit IRC
[06:46:32] *** rmustacc has quit IRC
[06:46:38] *** sjorge has quit IRC
[06:46:42] *** |woody| has quit IRC
[06:46:45] *** beau-_ has quit IRC
[06:46:46] *** ryao has quit IRC
[06:46:46] *** xenol has quit IRC
[06:46:47] *** wesolows has quit IRC
[06:46:49] *** Lumb has quit IRC
[06:46:50] *** Eimann has quit IRC
[06:46:53] *** jamesog has quit IRC
[06:46:54] *** jamesog has joined #smartos
[06:47:06] *** jamesog is now known as Guest46191
[06:49:54] *** denizr1 has quit IRC
[06:56:52] *** ryancnelson1 has joined #smartos
[06:57:10] *** ryancnelson has quit IRC
[06:59:00] *** ryancnelson has joined #smartos
[07:01:43] *** ryancnelson1 has quit IRC
[07:03:35] *** wolfeidau has quit IRC
[07:19:47] *** |woody_| is now known as |woody|
[07:20:49] *** ryancnelson1 has joined #smartos
[07:23:30] *** ryancnelson has quit IRC
[07:24:17] *** ryancnelson has joined #smartos
[07:25:00] *** wolfeidau has joined #smartos
[07:26:41] *** ryancnelson1 has quit IRC
[07:31:31] *** bixu has joined #smartos
[07:35:58] *** bixu has quit IRC
[07:39:58] *** ryancnelson has joined #smartos
[08:12:36] *** ryancnelson has quit IRC
[08:12:54] *** ryancnelson has joined #smartos
[08:15:32] *** kaladis has joined #smartos
[08:22:32] *** texarcana has joined #smartos
[08:37:10] *** ChrisPartridge has quit IRC
[08:40:46] *** scarcry_ is now known as scarcry
[08:43:34] *** beau-_ has joined #smartos
[08:45:39] *** Eimann has joined #smartos
[08:50:48] *** bens1 has joined #smartos
[09:03:19] *** blindcoder has joined #smartos
[09:04:14] <blindcoder> morning
[09:04:29] <blindcoder> is it possible to permanently create normal users in the global zone?
[09:04:47] <jesse_> with much effort
[09:05:21] <jesse_> you can add a smf service that runs a script that adds the users to /etc files, for example
[09:05:40] <jesse_> or use lofi mounts(?)
[09:05:54] <jesse_> or even modify the platform image
[09:06:02] <jesse_> or, the best solution
[09:06:06] <jesse_> don't do it=)
[09:06:41] <LeftWing> It's also possible to configure the LDAP client in the global zone, if you have admin users in LDAP.
[09:07:12] <jesse_> dang, I forgot that one
[09:07:31] *** alcir has joined #smartos
[09:08:00] <LeftWing> You shouldn't be running your workloads in the global, but it makes sense to me to have user accounts in there so that you can get in and manage/instrument things when required.
[09:09:45] <jesse_> another case is, if you really need to run nfs server
[09:13:46] <blindcoder> okay, let's take one step back :)
[09:14:00] <blindcoder> I have two public IPs, one in the GZ, one in a Linux KVM
[09:14:14] <blindcoder> on the linux server I run apache on 80 and 443 and on the GZ I run ssh on 443
[09:14:43] <blindcoder> now what I want to do is do a HTTP CONNECT to the ssh on 443 via the company proxy
[09:15:00] <blindcoder> so for now I CONNECT to the GZ and from there ssh to the linux kvm
[09:15:04] <jesse_> stupid question: why run kvm linux if you just run apache?
[09:15:22] *** ipalreadytaken has quit IRC
[09:15:23] <blindcoder> jesse_: there's lots more running there, it's my "catchall" service
[09:15:33] <blindcoder> jesse_: apache is just occupying the port 443, obviously
[09:15:43] <jesse_> ah, ok.
[09:16:19] <blindcoder> I tried redirecting the port 443 of the gz to 22 on the kvm, but that didn't quite work
[09:16:41] <blindcoder> rdr rge0 <gz.IP> port 443 -> <kvm.IP> port 22 tcp
[09:17:02] <blindcoder> it works only when connecting froem the GZ, when connectign from outside no packet are going back out
[09:24:22] <jesse_> rdr should work
[09:24:42] <jesse_> I wonder if you need to enable ip forwarding with that
[09:24:53] <jesse_> I've only used that with nat
[09:24:58] <blindcoder> jesse_: interestingly, this works: rdr rge0 78.46.75.109 port 1194 -> 172.29.1.5 port 1194 tcp
[09:25:10] <blindcoder> that's for a small vpn zone
[09:25:30] <blindcoder> jesse_: do rdr have to come before or after map lines?
[09:26:03] <jesse_> no idea if it matters, but seems to work when rdr is after map=)
[09:26:30] <blindcoder> okay, I'll swap that around and try again
[09:31:29] *** dalethion has joined #smartos
[09:31:30] *** bens1_ has joined #smartos
[09:31:55] *** bixu has joined #smartos
[09:33:48] *** kaladis has quit IRC
[09:34:05] *** Lumb has joined #smartos
[09:35:20] *** kaladis has joined #smartos
[09:35:40] *** amgrecu has quit IRC
[09:35:41] *** trentster has quit IRC
[09:36:08] *** alcir_ has joined #smartos
[09:36:41] *** bixu has quit IRC
[09:39:33] *** jamesog has joined #smartos
[09:39:34] *** trentster_ has joined #smartos
[09:39:37] *** trentster_ is now known as trentster
[09:39:42] *** richlowe` has joined #smartos
[09:39:55] *** JT-EC has quit IRC
[09:39:56] *** aszeszo has quit IRC
[09:39:58] *** Lumb_ has quit IRC
[09:39:58] *** |woody| has quit IRC
[09:40:00] *** wolfeidau has quit IRC
[09:40:00] *** samu has quit IRC
[09:40:01] *** jamesog is now known as Guest89260
[09:40:06] *** wesolows_ has quit IRC
[09:40:07] *** jgrafton has quit IRC
[09:40:10] *** wolfeidau has joined #smartos
[09:40:17] *** jgrafton has joined #smartos
[09:40:27] *** richlowe has quit IRC
[09:40:32] *** blindcoder has quit IRC
[09:40:32] *** texarcana has quit IRC
[09:40:33] *** alcir has quit IRC
[09:40:33] *** bens1 has quit IRC
[09:40:36] *** Guest46191 has quit IRC
[09:40:39] *** Xenith has quit IRC
[09:40:43] *** bens1_ is now known as bens1
[09:41:40] *** bluezenix has joined #smartos
[09:41:41] *** |woody| has joined #smartos
[09:41:43] *** bluezenix has quit IRC
[09:41:48] *** bluezenix has joined #smartos
[09:42:57] *** blindcoder has joined #smartos
[09:43:12] <blindcoder> okay, I was going at this problem the wrong way.
[09:43:22] <blindcoder> the forward from the GZ to the KVM works
[09:43:28] *** wesolows has joined #smartos
[09:43:42] <blindcoder> it's just that the packets coming from the KVM don't leave the system...
[09:43:57] <blindcoder> I see the SYN/ACK in the tcpdump
[09:44:26] *** Xenith has joined #smartos
[09:45:33] *** JT-EC has joined #smartos
[09:45:33] *** aszeszo has joined #smartos
[09:45:45] <blindcoder> aaah, I think I got it
[09:45:57] <blindcoder> the KVM has a public and a private IP address
[09:46:24] <blindcoder> when I forward the port to the private IP address, it tries to respond with that directly to the client
[09:46:46] <blindcoder> so: client --> GZ.publicIP
[09:46:53] <blindcoder> BUT: kvm.publicIP --> client
[09:47:50] *** samu has joined #smartos
[09:48:09] <blindcoder> err, kvm.privateIP --> client
[09:49:06] <jesse_> you'll need nat
[09:49:13] <blindcoder> SOLVED
[09:49:31] <blindcoder> I only had a route do 0.0.0.0/0 via the public IP address
[09:49:48] <blindcoder> I now added a route to 0.0.0.0/0 via the private IP address and now I can connect \o/
[09:51:40] <blindcoder> stupid beginners mistake :/
[09:54:10] *** JT-EC has quit IRC
[09:54:11] *** aszeszo has quit IRC
[09:56:28] *** aszeszo has joined #smartos
[09:57:04] *** samu has quit IRC
[09:59:17] *** JT-EC has joined #smartos
[10:02:14] *** samu has joined #smartos
[10:02:36] *** KermitTheFragger has joined #smartos
[10:16:20] *** ipalreadytaken has joined #smartos
[10:21:06] *** ipalreadytaken has quit IRC
[10:55:51] *** bluezenix has quit IRC
[10:56:42] *** scarcry has quit IRC
[10:59:39] *** bluezenix has joined #smartos
[11:03:57] *** Guest89260 has quit IRC
[11:04:15] *** jamesog has joined #smartos
[11:04:39] *** jamesog is now known as Guest68626
[11:06:42] *** scarcry has joined #smartos
[11:07:11] *** Hazel|ar1emis is now known as Hazel|artemis
[11:11:50] *** xenol_ is now known as xenol
[11:16:51] *** ipalreadytaken has joined #smartos
[11:18:50] *** olafm_ is now known as olafm
[11:21:16] *** ipalreadytaken has quit IRC
[11:32:23] *** bixu has joined #smartos
[11:33:27] *** mrvisser has joined #smartos
[11:33:51] <mrvisser> Hi all, I was just wondering how one can protect against a syn dos attack when you have a smartos machine accessible to the public
[11:34:44] <mrvisser> I see a lot of techniques, like reducing the amount of half-open connections you can have open, but I don't see anything that will not still result in a denial of service
[11:36:40] *** bixu has quit IRC
[11:58:43] <blindcoder> mrvisser: that's what syn cookies are for, which I think smartos has included
[11:59:21] <mrvisser> blindcoder: I see. I'll have a read about thouse
[12:01:01] * MerlinDMC loves cookies
[12:02:22] <mrvisser> so if a syn attack is coming from a single IP, will it still stop the system from becoming overload with connection and denying connections from valid users?
[12:11:27] *** system5 has joined #smartos
[12:17:25] *** ipalreadytaken has joined #smartos
[12:23:59] *** ipalreadytaken has quit IRC
[12:26:17] *** ira has joined #smartos
[12:27:53] *** dimitarivanov has joined #smartos
[12:28:09] *** dimitarivanov has left #smartos
[12:35:10] *** szaydel has joined #smartos
[12:35:40] *** sachinsharma has quit IRC
[13:01:43] *** bluezenix has quit IRC
[13:02:52] *** daleg has joined #smartos
[13:07:51] *** porkbelt has quit IRC
[13:08:49] <blindcoder> mrvisser: the idea with syn flood is to fill up the connection database of the server
[13:09:19] <blindcoder> mrvisser: the syn cookie is send back a token with the syn/ack and then forgetting about the connection again
[13:09:45] <blindcoder> mrvisser: that way the connection database of the server never fills up in the first place
[13:10:01] <blindcoder> mrvisser: be aware that this is just a simplified description of my understanding of syncookies
[13:12:16] <mrvisser> blindcoder: I see, that makes sense
[13:12:41] <mrvisser> I can see how it resolves syn floods
[13:13:10] *** porkbelt has joined #smartos
[13:13:19] <mrvisser> a lot of "avoid DoS from syn attacks!" documentation I've read still results in the service becoming unavailable, it just avoids the machine from crashing :P
[13:13:30] <mrvisser> but syn cookies make a lot of sense. thanks for the explanation
[13:19:02] <blindcoder> you can't really defend about a load of unintelligible data coming in
[13:19:26] <blindcoder> even if it's just random data, your inbound pipe would still be satiated, renderign your server inaccessible
[13:27:03] *** mrvisser has quit IRC
[13:28:05] *** daleg has quit IRC
[13:30:22] *** enmand has joined #smartos
[13:32:42] *** bixu has joined #smartos
[13:35:11] *** porkbelt has quit IRC
[13:36:59] *** bixu has quit IRC
[13:38:33] *** porkbelt has joined #smartos
[13:43:38] *** patkoscsaba has joined #smartos
[13:45:28] <patkoscsaba> hi guys. i'm running a virtual machine on an openindiana distribution using the kvm port. I keep getting this error messages in /var/log/messages: Mar 25 13:54:47 unknown kernel: kvm: [ID 177374 kern.warning] WARNING: kvm: emulating exchange as write. Does anyone have any idea what this means?
[13:46:55] <patkoscsaba> I tried to browse the source code but I can't find any references to this message
[13:49:25] <alcir_> so
[13:49:31] <alcir_> I want to use a nic
[13:49:38] *** noahmehl has joined #smartos
[13:49:39] <alcir_> for tagged vlan traffic
[13:49:55] <alcir_> I must forcibly assign an address also in the GZ?
[13:52:11] *** bluezenix has joined #smartos
[13:54:14] *** Teknix has joined #smartos
[14:01:13] *** dimitarivanov has joined #smartos
[14:03:23] *** AlainODea has joined #smartos
[14:05:09] <jesse_> afaik, no
[14:05:38] <alcir_> ok
[14:05:55] <alcir_> and
[14:06:05] <alcir_> I must specify the vlan in the global zone?
[14:06:23] <alcir_> I cannot define, in the GZ, only the nic
[14:06:34] <alcir_> like external=<MAC>
[14:06:44] <alcir_> and in each vm define the vlan?
[14:07:23] <jesse_> alcir_, see http://wiki.smartos.org/display/DOC/Managing+NICs
[14:07:39] <alcir_> yes, but it is not clear...
[14:07:39] <jesse_> "Exposing Additional NICs in VMs" looks what you're after?
[14:07:55] *** mrvisser has joined #smartos
[14:08:10] <alcir_> or I have problem with the switch ...
[14:08:15] <AlainODea> What is the equivalent of Ubuntu's /etc/environment (persistent environment variables) in a SmartMachine (SmartOS zone)?
[14:09:18] <jesse_> shell env? whatever startup scripts your shell runs?
[14:09:36] <mrvisser> I have the following ipfilter rules: https://gist.github.com/mrvisser/5236971   though it seems I can still ssh into the public interface (net0) on this machine (ipfilter has been restarted). Is something not right there?
[14:10:14] <jesse_> alcir_, I don't see what's not clear...=)
[14:11:50] <AlainODea> jesse_: not exactly. I'm not looking at my interactive shell, I am looking at establishing the global environment for all processes on the SmartMachine.  Is that possible?  Is it wise?
[14:12:55] *** ajcdotme has joined #smartos
[14:14:25] <jesse_> hmmm, I wonder if smartos uses environment for that
[14:14:35] *** enmand has quit IRC
[14:15:26] <jesse_> no svc:/system/environment, apparently not
[14:16:17] <jesse_> /etc/default/init
[14:16:36] *** axonpoet has quit IRC
[14:16:37] <jesse_> is the old[er] solaris location
[14:17:10] <AlainODea> jesse_: it looks like it applies.  I'll give it a shot.  Thank you :)
[14:17:33] <jesse_> looking at the file it seems to have some restriction what can be put in it
[14:18:16] <AlainODea> Indeed...
[14:18:29] <jesse_> the only thing that I normally need to set machine wide is TZ and LANG/LC_*
[14:18:45] <AlainODea> I'll stick to ~/.profile and SMF for setting environment.  Probably cleaner that way anyway.
[14:20:37] <jesse_> user-wide settings can be put into /etc/profile for ba/sh
[14:20:42] *** avrntsv has joined #smartos
[14:20:48] *** ira has quit IRC
[14:21:07] <jesse_> (and /etc/zshenv & /etc/zshrc for zsh)
[14:25:39] <AlainODea> jesse_: using standard config locations! What a novel concept.  I wish Ubuntu hadn't introduced /etc/environment.  It's more of an annoyance than a help
[14:26:00] <AlainODea> jesse_: Thank you for your help :)  /etc/profile worked perfectly
[14:26:23] <system5> AlainODea: Ubuntu Linux becomes more of a mess with each passing year
[14:26:57] <system5> AlainODea: I still don't get why Linux decided to not install 3rd party software in /opt/ so it's easy to find it
[14:27:28] <jesse_> depends on how you define 3rd party?=)
[14:28:21] <system5> jesse_: I dunno "/usr/local" just seems like a dumb idea, the more traditional /opt seemed better
[14:28:31] <jesse_> uhm...
[14:28:42] <jesse_> /usr/local is the traditional
[14:28:45] <system5> jesse_: referring to this - http://www.linuxjournal.com/magazine/pointcounterpoint-opt-vs-usrlocal
[14:28:48] <jesse_> /opt is some newfangled thing
[14:29:13] <system5> jesse_: uh, older UNIX operating systems like BSD and Solaris use /opt
[14:29:45] <MerlinDMC> I don't care about where software is installed as long as chef/puppet and the package tools know how to install/uninstall it ;)
[14:29:49] <system5> jesse_: I have an old AT&T UNIX book that was published from like 1983, I should go check if /opt is in there
[14:29:58] <jesse_> it's not
[14:30:11] <system5> jesse_: the weird thing is in the old AT&T UNIX book I think the home directories might have been in /usr instead of in /home which was odd
[14:30:18] <jesse_> it's going to tell you, among other things, that home directories go to /usr
[14:30:28] <jesse_> it's not odd at all
[14:30:30] <jesse_> just a sec
[14:31:26] <system5> jesse_: oh, so /usr/local was like BSD and /opt was some new fangled thing that SVR4 came up with, is that what you're saying?
[14:31:43] <olafm> I read this explanation for the split between / and /usr a while back and find it plausible enought: http://lists.busybox.net/pipermail/busybox/2010-December/074114.html
[14:32:18] <system5> jesse_: aha, SVR4 standard said to use "/opt"
[14:32:19] <olafm> So it was less of an organizational matter and more a "we ran out of space on /".
[14:32:24] <jesse_> system5, http://lists.busybox.net/pipermail/busybox/2010-December/074114.html
[14:32:27] <jesse_> read that first
[14:32:47] <system5> jesse_: so confirm or deny this, /opt came about as part of the SVR4 standard?
[14:32:47] <jesse_> dammit, olafm was faster
[14:33:17] * system5 reads jesse's link
[14:33:20] <mrvisser> sorry for a repost, but it seems there is much more activity now than before :)  https://gist.github.com/mrvisser/5236971  <== in these rules, I can still SSH on the net0 interface
[14:33:27] <jesse_> system5, no idea which brought it, but I think the bsds I used around 90s didn't have /opt
[14:33:38] <mrvisser> is that expected? Just want to rule out a configuration issue before I look into weirder possibilities
[14:33:43] <mrvisser> (it's ipfilter....)
[14:33:50] <system5> jesse_: it says in other places that AT&T created /opt as part of SVR4
[14:34:29] <system5> jesse_: and Solaris is based on AT&T SVR4  which explains why it traditionally uses /opt, the older BSD based SunOS wouldn't have an /opt then
[14:35:10] <system5> jesse_: I generally like the way SVR4 lays things out, which explains why I innately always preferred /opt even though I didn't know that it came about as part of UNIX SVR4
[14:35:55] <system5> jesse_: me preferring the SVR4 type of layout also explains why I idle in all these solaris related channels
[14:36:10] *** Teknix has quit IRC
[14:36:19] <jesse_> no idea what you mean
[14:36:30] <system5> jesse_: you know about the "UNIX wars" right?
[14:36:55] <system5> jesse_: there was AT&T and Sun MIcrosystems on one side and all of the other UNIX vendors (HP, IBM, etc.) on the other side
[14:37:34] *** Teknix has joined #smartos
[14:38:22] <system5> jesse_: AT&T tried to standardize what UNIX was supposed to be about with their SVR4 standard, like you can look at US government patent documents and you can see patents for Solaris commands like "pkgadd" and stuff belonging to AT&T as part of their development of SVR4 UNIX (Unix system V release 4)
[14:38:56] <system5> jesse_: like pkgadd was part of the SVR4 packaging system, and /opt was part of that standard that AT&T wrote also
[14:39:57] <system5> jesse_: there was an effor called "UNIX System Unification" described here -> http://en.wikipedia.org/wiki/UNIX_System_V
[14:40:01] <jesse_> http://objectroot.org/articles/brief-history-of-hier/
[14:40:38] <jesse_> I know of unix history
[14:41:41] <system5> jesse_: your objectroot.org article is misleading, it's not "some UNIX vendors" that implemented /opt it was AT&T and Sun Microsystems trying to redo everything from scratch to fix the mess at the very beginning of the "UNIX wars"
[14:42:34] <system5> jesse_: look at this -> http://en.wikipedia.org/wiki/Unix_wars
[14:43:39] <system5> jesse_: quote - "To increase the uniformity of Unix, AT&T and leading BSD Unix vendor Sun Microsystems started work in 1987 on a unified system... This was eventually released as System V Release 4 (SVR4)."
[14:45:14] *** ryao_ has quit IRC
[14:45:14] *** ryao_ has joined #smartos
[14:45:17] *** daleg has joined #smartos
[14:45:18] *** ryao_ is now known as ryao
[14:47:51] <system5> jesse_: I think AT&T and Sun did an somewhat ok job fixing the mess with UNIX when they came out with SVR4, but then Linux came around and had less standardization and made an even bigger mess
[14:48:13] <jesse_> oh, linux does have standard
[14:48:20] <jesse_> one (or more) per distribution
[14:48:23] <jesse_> =)
[14:49:26] <system5> jesse_: anyway, whoever came up with idea for "/opt" not sure if it was Bill Joy or Rob Pike or whoever, I think it was a good idea, if you ever set up sparse root zones in Solaris 10 with read only parts of /usr you'll know what I'm talking about
[14:50:52] <system5> jesse_: pkgadd command and the /opt directory came about from people at AT&T Bell Labs and Sun Microsystems saying- "This shit sucks and will be unusable for large enterprise / corporate customers, we need to re-standardize it in a way that makes more sense and is more usable"
[14:51:24] <system5> jesse_: other UNIX vendors like HP said- "Screw you AT&T and Sun Microsystems" and hence you have the UNIX wars
[14:51:34] *** Takimoto has joined #smartos
[14:51:52] *** mamash has joined #smartos
[14:51:57] *** mrvisser has quit IRC
[14:53:45] *** denizr has joined #smartos
[14:54:06] *** mamash has left #smartos
[14:55:10] *** Teknix has quit IRC
[14:56:35] *** Teknix has joined #smartos
[15:01:36] *** avrntsv has quit IRC
[15:05:11] <leecallen35> If my SMF service 'start' section takes 5 minutes to run, would that cause any problems?
[15:05:39] <MerlinDMC> leecallen35, is the timeout less than those 5 minutes? :P
[15:06:50] <leecallen35> I have that covered.  I just wondered whether there are any non-obvious (to a SunOS noob) consequences?
[15:07:03] *** patkoscsaba has left #smartos
[15:07:14] <jesse_> things that depend on it wait until it has started?
[15:07:57] <jesse_> (that's in the 'obvious' category, though)
[15:07:58] <leecallen35> well thats' my other question: what dependency should I set to make this run LAST? multi-user-server ?
[15:08:27] <jesse_> 'what does this need to run?'
[15:08:49] <leecallen35> it really wants some VMs to be started
[15:09:04] <leecallen35> but the script has logic to wait & check for that
[15:09:11] <leecallen35> (that's why it might take a few minutes)
[15:09:41] <jesse_> well, svc:/system/zones looks like a good start...
[15:10:23] <leecallen35> what, check to see whether that service is started?
[15:10:48] <leecallen35> or can that be coded as a dependency check in the SMF XML ?
[15:11:35] <jesse_> you can define dependencies in the xml
[15:12:29] <leecallen35> I am having a little trouble determine what are the valid dependencies I can put in the XML
[15:14:45] <jesse_> http://wiki.smartos.org/display/DOC/SMF+Manifest+Recipes
[15:15:06] <jesse_> (or the 'Using manifold' page above it on the left)
[15:17:15] <leecallen35> thanks, but, neither of those tells me what the valid values for 'dependency' are
[15:17:55] <jesse_> ...see the examples?
[15:18:00] <leecallen35> do you suppose that multi-user-server is sufficiently late in the startup process that no other services are dependent on it?
[15:18:12] <leecallen35> I mean, standard services
[15:18:28] *** axonpoet has joined #smartos
[15:18:28] <jesse_> no idea
[15:19:13] <jesse_> when I write a smf xml, I check 'what does this service need' (normally: fs, some networking, maybe a database)
[15:19:31] <jesse_> and I let smf decide when to start it up
[15:20:02] <leecallen35> but my service is going to wait until a VM starts.  I am afraid it will prevent some subsequent service from starting - potentially a service that starts VMs !
[15:20:48] <leecallen35> I will hunt around some more for definitions of those dependencies
[15:21:15] *** ipalreadytaken has joined #smartos
[15:21:24] <jesse_> things that you depend on cannot be blocked by you
[15:22:02] <jesse_> actually, if they're not up, your service start will not be called at all
[15:23:37] <leecallen35> okay, thanks
[15:25:58] *** ipalreadytaken has quit IRC
[15:26:30] *** ajcdotme has quit IRC
[15:29:12] *** ajcdotme has joined #smartos
[15:30:38] *** daleg has joined #smartos
[15:33:02] *** bixu has joined #smartos
[15:33:33] *** neophenix has joined #smartos
[15:35:17] *** sachinsharma has joined #smartos
[15:37:13] *** bixu has quit IRC
[15:49:22] *** jwk404 has quit IRC
[15:54:09] *** Teknix has quit IRC
[15:55:34] *** Teknix has joined #smartos
[15:55:36] <ajcdotme> how does smartos handle ram limits in kvm? are machines prone to overrunning into swap and crashing / throwing exception in this format?
[16:00:52] <MerlinDMC> ajcdotme, kvm runs inside a zoe which itself is limited in ram (the configured ram + 1G extra for the qemu process and stuff it might do) If you overcommit memory you might get some problems but the point where qemu is started it allocates all the ram it has configured so that is locked to the one process
[16:01:26] <ajcdotme> thanks!
[16:03:11] <MerlinDMC> also allocation is handled different on illumos than in linux ... all ram is backed by swap - there is no oom-killer (don't ask me about this I can't explain it) ;)
[16:03:18] <rmustacc_> You can't overcommit qemu dram.
[16:03:38] <rmustacc_> Memory is locked forkvm guests.
[16:03:40] <rmustacc_> *for kvm
[16:04:09] <ajcdotme> excellent. i wanted to make sure i didn't need any other protections in place before going into early production testing
[16:05:26] *** sachinsharma has quit IRC
[16:06:53] <MerlinDMC> if I have a filesystem set for a zone (lofs mount) and I zfs recv on the directory from inside the GZ will the zone see the new data that gets written or does the lofs mount stay at the old copy?
[16:08:55] <jesse_> you can't zfs recv to a lofs mount?
[16:09:24] <jesse_> ah, sorry, from inside gz...
[16:10:01] <jesse_> afaik lofs is just a mapping to the underlying fs
[16:10:11] <jesse_> so everything shows right away
[16:10:44] <MerlinDMC> that would be cool - thx
[16:10:50] *** dalethion has quit IRC
[16:11:11] *** mrvisser has joined #smartos
[16:11:15] <jesse_> (I've only copied files from zone to zone in gz, but I don't see why zfs recv would differ)
[16:17:26] *** nefilim has joined #smartos
[16:18:20] *** ajcdotme has quit IRC
[16:18:48] *** ajcdotme has joined #smartos
[16:21:42] *** ipalreadytaken has joined #smartos
[16:24:31] *** richlowe` is now known as richlowe
[16:26:01] *** ipalreadytaken has quit IRC
[16:26:09] *** marsell has joined #smartos
[16:56:58] *** alpharender has joined #smartos
[16:57:16] *** ryancnelson has joined #smartos
[17:00:21] *** denizr has quit IRC
[17:01:20] *** denizr has joined #smartos
[17:05:10] *** Teknix has quit IRC
[17:05:50] *** bluezenix has quit IRC
[17:06:35] *** Teknix has joined #smartos
[17:06:42] *** Takimoto has left #smartos
[17:11:51] *** dimitarivanov has quit IRC
[17:13:47] *** alcir_ has quit IRC
[17:14:54] *** chris--- has joined #smartos
[17:15:53] *** rmustacc_ is now known as rmustacc
[17:16:05] *** dgillespie has joined #smartos
[17:17:33] *** dap has joined #smartos
[17:19:12] *** dgillespie has quit IRC
[17:33:24] *** bixu has joined #smartos
[17:37:38] *** bixu has quit IRC
[17:38:30] *** denizr has left #smartos
[17:40:53] *** alpharender has quit IRC
[17:42:59] <vsomes_> wc
[17:43:01] *** vsomes_ has left #smartos
[17:46:13] *** bluezenix has joined #smartos
[17:50:48] *** dap has quit IRC
[17:54:11] *** Teknix has quit IRC
[17:54:27] *** szaydel has quit IRC
[17:55:36] *** Teknix has joined #smartos
[17:57:36] *** jeffpc_ is now known as jeffpc
[18:00:26] *** szaydel has joined #smartos
[18:01:36] *** ktk has joined #smartos
[18:02:06] *** psanford has joined #smartos
[18:02:56] <ktk> I'm trying to get ipv6 autoconfiguration to work on smartos, when I start in.ndpd it does discover the correct prefix but no address is assigned in that network. solaris docs couldn't give me hints what I'm doing wrong
[18:03:06] <ktk> is that different behaviour on smartos?
[18:03:47] <jesse_> ipv6 is not supported
[18:04:08] <jesse_> but it does work if you disable ip spoofing protection
[18:04:18] <jesse_> (and start in.ndpd)
[18:04:25] <ktk> ah interesting
[18:04:56] <ktk> how would I disable ip spoofing protection?
[18:05:21] <ktk> ah are you talking about a kvm host?
[18:05:26] <ktk> I mean the global zone itself
[18:05:37] <jesse_> I'm talking about a zone
[18:05:42] <ktk> ok
[18:05:43] <jesse_> and same applies to kvm instances
[18:05:47] *** jim80net has joined #smartos
[18:05:51] <ktk> I don't get it in the global zone either
[18:05:59] <ktk> which is confusing me a bit
[18:06:04] <jesse_> let's see....
[18:06:23] <jesse_> in a zone,
[18:06:24] <jesse_> ipadm create-addr -t -T addrconf net0/v6
[18:06:24] <jesse_> /usr/lib/inet/in.ndpd
[18:07:30] <jesse_> and, ofcourse, echo '{"update_nics": [{"mac": "<mac>", "allow_ip_spoofing": true}]}' | vmadm update <uuid> in the gz
[18:08:05] <jesse_> (should probably have used smf to start in.ndpd there, though=))
[18:08:15] <ktk> right that one I found. letme try ipadm
[18:08:39] <ktk> net0 is interface name?
[18:08:45] <jesse_> yes
[18:09:17] <jesse_> that was the only interface in the zone -> net0
[18:09:18] <ktk> and /v6 makes sure I don't screw up my v4 config?
[18:09:41] <ktk> tried assigning it manually last week, that didn't end well :-D
[18:09:47] *** melinko-CR24 has joined #smartos
[18:09:49] <jesse_> it shouldn't=)
[18:10:19] *** Melinko has quit IRC
[18:10:52] <jesse_> you should try it first in a zone?
[18:11:04] <ktk> ah interesting, didn't think of htat
[18:11:11] <ktk> I'm pretty new to solaris so :)
[18:11:12] <jesse_> after you see it working there, you can try it in gz
[18:11:17] <ktk> good idea
[18:11:46] <jesse_> zones are expendable=)
[18:13:07] <ktk> jesse_: thanks for the hint, will try later
[18:13:23] *** leecallen35 has quit IRC
[18:13:43] *** leecallen has joined #smartos
[18:16:45] *** dysinger has joined #smartos
[18:20:16] <ktk> jesse_: hmm the ip spoofing hint fixed the issue that my ip aliases didn't work, thanks for the hint :-D
[18:23:30] <ryancnelson> an ethernet frame that's carrying a tcp packet with a src address that doesn't match your allowed-ips list will be dropped, unless you've turned off anti-spoofing
[18:23:36] *** Trixboxer has joined #smartos
[18:24:00] <Trixboxer> Hi, can I do an Active-Passive bonding in smartos ?
[18:24:05] <ryancnelson> that apparently means ipv6, too.  and since there is no ipv6 address in the allowed-list , you need to turn that off
[18:24:11] <ktk> ryancnelson: yeah I had the config on "dhcp" and assumed it will work that way
[18:25:12] <ryancnelson> "dhcp" should also imply anti-spoofing is off
[18:25:32] <ktk> definitely didn't work with dhcp only
[18:25:52] <ktk> I saw arp who-has and the kvm host sent is-at but they never made it through
[18:27:09] *** bluezenix has quit IRC
[18:27:55] *** dap has joined #smartos
[18:28:03] <ktk> ah ipadmin is persistent
[18:28:04] <ktk> that's neat
[18:28:09] <ktk> ipadm I mean
[18:29:02] *** mrvisser has quit IRC
[18:29:19] *** ryancnelson has left #smartos
[18:32:10] <jesse_> ktk, I thought that -t was for 'temporary'
[18:32:22] <jesse_> I was about to write about it
[18:32:23] <ktk> jesse_: in terms of I can set it persistent
[18:32:28] *** bens1 has quit IRC
[18:37:15] *** enmand has joined #smartos
[18:42:38] *** sachinsharma has joined #smartos
[18:45:59] *** ipalreadytaken has joined #smartos
[18:47:36] *** sachinsharma has quit IRC
[18:53:21] *** ahaydock has joined #smartos
[18:55:52] *** bixu has joined #smartos
[18:57:26] *** ajcdotme has quit IRC
[18:58:09] *** ajcdotme has joined #smartos
[19:05:11] *** Teknix has quit IRC
[19:06:36] *** Teknix has joined #smartos
[19:18:01] *** ira has joined #smartos
[19:19:20] *** Teknix has quit IRC
[19:21:06] *** KermitTheFragger has quit IRC
[19:21:20] *** ajcdotme has quit IRC
[19:21:49] *** Teknix has joined #smartos
[19:23:17] *** denizr has joined #smartos
[19:26:28] *** ipalreadytaken has quit IRC
[19:28:29] *** ipalreadytaken has joined #smartos
[19:29:57] *** avrntsv has joined #smartos
[19:32:40] *** ira has quit IRC
[19:34:05] *** ajcdotme has joined #smartos
[19:37:31] *** ryancnelson has joined #smartos
[19:41:06] *** ajcdotme_ has joined #smartos
[19:41:35] *** mAniAk-_- has quit IRC
[19:43:43] *** ajcdotme has quit IRC
[19:45:07] *** mAniAk-_- has joined #smartos
[19:46:26] *** mamash has joined #smartos
[19:46:45] *** ipalreadytaken has quit IRC
[19:50:40] *** Teknix has quit IRC
[19:50:49] *** bluezenix has joined #smartos
[19:52:04] *** Teknix has joined #smartos
[19:55:59] *** ajcdotme has joined #smartos
[19:58:28] *** ajcdotme_ has quit IRC
[20:00:40] *** xmerlin has joined #smartos
[20:04:47] *** noahmehl has quit IRC
[20:11:10] *** Gruu_ has joined #smartos
[20:12:29] *** mamash has left #smartos
[20:16:33] <ajcdotme> greetings. in /usbkey/config i see admin_network=...
[20:16:37] <ajcdotme> what does this imply?
[20:18:12] <ryancnelson> "imply"?
[20:19:24] <ajcdotme> should it be set to a network? what does "..." mean here?
[20:19:43] *** Gruu_ has quit IRC
[20:19:50] <ajcdotme> apologies for what is probably a newb question, but it seems documentation on the subject is somewhat lacking
[20:20:09] <ryancnelson> it means "network address"... as in "broadcast address, network address, subnet"
[20:20:21] <ryancnelson> "..." means "calculate that normally"
[20:20:59] <ajcdotme> alright, so the only reason to specify it any further there is if the subnet/mask don't align normally?
[20:21:11] <ajcdotme> is that the case for other created networks as well?
[20:21:27] <ryancnelson> if your network is 10.33.44.0/24 , your network address is probably 10.33.44.0 ... yeah, just let that be, 99.99% of the time
[20:21:36] <ajcdotme> great. thanks much.
[20:26:13] *** mamash has joined #smartos
[20:28:11] *** avrntsv1 has joined #smartos
[20:31:40] *** avrntsv has quit IRC
[20:32:10] *** avrntsv1 has quit IRC
[20:35:26] *** wramthun has joined #smartos
[20:35:28] *** AlainODea has quit IRC
[20:52:12] *** enmand_ has joined #smartos
[20:54:36] *** enmand has quit IRC
[20:56:36] *** szaydel has quit IRC
[20:57:32] *** szaydel has joined #smartos
[21:09:30] *** enmand has joined #smartos
[21:11:41] *** Teknix has quit IRC
[21:11:55] *** enmand_ has quit IRC
[21:13:23] *** Teknix has joined #smartos
[21:31:42] *** Trixboxer has quit IRC
[21:32:56] *** enmand has quit IRC
[21:39:12] *** opeth___ has quit IRC
[21:39:53] *** noahmehl has joined #smartos
[21:49:21] *** enmand has joined #smartos
[21:59:16] *** Verilium has quit IRC
[22:00:17] *** ira has joined #smartos
[22:02:55] *** kfr_ has joined #smartos
[22:08:59] <kfr_> So ? Chef or Puppet?
[22:09:32] <wesolows> vi or emacs? OpenBSD or NetBSD? Coke or Pepsi?
[22:11:15] <ira> vi, NetBSD, Coke.
[22:12:32] <arai> Well, it's been fun
[22:17:02] <ryancnelson> gold star for ira
[22:17:27] *** tonyarkles has joined #smartos
[22:17:43] <kwork> vi, openbsd, coke
[22:17:52] <ira> ryancnelson: Thanks ;0
[22:17:54] <ira> ;)
[22:18:16] <jperkin> there was a talk on ansible at pkgsrccon, it looks like the most reasonable config mgmt option to me
[22:18:26] *** schw3rt has joined #smartos
[22:19:11] <schw3rt> my zones can ping each others but they can not access the rest of the network, did I miss something ?
[22:19:18] <jperkin> I'm just waiting for someone to write a hackers one which requires ed(1) scripts instead of custom 'if blah replace with foo'
[22:20:10] <wesolows> python :-(
[22:20:12] *** denizr has quit IRC
[22:20:15] <wesolows> still better than ruby though
[22:20:17] <ryancnelson> schw3rt: what's your config file look like?  are you using tagged vlans?  can they ping their gateway?
[22:20:33] *** denizr has joined #smartos
[22:20:38] <ryancnelson> "yes", you obviously missed something.  let's figure out what. :)
[22:20:40] <ira> wesolows: Why no Python?
[22:20:41] <schw3rt> no vlans and no they cant ping anything
[22:21:12] <jperkin> wesolows: right, was the main drawback I saw
[22:21:29] <wesolows> ira: the simplest and least controversial response is get off my lawn.
[22:21:30] <ryancnelson> schw3rt: you need to tell us *something* about how you have things set up
[22:21:34] *** enmand has quit IRC
[22:21:34] <schw3rt> http://dpaste.org/uMwRH/
[22:21:39] <jperkin> it has pkgin support, which is nice
[22:22:05] <ira> wesolows: I can respect that… But you didn't seem the perl type ;)
[22:22:11] <ryancnelson> that's your zone.  i don't care about that.   what's your /usbkey/config file look like?
[22:22:13] <wesolows> I'm not.  We don't even use perl.
[22:22:33] <wesolows> it exists in the gate solely to break us.
[22:22:37] <ryancnelson> you're not in virtualbox or something crazy like that, correct?
[22:23:28] <schw3rt> vmware host
[22:23:31] <schw3rt> http://dpaste.org/ATO8t/
[22:23:40] <ira> wesolows: And node is ok? :)  (Making sure I understand 100%)
[22:24:18] *** mamash has left #smartos
[22:25:17] <ryancnelson> schw3rt: so, your headnode is getting *its* IP from dhcp...
[22:25:32] <ryancnelson> but your vm's are specifying an ip address
[22:25:37] <schw3rt> yes, static on the dhcp server
[22:26:18] <ryancnelson> ... but those vm's are creating vnics on your admin nic.
[22:26:34] <wesolows> ira: I sometimes tolerate node.  At least it has C-like syntax.
[22:26:36] <ryancnelson> does your headnode have an IP in 192.168.222.x?
[22:26:42] <schw3rt> yes
[22:26:43] <wesolows> Really I'd just rather write C.
[22:27:13] <ira> wesolows: Ok… At least you are consistent. :)  As one of my co-workers so eloquently put it: "They all suck."
[22:27:22] *** wolfeidau has quit IRC
[22:27:37] <ryancnelson> but they can't ping 192.168.222.254, right?  from inside the vm
[22:27:42] <schw3rt> correct
[22:28:02] <ryancnelson> vmware fusion, or somethign like esx?
[22:28:02] *** opeth__ has joined #smartos
[22:28:07] <opeth__> whee
[22:28:08] <schw3rt> workstation
[22:28:49] <opeth__> it's been quite a ride to construct my zones pool the way I wanted it to be, but it's up
[22:29:45] <schw3rt> does the headnode needs promiscous mode ?
[22:29:59] <ryancnelson> sort of...
[22:30:06] <ryancnelson> well, it doesn't, but vmware probably does
[22:30:13] <ryancnelson> vnics have their own mac addresses
[22:30:23] <ryancnelson> what does
[22:30:23] <ryancnelson> dladm show-bridge
[22:30:24] <ryancnelson> say?
[22:30:36] <ryancnelson> in the GZ
[22:30:49] <schw3rt> vmwarebr    stp     32768/0:c:29:5a:15:a4 32768 32768/0:c:29:5a:15:a4
[22:31:32] <ryancnelson> well, that's good at least
[22:32:28] <ryancnelson> you might try running "snoop" in the background, to force the emulated vmware nic into permisc mode.  (do it in "screen" or something, and background it)
[22:32:54] <ryancnelson> but vmware workstation isn't something i've tested with, only vmware fusion on a mac
[22:35:32] <schw3rt> ok, it was the promiscous mode on vmware workstation
[22:35:38] <ryancnelson> your vnics have their own mac addrs, which are then on your vmware's virtual nic, which are then on your actual nic...
[22:35:42] <ryancnelson> oh, that's a setting?
[22:36:11] <schw3rt> you need to change some rights on the adapter on linux hosts
[22:36:46] <ryancnelson> a good read is /lib/svc/method/net-physical ... search for "vmwarebr"
[22:36:47] *** opeth__ has quit IRC
[22:36:58] <ryancnelson> ... it's a hack to make working in vmware work.
[22:37:02] <ryancnelson> so you're working now?
[22:37:11] <schw3rt> yes
[22:37:16] <ryancnelson> cool
[22:37:21] <schw3rt> thank you for the troubleshooting
[22:39:52] *** opeth__ has joined #smartos
[22:44:45] *** wolfeidau has joined #smartos
[22:47:11] *** neophenix has quit IRC
[22:52:39] *** bluezenix has quit IRC
[22:56:58] <kfr_> in the kvm json file, how can I specify the domain that dhcpclient will use in the kvm guest?
[22:57:28] *** denizr has quit IRC
[22:57:41] <kfr_> I mean in the linux kvm I want the search in /etc/resolv.conf to be populated
[22:57:45] <rmustacc> kfr_: You cannot right now.
[22:58:19] <kfr_> Should I added it as an issue?
[22:58:36] <rmustacc> It already is.
[22:58:39] <kfr_> kewl
[22:58:57] <rmustacc> https://github.com/joyent/illumos-kvm-cmd/issues/14
[23:00:35] *** ktk is now known as ktkNA
[23:02:14] <kfr_> rmustacc thanks!
[23:02:20] <kfr_> I'm  now watching it
[23:06:32] <rmustacc> It's pretty low priority, so I wouldn't expect anything any time soon.
[23:09:11] <kfr_> I can work arround it.
[23:10:25] *** axonpoet has quit IRC
[23:10:25] <kfr_> http://www.cyberciti.biz/faq/dhclient-etcresolvconf-hooks/ seems to have a couple ways to work arround it.
[23:11:52] *** szaydel has quit IRC
[23:13:11] *** deirdres has joined #smartos
[23:20:17] <wesolows> window 2
[23:20:20] <wesolows> guh
[23:22:44] *** AlainODea has joined #smartos
[23:26:42] *** jim80net has quit IRC
[23:56:19] *** nikolam has joined #smartos
top

   March 25, 2013  
< | 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 | >