December 1, 2012 
< | 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 | >
December 1, 2012 [00:07:32] <konobi> kfr: there's always usbip[00:08:09] *** jim80net has quit IRC[00:10:25] *** miine__ has quit IRC[00:11:28] *** szaydel has quit IRC[00:11:41] *** jamesd has joined #smartos[00:14:51] *** kfr has quit IRC[00:21:31] *** Vod has quit IRC[00:39:01] *** jamesd has quit IRC[01:05:48] *** szaydel has joined #smartos[01:06:32] *** LucidDreamZzZ has quit IRC[01:07:50] *** LucidDreamZzZ has joined #smartos[01:09:28] *** ryancnelson has quit IRC[01:19:29] *** szaydel has quit IRC[01:24:24] *** ryancnelson has joined #smartos[01:28:46] *** szaydel has joined #smartos[01:40:50] *** jamesd has joined #smartos[02:04:53] *** izaki has joined #smartos[02:19:24] *** z0mbix has quit IRC[02:24:38] *** newlix has joined #smartos[02:32:44] *** cncfanatics1 has quit IRC[02:44:14] *** newlix has quit IRC[03:04:23] *** arekinath has quit IRC[03:05:52] *** dap has quit IRC[03:07:32] *** ryancnelson has quit IRC[03:08:03] *** arekinath has joined #smartos[03:12:17] *** trentm has quit IRC[03:13:26] *** ira has quit IRC[03:40:17] *** newlix has joined #smartos[03:44:05] <yofuh> any idea why gcc47 in smartos eats up PATH_MAX?[03:44:12] <yofuh> echo "#include <limits.h>\nPATH_MAX" |cpp[03:53:42] *** wdent has quit IRC[03:54:11] *** wdent has joined #smartos[03:58:29] *** wdent has quit IRC[04:00:14] *** rcamera has quit IRC[04:00:31] *** rcamera has joined #smartos[04:04:37] *** rcamera has quit IRC[04:05:53] *** szaydel has quit IRC[04:10:13] *** ryancnelson has joined #smartos[04:11:11] *** ryancnelson has quit IRC[04:12:01] *** trentm has joined #smartos[04:15:12] *** viridari has joined #smartos[04:15:12] *** viridari has joined #smartos[04:24:11] *** ryancnelson has joined #smartos[04:29:41] *** ryancnelson has quit IRC[04:48:18] *** notmatt has joined #smartos[04:51:53] *** chriss has joined #smartos[04:53:11] *** trentm has quit IRC[04:55:11] *** chriss has quit IRC[05:01:51] *** haydock__ has joined #smartos[05:01:59] *** haydock_ has quit IRC[05:08:09] *** notmatt has joined #smartos[05:12:42] <LeftWing> Try swapping echo for printf, there[05:12:48] <LeftWing> yofuh: ^^[05:16:26] *** viridari has quit IRC[05:42:57] *** szaydel has joined #smartos[06:46:19] *** dap has joined #smartos[07:32:07] *** e^ipi_laptop has quit IRC[07:53:30] *** szaydel has quit IRC[07:58:00] *** dap has quit IRC[08:10:05] *** jamesd has quit IRC[08:37:01] *** haydock___ has joined #smartos[08:37:03] *** haydock__ has quit IRC[09:08:33] *** elijah-m_ has joined #smartos[09:10:39] *** elijah-mbp has quit IRC[09:18:00] *** avrntsv has joined #smartos[09:22:37] *** texarcana has quit IRC[09:23:51] *** texarcana has joined #smartos[09:32:27] <jperkin> someone was trying to compile gcc 4.7.2 manually the other day, just to say that pkgsrc is now updated to 4.7.2[10:04:01] *** fsteinelX has joined #smartos[10:09:14] *** fsteinelX is now known as fsteinel[10:11:21] *** avrntsv has quit IRC[10:38:58] *** arekinath has quit IRC[10:39:16] *** arekinath has joined #smartos[10:39:16] *** arekinath has joined #smartos[10:45:39] *** bahamas10 has quit IRC[10:45:39] *** papertigers has quit IRC[10:45:41] *** matthewpucc has quit IRC[10:45:42] *** michal` has quit IRC[10:45:43] *** disharmony has quit IRC[10:45:44] *** z has quit IRC[10:45:45] *** joshie has quit IRC[10:45:45] *** richlowe has quit IRC[10:45:46] *** rodgort has quit IRC[10:45:48] *** fly__ has quit IRC[10:47:56] *** fsteinel has quit IRC[10:48:06] *** michal` has joined #smartos[10:48:06] *** bahamas10 has joined #smartos[10:48:07] *** papertigers has joined #smartos[10:48:07] *** matthewpucc has joined #smartos[10:48:07] *** disharmony has joined #smartos[10:48:07] *** z has joined #smartos[10:48:07] *** joshie has joined #smartos[10:48:08] *** richlowe has joined #smartos[10:48:08] *** rodgort has joined #smartos[10:48:08] *** fly__ has joined #smartos[11:12:53] *** mamash has joined #smartos[11:19:03] *** fsteinelX has joined #smartos[11:19:22] *** fsteinelX is now known as fsteinel[11:31:17] *** defa has joined #smartos[11:44:25] *** cncfanatics has joined #smartos[12:16:00] *** sachinsharma has joined #smartos[12:16:24] *** kevinykchan has quit IRC[12:22:33] *** kevinykchan has joined #smartos[12:25:38] *** z0mbix has joined #smartos[12:36:11] <jesser> i have a ssh PATH problem. adding /opt/local/bin to PATH in /etc/profile does not seem to be working. That is, doing "ssh foo env" shows that /opt/local/bin has not been added to PATH. How does one do this on a smartos os vm? (joyent brand)[12:37:20] <mamash> set PATH in ~/.ssh/environment?[12:37:53] <mamash> should be there by default if you created a user using the default skel dir[12:38:01] <jesser> yeah i was hoping to have it system wide[12:38:26] <jesser> i guess sshd is compiled to not load /etc/profile[12:38:45] <jesser> good to know about the default skel dir doing this though, thanks.[12:43:03] *** z0mbix has quit IRC[12:51:03] *** avrntsv has joined #smartos[13:08:00] *** nikolam has joined #smartos[13:24:35] *** sachinsharma has quit IRC[13:44:39] *** miine has joined #smartos[14:08:05] *** avrntsv has quit IRC[14:53:50] *** mamash has left #smartos[15:04:45] *** nikolam has quit IRC[15:07:49] *** viridari has joined #smartos[15:07:49] *** viridari has joined #smartos[15:10:43] *** nikolam has joined #smartos[15:34:24] *** jamesd has joined #smartos[15:38:00] *** jamesd_ has joined #smartos[15:39:49] *** szaydel has joined #smartos[15:40:42] *** jamesd has quit IRC[16:05:16] *** miine has quit IRC[16:23:17] *** mamash has joined #smartos[16:48:08] *** nikolam has quit IRC[16:48:26] *** mamash has left #smartos[16:55:29] *** kevinykchan has quit IRC[16:58:34] *** kevinykchan has joined #smartos[17:01:25] *** tonyarkles has joined #smartos[17:05:12] *** szaydel has quit IRC[17:08:47] *** szaydel has joined #smartos[17:10:29] *** szaydel has quit IRC[17:21:28] *** dap has joined #smartos[17:21:43] *** tonyarkles has quit IRC[17:27:09] *** wdent has joined #smartos[18:02:02] *** wdent has quit IRC[18:02:34] *** wdent has joined #smartos[18:07:18] *** wdent has quit IRC[18:26:23] *** tdmackey has quit IRC[18:26:33] *** tdmackey has joined #smartos[18:31:56] *** CostasGR has joined #smartos[18:34:51] *** CostasGR has left #smartos[18:39:01] *** jamesd_ has quit IRC[18:39:55] *** jamesd_ has joined #smartos[19:11:20] *** mamash has joined #smartos[19:13:06] *** wdent has joined #smartos[19:17:35] *** wdent has quit IRC[19:22:39] *** Kush- has left #smartos[19:35:26] *** Saskaloon has quit IRC[19:42:45] *** abhishekvasisht has joined #smartos[19:53:55] *** mamash has left #smartos[20:02:50] *** Saskaloon has joined #smartos[20:12:33] *** abhishekvasisht has quit IRC[20:31:32] *** ryancnelson has joined #smartos[20:36:07] *** ryancnelson has quit IRC[20:58:43] *** wdent has joined #smartos[20:59:06] *** trentster has quit IRC[21:08:12] *** chg has joined #smartos[21:14:59] *** tonyarkles has joined #smartos[21:22:25] <chg> Licenser: just installed fifo 0.3 alpha - however, no hypervisor shows up in wiggle-ui.[21:22:36] <Licenser> hi chg :)[21:23:26] <Licenser> try the following: put the hypervisors hostname/ip in the /etc/hosts of the fifo zone[21:23:57] <chg> great: you are around! hypervisor has set static hostname according to http://wiki.smartos.org/display/DOC/Setting+a+static+hostname+at+boot+time; could this interfere communication with fifo zone?[21:24:34] <Licenser> could - should not but could[21:24:52] <Licenser> we can go through a few things to find out what causes the problem if you have the time[21:25:26] *** defa has quit IRC[21:29:20] <chg> just put hypervisor hostname/ip in etc/hosts and rebooted fifo vm. still no hypervisor in wiggle-ui[21:29:31] <Licenser> okay :)[21:29:32] *** trentster has joined #smartos[21:29:47] <Licenser> the zone and hypervisor are on the same physical host right?[21:29:54] <chg> yes[21:29:55] *** defa has joined #smartos[21:30:05] <Licenser> okay, does the zone have more then one network interface?[21:30:25] <chg> no, both hypervisor and vm are in the same subnet[21:30:36] <Licenser> okay, this is strangelish[21:30:54] <Licenser> can you on the hypervisor do: /opt/chunter/bin/attach attach[21:31:01] <Licenser> then type: libsniffle:servers().[21:32:03] <chg> /opt/chunter/bin/attach attach[21:32:03] <chg> bash: /opt/chunter/bin/attach: No such file or directory[21:32:43] <Licenser> sorry ../bin/chunter attach[21:33:01] *** z0mbix has joined #smartos[21:35:39] <chg> okay: i get Attaching to /tmp/chunter (^D to exit) ; could you please repeat the libsniffle statement - I got it with a smiley in it[21:35:47] *** wdent has quit IRC[21:36:01] <Licenser> oh darn[21:36:12] <Licenser> libsniffle:servers followed by ( then ) then .[21:36:15] *** wdent has joined #smartos[21:37:51] <MerlinDMC> we all love smileys :)[21:38:29] *** trentster has quit IRC[21:38:59] <Licenser> we do![21:39:30] <Tekni> I've installed 0.3 alpha also. It doesn't seem to do much yet. :)[21:40:44] *** wdent has quit IRC[21:40:48] <chg> I killed the "chunter attach" process with ^D, now I get "Node is not running", even after trying to restart with "sh chunter-latest"[21:40:57] <Licenser> Tekni it does a lot - propably just things that are not yet visible :) working on a second push right now[21:41:03] <Licenser> chg do svcs -x[21:41:10] <Licenser> chunter should show up in there if it's not running[21:41:19] <Tekni> I didn't want to pass up an opportunity to use a tactfully placed smiley is all[21:41:20] <Licenser> which would be a good reason why it's not showing up :)[21:41:31] <Licenser> Tekni hehe[21:41:46] <chg> svc:/network/chunter:default[21:43:14] <Licenser> can you post the log of that?[21:43:41] <chg> svcadm restart didnt help; log follows[21:43:59] *** notmatt has quit IRC[21:44:06] <chg> [ Dec 1 18:40:43 Executing start method ("/opt/chunter/bin/chunter start"). ][21:44:07] <chg> [ Dec 1 18:40:43 Method "start" exited with status 0. ][21:44:07] <chg> [ Dec 1 20:36:50 Stopping because all processes in service exited. ][21:44:07] <chg> [ Dec 1 20:36:50 Executing stop method ("/opt/chunter/bin/chunter stop"). ][21:44:07] <chg> Node 'chunter at 192 dot 168.1.184' not responding to pings.[21:44:07] <chg> [ Dec 1 20:36:50 Method "stop" exited with status 1. ][21:44:07] <chg> [ Dec 1 20:36:50 Executing stop method ("/opt/chunter/bin/chunter stop"). ][21:44:08] <chg> Node 'chunter at 192 dot 168.1.184' not responding to pings.[21:44:08] <chg> [ Dec 1 20:36:51 Method "stop" exited with status 1. ][21:45:11] *** newb_ has joined #smartos[21:45:56] <Licenser> oi for the sake of people here please use something like refheap.com, gits o[21:46:05] <chg> by the way: smiley was in between libsniffle and ervers[21:46:06] <Licenser> r any other pastebin out there :)[21:46:21] <Licenser> oh didn't knew that :s is a smilie -.-[21:46:31] <Licenser> okay another thing: /opt/chunter/bin/chunter console[21:47:44] <chg> great thanks for clarification of the smiley -.-[21:48:03] <chg> what to put in the chunter console?[21:48:53] <Licenser> just see if it starts :)[21:49:01] <Licenser> if it does quit it with q().[21:49:24] <chg> it successfully started - so turning back to the fifo vm?[21:49:27] <Licenser> if it works clear the service with svcadm clear chunter[21:52:14] *** defa has quit IRC[21:54:29] *** pstoll|away has joined #smartos[21:54:30] *** pstoll|away is now known as pstoll[21:57:08] <newb_> I'm following the wiki on managing nics and vmadm but can't get a zone assigned to another phys nic & vnic. I edit usbkey/config with the external_nic=macaddr, reboot, create a addnic.json file with the external nic tag, physical interface e1000g1 and the static ip netmask etc. When I update addnic.json to the vmadm uuid and then get the vmadm uuid info back it shows the interface as net1. After I restart the zone ifconfig shows no ne[21:58:29] <chg> chunter running again - then started "chunter attach" and put libsniffle.servers() - what should I get; I actually get nothing ;)[22:01:08] <newb_> I want to connect wan to a physical eth port on the host and assign a vnic to a zone with a wan public static ip. According to wiki I believe I am performing the config correctly. Is there anything else I need to do inside the zone to get it to acquire the vnic and vnic ip settings?[22:03:57] *** szaydel has joined #smartos[22:04:32] <Licenser> chg any luck now?[22:05:42] <chg> still twiddleing - calling back later - thanks so far![22:06:09] <Licenser> kk :) glad I could help[22:11:47] <chg> Licenser: chunter process died again - /var/svc/log/network-chunter:default.log follows:[22:11:56] <Licenser> oi[22:12:10] <Licenser> can you package up the folder /var/log/chunter and mail it to me?[22:12:34] <chg> yes - where to mail it?[22:13:06] * Licenser points to the query :)[22:13:56] <Licenser> I'll dig through the logs and try to find the bugger[22:14:27] *** notmatt has joined #smartos[22:14:55] *** notmatt has quit IRC[22:15:09] *** notmatt has joined #smartos[22:22:52] *** szaydel has quit IRC[22:23:21] <Licenser> chg from the first glance it looks like a communicaiton problem either your zone can't reach the hypervisor or visa versa[22:24:37] <Licenser> hmm strange[22:25:18] * Licenser researches a bit on[22:26:01] <chg> just wait - seems you put me on the right track[22:27:59] *** jim80net has joined #smartos[22:28:30] <Licenser> I found the bug :)[22:29:08] <Licenser> the problem was a very sneaky one[22:29:17] <Licenser> it manifested because you had more VM's then my test system :P[22:34:23] <konobi> newb_: which distro?[22:35:17] *** trentster has joined #smartos[22:35:31] <newb_> smartos with plus zone.[22:36:47] <konobi> newb_: and vmadm get <uuid> is showing the nics all fine?[22:37:49] <Licenser> chg it's not a problem you can solve :) well you could remove a few zones[22:38:39] <Licenser> the bug is in a library, causing a crash in zone registration, it isn't causing any trouble with a low number of zones since the system recovers but you seem to have hit the treshold in which is goes bad fast :P[22:38:49] <chg> Licenser: kind of workaround I don't like too much ;) - beter I'm trying again after you commited the fix[22:39:30] <Licenser> chg will take a hour or so since it goes together with a bit of cleanup I'm doing right now[22:39:53] <Licenser> actually just grab the newest chunter build - might fix it already[22:41:37] <chg> not sure, if I am trying later, but I surely will tomorrow, will turn back to you then - thanks again for immediately digging into it :)[22:43:48] <Licenser> ah thank you for catching this ;)[22:43:54] *** trentster has quit IRC[22:45:04] <konobi> still in erlang, eh?[22:45:16] <Licenser> konobi ayup[22:45:19] * Licenser loves it[22:46:43] *** wdent has joined #smartos[22:47:24] <newb_> konobi: sorry for delay. vmadm get looks like the nics are fine.[22:48:26] <konobi> newb_: did you mark one of them as primary?[22:48:42] <newb_> the admin one looks marked as primary[22:48:48] <konobi> Licenser: surprised you haven't just tried updating the REST api in vmadmd[22:49:09] <konobi> newb_: should be fine then[22:49:15] <newb_> ive removed the admin nic and setup the external in its place and the zone does not see it.[22:49:20] <newb_> the zone just sees the l0[22:49:32] <Licenser> konobi there is a rest api in vmadmd? o.O[22:49:34] <konobi> you changed the mac?[22:49:42] <konobi> Licenser: yup... check the manpage[22:50:01] <newb_> yes the mac is changed to the vnic mac. i am probably overlooking some detail. i am trying to create some more nics and troubleshoot this.[22:50:20] <konobi> newb_: nopaste your json[22:50:34] <newb_> one moment[22:50:40] <Licenser> konobi neat, didn't know that but well it wouldn't go all the way in regards of what chunter does[22:51:17] <Licenser> but that's pretty cool :)[22:51:26] *** wdent has quit IRC[22:51:45] *** tonyarkles has quit IRC[22:51:47] <konobi> you could probably move much of what vmadm does into vmadmd since it's the same codebase[22:52:24] <konobi> newb_: oh[22:52:35] <konobi> newb_: have you run `sysinfo -f -u` ?[22:53:30] <Licenser> ture, still even vmadm doesn't do all the things of chunter, essecial parts of cause :) half of chunter is calling vmadm :P, but things like watching for zone changes, mdns service discoversy etc would make no sense to place there[22:54:09] <Licenser> but but it might be a good thing to investigate into in replacing calls to vmadm with rest calls to vmadmd[22:54:57] <newb_> konobi: just ran that. in the GZ i get no print out. in the zone i get an error.[22:57:56] <konobi> Licenser: https://github.com/orlandov/node-sysevent[22:58:22] <konobi> newb_: `sysinfo` do you see the nic_tags now?[22:59:51] <konobi> in the gz that is[22:59:54] <newb_> konobi: yes in the GZ i see the nic tag i named external[23:00:08] <konobi> reboot yer zone and see if it shows up[23:00:09] <konobi> =0)[23:00:13] <newb_> konobi: but the status shows unknown, not up. thats probably the problem[23:00:27] <newb_> even though the cable is connected and nic leds are lit up[23:00:50] <konobi> newb_: yeah... there's a page on the wiki about fixing that scenario[23:00:59] <Licenser> konobi sure but I'd have to do that there too ;)[23:01:24] <newb_> do you remember which wiki? i will look for it but havent ran across it yet.[23:01:43] <Licenser> plus as odd as it sounds - I found it easyer to get an erlang app to run in the GZ then a node app[23:01:49] <konobi> basically, while you need to edit the usbkey/config, you also need to update the sysinfo[23:04:26] <newb_> ok. quick question about these nics... when i add this external nic to usbkey/config and expose a vnic thru it to a zone is this phys and vnic exposing anything into the global zone? would i be foolish not to run ipf on this external interface? i will learn to setup ipf anyway but want to know whats being exposed by the external nic.[23:04:41] <konobi> Licenser: well, you can always write small things in node and then wrap them in a pipe for erlang[23:04:59] <konobi> newb_: no... a vnic is a L2 device[23:05:52] <konobi> newb_: so ie... the only thing exposed with be the zone[23:06:09] <Licenser> could do that true[23:06:23] *** jesser_ has joined #smartos[23:06:24] <konobi> unix philosophy ftw[23:06:43] <Licenser> konobi I do that already in some part :) like calling dtrace stuff, or vmadm[23:06:48] <newb_> ok. is it still advisable to run ipf? or what scenarios necessitate ipf?[23:06:59] <konobi> newb_: when you have an IP[23:07:00] <Licenser> but getting things like node-sysevent to run in the GZ isn't fun when you don't controll the image[23:07:25] <konobi> huh, i thought it was already in there somewhere[23:07:47] <Licenser> konobi not sure, I just had a runin with node_dtrace in that regard was ugly[23:08:07] <newb_> i intend on providing the zone a public wan ip. so i would run ipf for the vnic in the GZ or ipf in the wan exposed zone?[23:08:10] <konobi> node_dtrace is somewhat of an exception i'd think[23:08:22] <konobi> newb_: wherever the IP is[23:08:33] <newb_> thanks.[23:08:40] <Licenser> konobi problem was mostly the fact that it had C(++) code to compile and it had to be compiled with the RIGHT node version (the on on the hypervisor) to work correctly[23:08:53] *** jesser has quit IRC[23:08:53] *** jesser_ is now known as jesser[23:09:03] <konobi> Licenser: yup, that's true[23:09:22] <Licenser> with my lerang stuff it's not a issue since the package application bundles the erlang it was build wiht[23:09:38] <Licenser> sure you can argue if that's good or not :P but it solves this problem at least[23:09:41] <konobi> Licenser: you could always do the same with node[23:09:59] <konobi> but, you could ship a binary version of the module based on what you've built with[23:10:43] <Licenser> I know but again, that would mean extra effort with minor payoff. Since the entire FiFo system is build in erlang, buiding addiotnal components in erlang has big synergy effects[23:11:00] <konobi> actually i've been tempted to write a binary bundler[23:11:06] <Licenser> ^^[23:11:28] <Licenser> I find it a very nice principle, it holds a ton of advantages with the only downside being bundled apps are kind of big[23:11:40] <Licenser> ~20mb in the case of fifo's components[23:11:54] <Licenser> well that and that I think the smart shared library loading stuff isn't kicking in there[23:13:03] *** robins has joined #smartos[23:13:21] <Licenser> I guess it's pretty much the same reason why you folks write close to everything in node (the synergy effects that is)[23:14:39] <konobi> yeah... though everything's pretty small and concise[23:15:37] <Licenser> I try to do the same - even so don't succeed everywher :P sniffle is quite a beast[23:18:00] <Licenser> in the end it's the only sane way to build an bigger appliaiton :) lots of small bricks[23:18:56] <konobi> even sdc is the same idea[23:19:24] <Licenser> I guessed so :) would make no sense otherwise[23:19:49] <Licenser> I'd expect it to be broken up even further then fifo[23:22:56] <Licenser> but I guess you can't compare the two really :P[23:34:38] *** fsteinel has quit IRC[23:48:31] *** trentster has joined #smartos[23:54:38] *** newb_ has quit IRC[23:56:55] *** jim80net has quit IRC