February 26, 2014 
< | 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 | >
February 26, 2014 [00:02:21] *** ipalreadytaken has quit IRC[00:17:49] *** jim80net has quit IRC[00:24:20] *** ipalreadytaken has joined #project-FiFo[01:28:12] -msg- [chunter] Licenser pushed 1 new commit to dev: http://git.io/Ws4SNw[01:28:13] -msg- chunter/dev 6d35c7c Heinz N. Gies: Merge branch 'test' into dev[02:44:52] *** fridder has quit IRC[02:55:22] *** olauzon has quit IRC[03:52:39] *** barhom has quit IRC[04:52:27] -msg- [chunter] Licenser pushed 1 new commit to dev: http://git.io/r0tQdw[04:52:27] -msg- chunter/dev d63bf6e Heinz N. Gies: Merge branch 'test' into dev[04:56:21] -msg- [chunter] Licenser pushed 1 new commit to dev: http://git.io/icKC7Q[04:56:22] -msg- chunter/dev eecde01 Heinz N. Gies: Merge branch 'test' into dev[05:24:04] *** ipalreadytaken has quit IRC[05:24:32] *** ipalreadytaken has joined #project-FiFo[05:29:10] *** ipalreadytaken has quit IRC[05:30:39] -msg- [chunter] Licenser pushed 2 new commits to dev: http://git.io/7zGLcg[05:30:39] -msg- chunter/dev c179185 Heinz N. Gies: fxied bug in chunter locking server.[05:30:39] -msg- chunter/dev 8351cb7 Heinz N. Gies: Merge branch 'test' into dev[05:32:15] *** _lb_ has joined #project-FiFo[05:43:34] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/tBKS2g[05:43:34] -msg- sniffle/dev 05c06ca Heinz N. Gies: Bump.[05:45:19] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/r8Gt-g[05:45:19] -msg- sniffle/dev 8d5ee2b Heinz N. Gies: Corrected libchunter version.[05:58:52] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/VuHqjQ[05:58:52] -msg- sniffle/dev 0b249c8 Heinz N. Gies: Merge branch 'test' into dev[06:00:52] -msg- [chunter] Licenser pushed 1 new commit to dev: http://git.io/xshRwA[06:00:52] -msg- chunter/dev 2b202c2 Heinz N. Gies: Merge branch 'test' into dev[06:14:00] -msg- [chunter] Licenser pushed 1 new commit to dev: http://git.io/jbd1LQ[06:14:00] -msg- chunter/dev d90af6b Heinz N. Gies: Merge branch 'test' into dev[06:15:25] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/tgAS9A[06:15:25] -msg- sniffle/dev 3b10aa0 Heinz N. Gies: Merge branch 'test' into dev[06:25:05] <Licenser> killfill your simultanious create issue should be fixed![06:31:04] <Licenser> it's not since the problem is different from what I thought[06:34:21] <Licenser> problem for tomorrow[08:10:21] <bayoda> morning[08:12:27] <MerlinDMC> morning[08:23:28] *** barhom has joined #project-FiFo[08:24:15] *** aortmann has joined #project-FiFo[09:28:54] *** barhom has quit IRC[09:33:35] *** alcir has joined #project-FiFo[09:36:21] <bayoda> if anyone has seen it - joyent and ubuntu / cannonical made a partnership for certified ubuntu "images"[09:38:17] <bayoda> on the joyent cloud - sounds interesting - as it shows that smartos gains momentum[12:38:54] *** hugo has joined #project-FiFo[12:38:55] *** hugo has joined #project-FiFo[12:58:06] <killfill> hi[12:58:19] <killfill> Licenser: almost.. :P[12:58:53] <killfill> Licenser: i think i will target the test branch.. so the irc bot shows the actuall messages[12:59:06] <killfill> or maybe we should use rebase instead..[13:11:48] <killfill> uuh.. maybe someone that knows git better than me should do the rebase on jingles..[13:23:52] <killfill> Licenser: sending a PUT to /hypervisors/:uuid/services doesnt really work.. :S[13:32:05] -msg- [jingles] killfill fast-forwarded test from fb058cf to 0ba1a18: http://git.io/GUo-2g[13:34:46] <killfill> bayoda: ok just pushed the service on hypervisor thing. PD: when you change a service state, it wont work jet. (need Licencer's help)[14:19:15] <bayoda> killfill cool[14:19:47] <bayoda> I'm just fighting with ORG/GROUP/USER - the user doesn't see his vm[14:23:25] <bayoda> will update within 1 hour[14:35:38] *** _lb_ has quit IRC[14:39:00] *** _lb_ has joined #project-FiFo[14:57:50] *** _lb_ has quit IRC[14:59:21] *** _lb_ has joined #project-FiFo[15:05:23] <alcir> well[15:05:36] <alcir> why some vms are reported as provisioning (running)[15:05:56] <alcir> even if from the related GZ it is running (and not provisioning?)[15:12:46] <konobi> alcir: status from which tool?[15:13:58] <alcir> yes[15:14:04] <alcir> from the web page of fifo[15:16:21] <konobi> and status from the GZ is from where?[15:16:53] <alcir> from the gz where the vm live[15:17:03] <alcir> vmadm list[15:17:34] <konobi> sounds like a bug?[15:17:42] <alcir> don't know[15:29:59] *** jim80net has joined #project-FiFo[15:34:40] <killfill> alcir: thats becouse fifo knows if a vm is provisioning if the file 'provisioning' exists.[15:34:52] <killfill> Probably when creating the VM, the tools didnt wipe that file[15:34:58] <killfill> or maybe becouse you created it manually.. :P[15:35:15] <killfill> its in /zones/:uuid/root/svc/ something like that.[15:35:25] <killfill> /zones/<uuid>/root/var/svc/provisioning[15:43:28] <alcir> killfill thank you[15:46:07] <aortmann> something like this known If I want to restart a vm throw fifo? "[Error: UNABLE_TO_VERIFY_LEAF_SIGNATURE]"[15:46:22] <aortmann> throw fifo api[15:55:03] <aortmann> I just get this message when I do it with https[15:55:05] <aortmann> with http it works[15:55:11] <aortmann> is there a way to change this?[15:56:35] <killfill> aortmann: what base version are you using on the fifo zone?[15:56:46] <killfill> base64 13.2.1?[16:01:42] *** mattronix has quit IRC[16:03:33] *** mattronix has joined #project-FiFo[16:27:20] -msg- [sniffle] Licenser pushed 2 new commits to dev: http://git.io/uoL4eA[16:27:20] -msg- sniffle/dev 7922712 Heinz N. Gies: Fixed problem with only one create fsm being able to run at a time.[16:27:21] -msg- sniffle/dev b0a27e0 Heinz N. Gies: Merge branch 'test' into dev[16:37:25] <Licenser> the deletion of the provisioning file is something up to the zone if the maintainer of the zone file did not propperly do that or the zonefile is too old (joyent changed how that worked half a year back)[16:37:45] <Licenser> killfill the starting/stopping of GZ services is not yet existing[16:40:30] *** fridder has joined #project-FiFo[16:40:51] <killfill> Licenser: cool. will do implement that? or should i take the buttons out?[16:41:05] <Licenser> It will be there eventually[16:42:27] <killfill> ok will take it out then.[16:42:47] <killfill> ah btw, howl is sending services event frecuently, when no service is changed.[16:42:55] <killfill> (at least on my setup..)[16:44:52] <Licenser> hmmm should not[16:45:42] <Licenser> killfill fixed the multi creation thing[16:46:01] <killfill> yay![16:52:27] <killfill> it does![16:52:35] <killfill> now you have another fifo-fact.. :P[16:52:35] <Licenser> hypervisors will now lock on VM provisioning so you can send up a hundred VM requests and they will be dealt with one by one (or N by N if you have N valid hypervisors)[16:52:41] <Licenser> true![16:52:55] <Licenser> even so I don't like the interface[16:52:58] <killfill> (mashines-{1-20} will create 20 machines.. )[16:52:58] <Licenser> it's too magicy[16:53:17] <killfill> the interface of what=[16:53:18] <killfill> ?[16:53:24] <Licenser> of creating muyltiple machines[16:53:32] <Licenser> magic == bad, obvious == good[16:53:48] <killfill> ah yah.. its a little hacky i guess[16:54:32] <Licenser> at one point smoene wants to name their machiens system-{1-100} and end up getting 100 vms :P[16:54:33] <killfill> where was the problem? on vmadm? what where you locking?[16:54:36] *** olauzon has joined #project-FiFo[16:54:46] <Licenser> killfill multiple problems:[16:55:04] <Licenser> 1) The most stupid one, only 1 create FSM per sniffle instance could run since it was a named process == bug[16:55:21] * killfill would never name a machin is a 'moustache'.. :P[16:55:26] <Licenser> 2) When two creates happen very quickly they could end up claiming the same memory[16:55:48] <Licenser> 3) vmamd isn't all happy with working concurrently[16:56:16] <Licenser> what happehns now is that when trying to find a hypervisor sniffle requests a lock and only uses the hypervisor if it can get that lock - if it can't it tries the next one[16:56:26] <killfill> oh that 2) is a dirty one[16:56:56] <killfill> hope there is no situation where something fails and the lock doesnt end[16:57:52] *** bixu_ has joined #project-FiFo[16:58:26] <killfill> would be nice to take the time until creation ends.. to compare to other iaas's.. :P[16:58:53] <Licenser> killfill locks timeout after 2m[16:59:08] <killfill> i guess it will not be quick.. becouse of the locking.. maybe not a good idea.. :P[16:59:08] <Licenser> 1m sorry[16:59:17] <killfill> cool[16:59:37] <Licenser> killfill usually the lock ends very quickly[17:09:34] <killfill> hm..[17:11:25] <killfill> with 35 it didnt do very well.. :P[17:13:46] *** ira_ has joined #project-FiFo[17:17:18] <Licenser> killfill https://gist.github.com/Licenser/9232820[17:17:28] <Licenser> that are the locking for 4 machines[17:21:19] <Licenser> killfill define not too well?[17:22:17] <killfill> letme try again[17:22:51] *** aortmann has quit IRC[17:24:41] <killfill> its like fifo went out of sync with vmadm now..[17:24:45] * killfill cleaning the mess..[17:27:35] *** ira_ has quit IRC[17:27:47] *** alcir has quit IRC[17:28:59] <killfill> niffle_vm_vnode:handle_command:189 [vms] tried to write to a non existing vm: not_found :D[17:29:27] <killfill> i think i should have make a snapshot before playing with this stuff.. :P[17:29:28] *** fridder has quit IRC[17:29:46] *** conan_the_destro has joined #project-FiFo[17:46:51] <Licenser> killfill heh[17:56:14] *** ipalreadytaken has joined #project-FiFo[18:05:06] <killfill> Licenser: when a vm changes its package, the howl 'update' event doesnt contain the new package[18:05:36] <killfill> (nothing to do with this, but it may be a bug anyway)[18:23:09] -msg- [jingles] killfill fast-forwarded test from 0ba1a18 to 286f77c: http://git.io/1FU5YA[18:37:37] <killfill> Licenser: i think i see different things when creating > 10. like:[18:37:53] <Licenser> hmm okay cause 5 looked quite good[18:38:07] <killfill> 1.- Sometimes the browser after sending the put, and following the 303, it gets a 500[18:38:39] *** ipalread_ has joined #project-FiFo[18:38:40] <killfill> 2.- howl at some point stops sending the change state event. Some mashines finishes ok, but no event sent.[18:38:52] <killfill> 3.- some machiens finish on 'fail' state[18:39:13] *** ipalreadytaken has quit IRC[18:39:35] *** ipalreadytaken has joined #project-FiFo[18:39:40] <killfill> on case, the history shows 'Finding hypervisor etc, etc'[18:40:17] <killfill> you think it could be that becouse of the lock, someone is waiting and going out with a timeout, thous 'failing' ?[18:42:12] *** ipalrea__ has joined #project-FiFo[18:42:15] *** ipalreadytaken has quit IRC[18:42:25] <killfill> Licenser: not really sure if its worth the energy to go throught this tho... if people need to create much machine they probably are using scripts.. not the gui.. and they probably will do it serially not parallel[18:43:29] *** ipalread_ has quit IRC[18:48:27] <Licenser> killfill true[18:49:36] <Licenser> killfill any chance you can put in dataset upload API into jingles before 0.4.3 release?[18:51:02] -msg- [jingles] killfill fast-forwarded test from 286f77c to a4bc5a5: http://git.io/UR3-fQ[18:51:35] <killfill> Licenser: like for user to select zvol and manfest from the PC and upload them ?[18:52:21] * killfill getting lunch[18:53:48] <Licenser> killfill yea[18:53:59] <Licenser> it's 2 API calls, 1) to upload the manifest 2) to upload the zvol[19:06:45] -msg- [sniffle] Licenser pushed 2 new commits to dev: http://git.io/tcfQ7g[19:06:45] -msg- sniffle/dev db77315 Heinz N. Gies: Improved FSM's[19:06:45] -msg- sniffle/dev a23b6db Heinz N. Gies: Merge branch 'test' into dev[19:09:57] *** conan_the_destro has quit IRC[19:11:17] *** conan_the_destro has joined #project-FiFo[19:12:05] <Licenser> killfill versioning is broken :( no requests are send[19:19:39] <Licenser> killfill I'll need to add some throtteling on VM creation as like each FiFo server can provision X VM's at max per time[19:33:34] <killfill> Licenser: maybe we should file a bug on vmadm or such.. if the problem is there..[19:33:56] <Licenser> killfill it's a mixed problem I'm not sure if VMADM actually is the issue[19:35:16] <that0n3guy> is it bad to have fifo accessed via the world?[19:36:05] <killfill> that0n3guy: doesnt sound bad to me![19:36:19] <killfill> the gui and wiggle at least.. :P[19:37:32] <that0n3guy> so just firewall off everything except 8080 and 80? Its secure enough to let live?[19:38:18] <killfill> you probably would need just 443[19:39:17] <that0n3guy> Gotcha... thanks[19:41:52] <killfill> Licenser: XMLHttpRequest cannot load http://release.project-fifo.net/pkg/dev/sniffle.version. No 'Access-Control-Allow-Origin' header is present on the requested resource[19:42:09] <killfill> CORS stuff?[19:44:19] <Licenser> hmmmm[19:44:23] <Licenser> <3 nginx[19:44:31] <Licenser> propably yes[19:45:00] <killfill> add_header Access-Control-Allow-Origin *;[19:45:01] <killfill> :)[19:45:56] *** ipalrea__ has quit IRC[19:46:32] *** ipalreadytaken has joined #project-FiFo[19:50:56] <killfill> Licenser: btw.. where do i get a zvol?..[19:51:14] *** ipalreadytaken has quit IRC[19:51:15] <Licenser> killfill locally, when you upload it[19:52:08] <killfill> are they just dumps of /dev/zvol/dsk/zones/:uuid?[19:52:15] <killfill> with gzip?[19:55:11] *** ipalreadytaken has joined #project-FiFo[19:56:29] *** ipalreadytaken has quit IRC[19:56:57] *** ipalreadytaken has joined #project-FiFo[20:02:03] *** ipalreadytaken has quit IRC[20:05:54] *** ipalreadytaken has joined #project-FiFo[20:10:33] *** ipalreadytaken has quit IRC[20:11:00] *** ipalreadytaken has joined #project-FiFo[20:15:26] *** ipalreadytaken has quit IRC[20:15:50] *** ipalreadytaken has joined #project-FiFo[20:18:13] <Licenser> killfill huh nono, upload to fifo[20:22:13] <killfill> yeah, but what do you upload?.. a manifest file, and a zvol.. i guess you can build a zvol yourself with the mibe stuff of joyent.. but you could also just dd a current vm and gzip it. and upload that?[20:26:36] *** ipalreadytaken has quit IRC[20:27:03] *** ipalreadytaken has joined #project-FiFo[20:31:29] *** ipalreadytaken has quit IRC[20:47:45] *** olauzon has quit IRC[20:58:49] *** saltsa has quit IRC[21:02:10] *** jim80net has quit IRC[21:03:26] *** jim80net has joined #project-FiFo[21:05:02] *** saltsa has joined #project-FiFo[21:49:43] *** ipalreadytaken has joined #project-FiFo[21:50:14] *** ipalreadytaken has quit IRC[21:50:41] *** ipalreadytaken has joined #project-FiFo[21:53:28] <that0n3guy> can someone tell me what ipf pass in/out I need in smartos (global zone) to for chunter if my admin ip is 10.0.1.1?[21:55:00] *** ipalreadytaken has quit IRC[21:56:46] *** jim80net has quit IRC[21:58:32] *** jim80net has joined #project-FiFo[22:12:03] *** ipalreadytaken has joined #project-FiFo[22:23:18] *** echelog has joined #project-FiFo[22:27:37] *** saltsa_ has joined #project-FiFo[22:28:08] *** saltsa has quit IRC[22:36:11] *** venky has quit IRC[22:37:34] *** conan_the_destro has quit IRC[22:37:40] *** venky__ has joined #project-FiFo[22:48:42] <Licenser> that0n3guy the netowkrlayout in the wiki lists ports[22:49:20] <trentster> morning all, how goes things?[22:49:27] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/dv0-Jw[22:49:27] -msg- sniffle/dev d7fee54 Heinz N. Gies: Merge branch 'test' into dev[22:50:01] <that0n3guy> yeah, I figured it out... fifo was listening on my external rather than my internal so it wasn't working when it shuld have.[22:50:12] <trentster> Hi Licenser hows things with you mate? :^)[22:50:29] <Licenser> that0n3guy ah okay :)[22:50:37] <that0n3guy> Licenser: now I just have to figure out how to get it to see internal network instead of external[22:50:47] <Licenser> trentster tired, how about yourelf?[22:51:00] <Licenser> that0n3guy there is a configuration option for chunter what Ip to listen on[22:52:28] <trentster> Licenser: yeah me too, been busy with a house move last week, going to try re-setup servers etc today. Thats if I can muster up the energy to run a network cable to the server cabinet in the garage and find my crimping tools etc.[22:52:53] <Licenser> heh[22:53:09] <trentster> Licenser: why tired, isnt it only like 5pm there at the moment?[22:53:41] <Licenser> trentster not too much sleep :P[22:53:51] <Licenser> was up long fixing locking in FiFo[22:54:15] <Licenser> also my brother is visiting right now so I've been walking a hell lot latelyu[22:54:37] <Licenser> trentster: https://cloudup.com/iaMhrBfeyXm[22:55:11] <trentster> amazing pic![22:57:24] <trentster> It must be great to have your bro there, and get to explore the city together.[22:58:25] <Licenser> yea it is[22:58:39] <Licenser> But exhausting, we walked SO much thoguht the citi :P[23:15:46] *** conan_the_destro has joined #project-FiFo[23:19:24] <Licenser> trentster the fifo page is now on Lucera and I added full SSL to it :)[23:19:36] <that0n3guy> Licenser, I can't get fifo to see chunter on interal... Do I set the ip in app.config?[23:19:54] -msg- [sniffle] Licenser pushed 1 new commit to dev: http://git.io/Km6zSg[23:19:54] -msg- sniffle/dev 72afcf3 Heinz N. Gies: Merge branch 'test' into dev[23:20:04] <Licenser> that0n3guy yap you set up the IP in the app.conf[23:20:11] <Licenser> or if you're ion dev in the chunter.conf[23:21:15] <that0n3guy> Licenser: I noticed it has commas instead of decimals... that seems odd.[23:21:24] <Licenser> it's just the syntax of the file[23:21:29] <Licenser> it's a datastructure[23:21:45] <Licenser> on dev we moved away from that formate, and made it a normal looking file[23:26:04] <trentster> Licenser: tried to pvt message you says you are offline, weird.[23:27:17] <Licenser> o.O I am not![23:37:25] -msg- [sniffle] Licenser pushed 2 new commits to dev: http://git.io/J0KemQ[23:37:25] -msg- sniffle/dev b00bca8 Heinz N. Gies: Fixed missing start for create pool[23:37:26] -msg- sniffle/dev f2812ba Heinz N. Gies: Merge branch 'test' into dev[23:42:55] *** trentster has left #project-FiFo