February 11, 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 11, 2014 [00:14:15] *** olauzon has quit IRC[00:17:56] *** conan_the_destro has quit IRC[00:22:41] *** conan_the_destro has joined #project-FiFo[01:07:43] *** conan_the_destro has quit IRC[01:08:14] *** conan_the_destro has joined #project-FiFo[01:50:06] *** conan_the_destro has quit IRC[02:03:50] *** jim80net has quit IRC[02:10:00] *** _dumfries has quit IRC[02:15:21] *** _dumfries has joined #project-FiFo[02:26:46] *** ipalreadytaken has quit IRC[02:39:53] *** jim80net has joined #project-FiFo[05:20:36] *** ira has quit IRC[05:44:14] *** ira has joined #project-FiFo[05:54:09] *** ira has quit IRC[09:36:57] *** tomww has quit IRC[09:47:06] *** tomww has joined #project-FiFo[10:21:08] *** tomww has quit IRC[10:22:13] *** tomww has joined #project-FiFo[10:22:14] *** tomww has joined #project-FiFo[10:28:12] *** tomww has quit IRC[10:33:19] *** tomww has joined #project-FiFo[11:02:36] *** bixu has joined #project-FiFo[11:33:41] *** tomww has quit IRC[11:33:49] *** tomww has joined #project-FiFo[11:33:49] *** tomww has joined #project-FiFo[11:36:02] *** wiedi has quit IRC[11:37:21] *** wiedi has joined #project-FiFo[11:51:23] *** tomww has quit IRC[11:56:31] *** tomww has joined #project-FiFo[12:04:01] *** tomww has quit IRC[12:04:10] *** tomww has joined #project-FiFo[12:08:24] *** tomww has quit IRC[12:13:27] *** ira has joined #project-FiFo[12:14:11] *** tomww has joined #project-FiFo[13:24:43] *** bixu_ has joined #project-FiFo[13:26:14] *** bixu has quit IRC[13:56:32] *** tomww has quit IRC[13:59:38] *** bixu has joined #project-FiFo[13:59:51] *** bixu_ has quit IRC[14:06:39] *** tomww has joined #project-FiFo[14:08:11] *** bixu_ has joined #project-FiFo[14:08:58] *** bixu has quit IRC[14:11:20] *** tomww has quit IRC[14:16:40] *** bixu has joined #project-FiFo[14:17:53] *** bixu__ has joined #project-FiFo[14:18:10] *** bixu_ has quit IRC[14:21:05] *** bixu_ has joined #project-FiFo[14:21:21] *** bixu has quit IRC[14:21:45] *** tomww has joined #project-FiFo[14:21:45] *** tomww has joined #project-FiFo[14:23:16] *** alcir has joined #project-FiFo[14:23:28] *** bixu__ has quit IRC[14:25:08] *** bixu has joined #project-FiFo[14:27:40] *** bixu_ has quit IRC[14:36:38] *** bixu_ has joined #project-FiFo[14:38:00] *** bixu__ has joined #project-FiFo[14:38:52] *** bixu has quit IRC[14:40:53] *** bixu_ has quit IRC[14:47:45] *** bixu__ has quit IRC[15:03:04] *** jim80net has quit IRC[15:34:20] *** jim80net has joined #project-FiFo[15:46:18] *** tomww has quit IRC[15:46:31] *** tomww has joined #project-FiFo[16:00:00] *** _lb_ has joined #project-FiFo[16:17:49] *** _lb_ has quit IRC[17:07:25] *** alcir has quit IRC[17:09:34] *** conan_the_destro has joined #project-FiFo[17:10:54] *** _jared_ has quit IRC[17:11:40] *** _jared_ has joined #project-FiFo[17:22:56] *** _lb_ has joined #project-FiFo[18:03:01] *** _lb_ has quit IRC[18:35:23] *** _lb_ has joined #project-FiFo[20:41:03] *** wiedi_ has joined #project-FiFo[20:41:34] *** wiedi has quit IRC[21:08:12] *** ipalreadytaken has joined #project-FiFo[21:19:55] <killfill> whos steven on irc?..[21:20:23] <killfill> bulk of mails.. already feeling im working on a helpdesk of something.. :P[21:27:26] <MerlinDMC> HAHA![21:27:57] <MerlinDMC> maybe I should get some ghosting accounts as well :P[21:30:05] <killfill> heh[21:40:54] <MerlinDMC> I need a faster dev server here[21:41:09] <MerlinDMC> compiling stuff takes way too long[21:50:38] *** mgls has joined #project-FiFo[21:56:09] <bayoda1> hi killfill[21:56:59] <bayoda1> you asked for the problem about fifo filling it's ZFS[22:11:06] <killfill> hey bayoda1[22:11:09] <bayoda1> hi[22:11:22] <bayoda1> filled the ZFS get filled bug immediatly again ...[22:11:23] <killfill> yes, well im not sure if the zfs thing has to do with the vm creation problem[22:11:32] <bayoda1> maybe[22:11:55] <bayoda1> i installed complete fresh - imported some datasets from datasets.at[22:12:23] <killfill> ah and it started to fillit?[22:12:44] <bayoda1> (just base13.3.1 / owncloud / ghost / gitlab)[22:12:51] <bayoda1> just 4.6 gb or something -[22:13:02] <bayoda1> sunday it started to grow[22:13:20] <bayoda1> today - 40GB is full... zone is "dead"[22:13:38] <MerlinDMC> it fills without you doing anything?[22:13:44] <bayoda1> cause filled up[22:13:51] <bayoda1> yes correct[22:14:30] <MerlinDMC> hmm ... could use smth like that for my credit card[22:14:37] <MerlinDMC> but weird behaviour[22:14:46] <bayoda1> i just did - ONE thing -> klicked on an old Machine (not having the dataset) - and clicked on import - (to get the information)[22:15:13] <bayoda1> merlin if this algo can work for my bank account too - I'm in ... (have some 38TB free here :D)[22:15:25] <killfill> heh[22:15:42] <bayoda1> the /var/db is filled up[22:15:44] <MerlinDMC> would at least be smth useful and new[22:15:47] <bayoda1> in ./sniffle[22:16:06] <bayoda1> with files like this[22:16:07] <bayoda1> 15G ./sniffle/365375409332725729550921208179070754913983135744.img[22:16:27] <MerlinDMC> is that sniffle db hanoidb?[22:16:43] <bayoda1> just basic install - no other config[22:16:49] <MerlinDMC> someone asked about hanoidb stability a while ago[22:17:10] <bayoda1> no leoFS (damn good - otherwise my TB cluster would be full)[22:17:35] <bayoda1> that bug is new - it did not happen - in 0.4.1[22:17:40] <bayoda1> and 0.4.2[22:18:05] <bayoda1> what are this *.img files[22:18:18] <killfill> i guess Licenser would now better about this..[22:18:40] <killfill> shuold be the datasets downloaded somewhat in hanoidb format?[22:18:42] <MerlinDMC> i'll grep the code while my freebsd is building -.-[22:18:56] <killfill> i know its partitioned somewhere.. maybe its there.. (somewhat 3x the original size)[22:19:19] <bayoda1> but be sure - biggest dataset is 590mB[22:19:20] <Licenser> bayoda1 0.4.1 had more files so what you'd have seen is 64 smaller files instead of 8 bigger ones[22:19:29] <Licenser> space consumption has not changed[22:19:42] <bayoda1> but it didn't fill lice[22:19:59] <bayoda1> it didn't kill my fifo zone before[22:20:23] <Licenser> bayoda1 what are you storing in there?[22:20:34] <bayoda1> you should tell me :D[22:20:46] <bayoda1> it's just plain - basic - fifo - one zone[22:20:58] <bayoda1> 21 machines on 5 hypervisors[22:21:12] <Licenser> bayoda1 what datasets was the question sorry[22:21:33] * killfill just noticed dataset import broked on his install...[22:21:34] <bayoda1> 2 windows machines - (that stuck in provisined - running )[22:21:56] <bayoda1> all from datasets.at[22:22:02] <bayoda1> no own build[22:22:20] <bayoda1> github / ghost / base 13.3.1 / owncloud[22:22:25] <bayoda1> i imported[22:22:45] <bayoda1> AND - i klicked on an old (1.8.3 VM and klick there on import - maybe this kills[22:23:26] <killfill> [you can rm /zones/<id>/root/svc/provisioning to while the privisinoed state btw][22:23:37] <bayoda1> so if an Machine do not have the corresponding dataset already imported - it imports it - and i (just my stomach says that to me) - there's the treasure burried[22:24:24] <bayoda1> i (1 week ago) rm the img files - and it worked till next day - filled up again[22:24:57] <bayoda1> thought about a defect in my installation - installed again .. but aehmm not really :)[22:25:01] <bayoda1> thx kill[22:25:44] <bayoda1> to replay - what does this *.img files contain?[22:28:31] <Licenser> bayoda1 the image files store: 1) imported images 2) created images[22:28:59] <Licenser> are there any reports about handoffs?[22:29:00] <bayoda1> but 40GB - for just that handfull of imported datasets?[22:29:11] <bayoda1> where to look for that reports?[22:29:35] <bayoda1> handoff reports?[22:30:05] <bayoda1> 9.9G ./sniffle/730750818665451459101842416358141509827966271488.img[22:30:05] <bayoda1> 1.5K ./sniffle/1096126227998177188652763624537212264741949407232/sst_5[22:30:05] <bayoda1> 90K ./sniffle/1096126227998177188652763624537212264741949407232/sst_2[22:30:05] <bayoda1> 1.5K ./sniffle/1096126227998177188652763624537212264741949407232/sst_6[22:30:05] <bayoda1> 12K ./sniffle/1096126227998177188652763624537212264741949407232/sst_1[22:30:05] <bayoda1> 1.5K ./sniffle/1096126227998177188652763624537212264741949407232/sst_3[22:30:06] <bayoda1> 1.5K ./sniffle/1096126227998177188652763624537212264741949407232/sst_4[22:30:06] <bayoda1> 23K ./sniffle/1096126227998177188652763624537212264741949407232/sst_0[22:30:07] <bayoda1> 3.9M ./sniffle/1096126227998177188652763624537212264741949407232[22:30:09] <bayoda1> ?[22:31:28] <Licenser> in the logs /var/log/sniffle/*[22:31:43] <bayoda1> let me check[22:32:51] <bayoda1> console.log ?[22:33:01] <bayoda1> crash.log?[22:33:06] <bayoda1> error.log?[22:36:01] <bayoda1> maybe intresting[22:36:02] <bayoda1> ** Reason for termination =[22:36:02] <bayoda1> ** timeout[22:36:02] <bayoda1> 2014-02-11 12:19:59 =CRASH REPORT====[22:36:02] <bayoda1> crasher:[22:36:02] <bayoda1> initial call: sniffle_entity_read_fsm:init/1[22:36:03] <bayoda1> pid: <0.337.184>[22:36:03] <bayoda1> registered_name: [][22:36:04] <bayoda1> exception exit: {timeout,[{gen_fsm,terminate,7,[{file,"gen_fsm.erl"},{line,611}]},{proc_lib,init_p_do_apply,3,[{file,"proc_lib.erl"},{line,239}]}]}[22:36:04] <bayoda1> ancestors: [sniffle_entity_read_fsm_sup,sniffle_sup,<0.360.0>][22:36:05] <bayoda1> messages: [][22:36:05] <bayoda1> links: [<0.367.0>][22:36:06] <bayoda1> dictionary: [][22:36:35] <bayoda1> hmmm[22:37:48] <bayoda1> and i got: run_erl [67079] Tue Feb 11 16:59:23 2014[22:37:48] <bayoda1> Error in writing to log.[22:38:04] <bayoda1> okay cause zfs is full[22:38:40] <killfill> i guess you can give it more space.. but that is not the cause of the problem.. :P[22:39:03] <bayoda1> i already tried that last week - 20GB more[22:39:15] <bayoda1> filled within another 10 hours[22:39:53] <bayoda1> on this machine up to 400gb available than this sun fire is filled up[22:39:57] <MerlinDMC> 20G/10h so you pretty much get your 38TB filled up till summer? :P ... would be really weird[22:40:30] <bayoda1> its growing faster the bigger it is ;-) kind of "Evolution" bug[22:40:40] <MerlinDMC> weird[22:40:42] <bayoda1> but head and shoulders may not work[22:40:58] <MerlinDMC> reminds me to get that movie again[22:41:33] <MerlinDMC> but building my freebsd image builder first! :P[22:42:26] <bayoda1> i think it happens on the "import" when there's already one installed vm with this base dataset[22:42:49] <MerlinDMC> it should not import again then ... right?[22:43:21] <bayoda1> if it isn't imported - it imports - but i think it imports and multiplies[22:44:33] <bayoda1> the funny thing . if i delete the img file - fifo works as expected[22:44:42] <bayoda1> but filling again[22:44:54] <MerlinDMC> cronjob *couch*[22:46:18] <bayoda1> hmmm no cronjob in crontab that may do this[22:47:23] <bayoda1> maybe i should rename fifo zone to fiLo zone[22:48:35] *** mgls has quit IRC[22:49:45] <MerlinDMC> (it was more the part to create a cronjob that does the deletion)[22:51:21] <bayoda1> LOL[22:51:22] <bayoda1> okay[22:52:05] <bayoda1> cronjob scp *.img file to IP of Licencer[22:52:38] <bayoda1> O:-)[22:52:47] <bayoda1> i have 200mbit connection ;-)[22:53:24] *** _lb_ has quit IRC[22:53:52] <bayoda1> seems that killfill and licencer are digging trough some code[22:54:24] *** mgls has joined #project-FiFo[22:55:39] <Licenser> bayoda1 can you share your sniffle.conf please and do you have more then 1 physical node? also can you do fifoadm datasets list[22:56:55] <bayoda1> paste in comments on jira?[22:57:06] <Licenser> bayoda1 btw: grep handoff /var/log/sniffle/*.log[22:57:15] <Licenser> or gists both fine[22:58:48] *** mgls has quit IRC[23:00:46] *** mgls has joined #project-FiFo[23:02:02] <Licenser> killfill we need to get more people to help eachother[23:02:08] <Licenser> support desk duty sucks![23:03:31] <MerlinDMC> if that thing now runs I'm fine with the day :) automatically created freebsd image for smartos[23:04:43] *** mgls has quit IRC[23:05:51] <Licenser> MerlinDMC nice[23:06:17] <MerlinDMC> Licenser, yep ... and only 700MB uncompressed ... not 2G compressed :P[23:06:19] *** ipalreadytaken has quit IRC[23:06:45] <Licenser> heh[23:08:05] <bayoda1> okay my zone is dead[23:08:05] <bayoda1> [Connected to zone '95c850f9-d0e6-4e00-9b8f-846f4b3bffae' pts/11][23:08:05] <bayoda1> No utmpx entry. You must exec "login" from the lowest level "shell".[23:08:29] <bayoda1> this was fifo Zone new - R.I.P[23:08:31] <bayoda1> :D[23:08:37] <Licenser> bayoda1 something is very wrong[23:09:02] <Licenser> can you tar the content of /var/log/sniffle and mail it to me?[23:09:02] <bayoda1> yes that happens when /var/log is full[23:09:15] <bayoda1> zlogin is terminated[23:09:48] *** mgls has joined #project-FiFo[23:10:53] <bayoda1> licencer - no handoff in the log files[23:10:57] <Licenser> bayoda1 you still can just tar it up from the GZ[23:11:13] <bayoda1> i know ..[23:17:49] <bayoda1> Licencer send you mail[23:17:58] <bayoda1> with tar[23:19:29] <bayoda1> beware licencer I'm using aggregates on smartos :D[23:19:38] <bayoda1> but heck that really works now ![23:21:28] <MerlinDMC> :( forgot to install bash ... but the rest is fine so i'm pretty much done[23:21:44] <bayoda1> just put gnome :D[23:21:48] <bayoda1> why bash[23:21:49] <bayoda1> LOL[23:22:17] <MerlinDMC> bayoda1, because the scripts from joyent use bash as shell -> i need bash to use me-freebsd[23:22:34] <bayoda1> okay[23:22:47] <bayoda1> was a joke like head and shoulders ;-)[23:24:15] <bayoda1> aside watching a documentation on animal planet - white shark - diving - without cage ...[23:24:20] <bayoda1> brrrr[23:26:24] <Licenser> bayoda1 I think I know the error, simple workaround use LeoFS[23:26:49] <bayoda1> nice joke[23:27:05] <bayoda1> you mean filling up 38tb is easier with leofs[23:27:36] <Licenser> you can try increasing the number of VNodes[23:27:42] *** olauzon has joined #project-FiFo[23:27:45] <Licenser> I don't think why what happens happens but I hink I know what happens[23:28:05] <bayoda1> and what happens really[23:29:07] *** ipalreadytaken has joined #project-FiFo[23:29:32] <Licenser> bayoda1 what happens is that for some reason there seems to be repairs (I don't know why they happenb) but causes the statebox of the imgae (which I shouldn't be there) keeps growing[23:29:36] <Licenser> so it's nothing I can fix in 5m[23:30:09] * MerlinDMC sleep &[23:31:30] <bayoda1> okay the img could be deleted[23:33:03] <bayoda1> 1.bitcask.data 1.bitcask.hint 2.bitcask.data 2.bitcask.hint 3.bitcask.data 3.bitcask.hint 4.bitcask.data 4.bitcask.hint bitcask.write.lock[23:33:15] <bayoda1> this is what i have in the *. img/ directory[23:33:28] *** ipalread_ has joined #project-FiFo[23:33:30] <Licenser> bayoda1 the only thing that cahnged between the 4.1 and 4.2 was the default number of vnodes[23:33:45] <Licenser> so it might help to just set the vnodes to 16 etc[23:34:05] *** mgls has quit IRC[23:34:12] <bayoda1> could i rm the *.data files?[23:34:17] <bayoda1> or not[23:34:37] <Licenser> bitcask is only used for images[23:34:56] *** mgls has joined #project-FiFo[23:36:18] *** mgls has quit IRC[23:36:32] *** ipalreadytaken has quit IRC[23:41:39] *** ipalreadytaken has joined #project-FiFo[23:41:52] *** ipalread_ has quit IRC[23:51:21] <bayoda1> killfill -> (22:23:26) killfill: [you can rm /zones/<id>/root/svc/provisioning to while the privisinoed state btw][23:51:33] <bayoda1> this is not working on KVM datasets[23:51:37] <bayoda1> no svc there[23:55:49] *** olauzon has quit IRC