*** shalini has quit IRC | 00:08 | |
*** shalini has joined #openstack-trove | 00:08 | |
*** achampion has quit IRC | 00:26 | |
*** matsuhashi has joined #openstack-trove | 00:32 | |
datsun180b | hub_cap: yt? | 00:35 |
---|---|---|
*** achampion has joined #openstack-trove | 00:42 | |
*** yogesh has joined #openstack-trove | 00:44 | |
*** tanisdl has quit IRC | 00:52 | |
*** adrian_otto has joined #openstack-trove | 00:54 | |
cp16net | datsun180b: yo | 01:02 |
datsun180b | sup | 01:02 |
*** nosnos has joined #openstack-trove | 01:09 | |
datsun180b | oh this is the dumbest solution and i'm mad it works | 01:10 |
datsun180b | OLD_SAVE = bkup_models.DBBackup.save ... bkup_modes.DBBackup.save = OLD_SAVE | 01:11 |
datsun180b | testtools.TestCase got some splainin to do | 01:12 |
hub_cap | datsun180b: ModelNotFoundError: DBBackup Not Found | 01:12 |
hub_cap | thats the real error just so u know | 01:13 |
datsun180b | and where'd you find that | 01:13 |
*** achampio1 has joined #openstack-trove | 01:14 | |
datsun180b | so why wouldn't the model be found in the unittests? is that table not being initialized properly? | 01:14 |
hub_cap | datsun180b: https://gist.github.com/hub-cap/7651859 | 01:14 |
hub_cap | datsun180b: that, im unsure about | 01:14 |
hub_cap | everythign seems a-ok | 01:14 |
datsun180b | so there's some kind of model switch i can't find just yet | 01:15 |
*** achampion has quit IRC | 01:15 | |
* hub_cap shrugs | 01:16 | |
datsun180b | also how'd you manage to get any kind of useful banana debugging out of tox? it seems to eat all of that | 01:16 |
hub_cap | it _must_ be timing | 01:16 |
datsun180b | it can't be db locks because a lock would cause a wait, not a failure | 01:16 |
hub_cap | im not using tox | 01:16 |
hub_cap | ive dropped down into testr's venv | 01:17 |
hub_cap | i find it much faster / easier | 01:17 |
hub_cap | and its showing me the errors | 01:17 |
datsun180b | yeah i'm going to just start using testr now, this is chicken-fried bullshit | 01:17 |
hub_cap | ++ | 01:17 |
datsun180b | so if i save the save method and restore it as the first action in my test's setUp it appears to be working all right | 01:18 |
*** adrian_otto has quit IRC | 01:18 | |
*** achampion has joined #openstack-trove | 01:19 | |
datsun180b | it's okay right up until the body of setUp, even in the same file anywhere else on the left edge of the file | 01:19 |
*** achampio1 has quit IRC | 01:19 | |
openstackgerrit | Ed Cranford proposed a change to openstack/trove: Conductor proxies host db access for guests https://review.openstack.org/45116 | 01:20 |
hub_cap | ok datsun180b im not sure what was going on w/ that error | 01:21 |
*** rongze has joined #openstack-trove | 01:21 | |
datsun180b | i'm sort of hoping this doesn't do it | 01:22 |
datsun180b | but really i hope it works so we can move on and revisit the problem with clearer heads | 01:23 |
hub_cap | link me to what u did | 01:25 |
datsun180b | https://review.openstack.org/#/c/45116/33..34/trove/tests/unittests/conductor/test_methods.py | 01:25 |
*** rongze has quit IRC | 01:26 | |
hub_cap | what huh | 01:27 |
datsun180b | yeah i know | 01:27 |
datsun180b | see if you were to assert that the save method wasn't None at about line 42 without this in there, you'd get an assertionerror | 01:27 |
datsun180b | zuul's got a bit on his plate, so now i get to be patient | 01:28 |
hub_cap | datsun180b: so i have confirmed what you said w/o seeing what you said lol | 01:30 |
hub_cap | save is None | 01:30 |
datsun180b | did that link not work for you? | 01:30 |
hub_cap | it did. but i wanted to be sure that my eyes did not defy me | 01:31 |
datsun180b | gotcha | 01:31 |
datsun180b | i'm aware this is a stupid fix | 01:31 |
hub_cap | cuz that sounds quite odd | 01:31 |
datsun180b | in a perfect world those two lines would basically be a noop | 01:32 |
datsun180b | seeing as that method shouldn't be nulled at any point | 01:32 |
hub_cap | is it possible that other tests could modify the state of the mock vs real object? as in, a diff unit test mocking baseModel.create ? | 01:36 |
hub_cap | i dont want a rogue test clobbering things | 01:36 |
datsun180b | ahahahaha py26 just passed on jenkins | 01:36 |
hub_cap | not saying its your test thats at fault :) | 01:36 |
datsun180b | it is certainly not | 01:36 |
datsun180b | i couldn't find anything in the tests that modified the save method | 01:36 |
datsun180b | trust me, i looked | 01:36 |
datsun180b | that is, i couldn't see anything that just straight up said "go find the save method and replace it with this None I found" | 01:37 |
datsun180b | yeah the jenkins gates passed | 01:38 |
*** rongze has joined #openstack-trove | 01:39 | |
hub_cap | fo sure | 01:40 |
datsun180b | all right, it's go-home oclock | 01:40 |
datsun180b | i'll read about conductor's stunning success in the morning | 01:41 |
datsun180b | thanks for your help | 01:41 |
*** datsun180b has quit IRC | 01:42 | |
hub_cap | np | 01:45 |
hub_cap | im still lookin | 01:45 |
hub_cap | its intriguing | 01:45 |
*** rongze has quit IRC | 01:46 | |
*** yogesh has quit IRC | 01:47 | |
*** amcrn has quit IRC | 01:52 | |
*** rnirmal has quit IRC | 01:55 | |
*** rongze has joined #openstack-trove | 02:01 | |
*** nosnos_ has joined #openstack-trove | 02:02 | |
*** yogesh has joined #openstack-trove | 02:04 | |
*** nosnos has quit IRC | 02:05 | |
*** coolsvap has quit IRC | 02:06 | |
*** yogesh_ has joined #openstack-trove | 02:09 | |
*** yogesh has quit IRC | 02:11 | |
*** erkules has quit IRC | 02:12 | |
*** yogesh_ has quit IRC | 02:16 | |
*** yogesh has joined #openstack-trove | 02:16 | |
*** yogesh has quit IRC | 02:21 | |
*** erkules has joined #openstack-trove | 02:26 | |
*** amytron has quit IRC | 02:32 | |
*** matsuhashi has quit IRC | 03:11 | |
*** coolsvap has joined #openstack-trove | 03:50 | |
*** matsuhashi has joined #openstack-trove | 04:19 | |
*** vipul-away is now known as vipul | 04:29 | |
*** coolsvap has quit IRC | 04:31 | |
*** nosnos_ has quit IRC | 04:32 | |
*** nosnos has joined #openstack-trove | 04:32 | |
*** rongze has quit IRC | 04:48 | |
*** yogesh has joined #openstack-trove | 04:59 | |
*** yogesh has quit IRC | 05:04 | |
*** SushilKM has joined #openstack-trove | 05:04 | |
*** coolsvap_ has joined #openstack-trove | 05:05 | |
*** SushilKM has quit IRC | 05:08 | |
*** zigo_ has quit IRC | 05:10 | |
*** zigo has joined #openstack-trove | 05:11 | |
*** rongze has joined #openstack-trove | 05:19 | |
openstackgerrit | Sushil Kumar proposed a change to openstack/trove: Removes security_groups from _create_server_volume_heat signature https://review.openstack.org/58203 | 05:21 |
*** coolsvap_ has quit IRC | 05:24 | |
*** rongze has quit IRC | 05:27 | |
*** coolsvap has joined #openstack-trove | 05:29 | |
*** matsuhashi has quit IRC | 05:50 | |
*** matsuhashi has joined #openstack-trove | 05:51 | |
*** matsuhas_ has joined #openstack-trove | 05:53 | |
*** matsuhashi has quit IRC | 05:53 | |
*** yogesh has joined #openstack-trove | 06:03 | |
*** yogesh_ has joined #openstack-trove | 06:08 | |
*** yogesh has quit IRC | 06:10 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove: setting fake mode to print startup errors https://review.openstack.org/58440 | 06:17 |
*** matsuhas_ has quit IRC | 06:20 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: fake mode script needs to be updated after changed were merged in trove https://review.openstack.org/58441 | 06:20 |
*** denis_makogon has joined #openstack-trove | 06:23 | |
*** rongze has joined #openstack-trove | 06:24 | |
*** yogesh_ has quit IRC | 06:25 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: changes to support configuration groups https://review.openstack.org/58445 | 06:25 |
*** yogesh has joined #openstack-trove | 06:26 | |
*** coolsvap has quit IRC | 06:26 | |
*** rongze has quit IRC | 06:28 | |
*** matsuhashi has joined #openstack-trove | 06:29 | |
*** yogesh has quit IRC | 06:31 | |
*** coolsvap has joined #openstack-trove | 06:33 | |
*** coolsvap has quit IRC | 06:36 | |
*** rongze has joined #openstack-trove | 06:38 | |
*** achampion has quit IRC | 06:39 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/python-troveclient: adds support for configurations management https://review.openstack.org/53169 | 06:40 |
*** yogesh has joined #openstack-trove | 06:42 | |
*** haomaiwa_ has quit IRC | 06:52 | |
*** haomaiwang has joined #openstack-trove | 06:52 | |
*** haomaiwang has quit IRC | 06:53 | |
*** matsuhashi has quit IRC | 06:53 | |
*** haomaiwang has joined #openstack-trove | 06:53 | |
*** yogesh has quit IRC | 06:54 | |
*** matsuhashi has joined #openstack-trove | 06:54 | |
*** yogesh has joined #openstack-trove | 06:54 | |
*** matsuhashi has quit IRC | 06:58 | |
*** yogesh has quit IRC | 06:59 | |
*** matsuhashi has joined #openstack-trove | 06:59 | |
*** achampion has joined #openstack-trove | 07:04 | |
*** yogesh has joined #openstack-trove | 07:07 | |
*** matsuhashi has quit IRC | 07:26 | |
*** matsuhashi has joined #openstack-trove | 07:27 | |
*** ikhudoshyn has joined #openstack-trove | 07:29 | |
*** matsuhas_ has joined #openstack-trove | 07:39 | |
*** matsuhashi has quit IRC | 07:42 | |
*** yogesh has quit IRC | 07:58 | |
*** yogesh has joined #openstack-trove | 07:59 | |
*** ashestakov has joined #openstack-trove | 08:00 | |
*** ashestakov has quit IRC | 08:00 | |
*** matsuhas_ has quit IRC | 08:05 | |
*** matsuhashi has joined #openstack-trove | 08:05 | |
*** SushilKM has joined #openstack-trove | 08:08 | |
*** flaper87|afk is now known as flaper87 | 08:10 | |
*** yogesh has quit IRC | 08:16 | |
*** yogesh_ has joined #openstack-trove | 08:19 | |
*** yogesh has joined #openstack-trove | 08:22 | |
*** yogesh_ has quit IRC | 08:24 | |
*** yogesh_ has joined #openstack-trove | 08:26 | |
*** yogesh has quit IRC | 08:29 | |
*** SushilKM has quit IRC | 08:35 | |
*** denis_makogon has quit IRC | 08:39 | |
*** yogesh has joined #openstack-trove | 08:45 | |
*** yogesh_ has quit IRC | 08:49 | |
*** _coolsvap_ has joined #openstack-trove | 09:06 | |
*** yogesh has quit IRC | 09:09 | |
*** yogesh_ has joined #openstack-trove | 09:14 | |
*** yogesh_ has quit IRC | 09:17 | |
*** yogesh has joined #openstack-trove | 09:18 | |
*** yogesh_ has joined #openstack-trove | 09:22 | |
*** yogesh has quit IRC | 09:23 | |
*** achampion has quit IRC | 09:29 | |
*** _coolsvap_ has quit IRC | 09:37 | |
*** yogeshmehra has joined #openstack-trove | 09:51 | |
*** matsuhas_ has joined #openstack-trove | 09:58 | |
*** matsuhas_ has quit IRC | 09:58 | |
*** matsuhashi has quit IRC | 10:02 | |
*** yogesh_ has quit IRC | 10:06 | |
*** achampion has joined #openstack-trove | 10:07 | |
*** yogesh has joined #openstack-trove | 10:07 | |
*** yogesh has quit IRC | 10:11 | |
*** achampion has quit IRC | 10:12 | |
*** away-gone is now known as denis_makogon | 10:17 | |
denis_makogon | /msg NickServ identify dmakogon | 10:17 |
*** rongze has quit IRC | 10:29 | |
*** yogeshmehra has quit IRC | 10:30 | |
openstackgerrit | Kevin Conway proposed a change to openstack/trove-integration: Add support for kick-starting new datastores https://review.openstack.org/57780 | 11:50 |
*** flaper87 is now known as flaper87|afk | 12:11 | |
*** nosnos has quit IRC | 12:11 | |
*** rongze has joined #openstack-trove | 12:15 | |
*** flaper87|afk is now known as flaper87 | 12:16 | |
*** SushilKM has joined #openstack-trove | 12:58 | |
*** pdmars has joined #openstack-trove | 12:59 | |
*** SushilKM has quit IRC | 13:09 | |
*** rongze has quit IRC | 13:17 | |
*** SushilKM has joined #openstack-trove | 13:28 | |
*** rongze has joined #openstack-trove | 13:29 | |
*** isviridov has left #openstack-trove | 13:57 | |
*** isviridov has joined #openstack-trove | 14:00 | |
*** SushilKM has quit IRC | 14:05 | |
*** SnowDust has joined #openstack-trove | 14:07 | |
*** achampion has joined #openstack-trove | 14:15 | |
*** isviridov is now known as help | 14:15 | |
*** help is now known as isviridov | 14:15 | |
*** isviridov has left #openstack-trove | 14:19 | |
openstackgerrit | Illia Khudoshyn proposed a change to openstack/trove: Initial support for single instance MongoDB support https://review.openstack.org/50597 | 14:21 |
*** SnowDust has quit IRC | 14:30 | |
*** radez_g0n3 is now known as radez | 14:36 | |
*** isviridov has joined #openstack-trove | 14:39 | |
*** NehaV has joined #openstack-trove | 14:44 | |
*** yogeshmehra has joined #openstack-trove | 14:49 | |
*** adrian_otto has joined #openstack-trove | 14:59 | |
*** adrian_otto has quit IRC | 14:59 | |
*** robertmyers has joined #openstack-trove | 15:02 | |
*** adrian_otto has joined #openstack-trove | 15:04 | |
*** Barker has joined #openstack-trove | 15:16 | |
*** jasonb365 has joined #openstack-trove | 15:23 | |
*** ashestakov has joined #openstack-trove | 15:25 | |
*** datsun180b has joined #openstack-trove | 15:29 | |
*** yogeshmehra has quit IRC | 15:34 | |
*** rnirmal has joined #openstack-trove | 15:45 | |
*** achampio1 has joined #openstack-trove | 15:47 | |
*** achampion has quit IRC | 15:50 | |
datsun180b | https://review.openstack.org/#/c/45116/33..34/trove/tests/unittests/conductor/test_methods.py took way too long to figure out | 15:52 |
datsun180b | in other news conductor is green again and anxiously awaiting approval and merging | 15:53 |
*** jmontemayor has joined #openstack-trove | 15:55 | |
denis_makogon | datsun180b, already +1 it)) | 15:59 |
datsun180b | saw that, thanks | 15:59 |
datsun180b | it's a stupid bug, the model appears to be corrupted but only when testcase is invoked | 15:59 |
*** amytron has joined #openstack-trove | 16:02 | |
*** achampion has joined #openstack-trove | 16:04 | |
*** achampio1 has quit IRC | 16:05 | |
*** jasonb365 has quit IRC | 16:06 | |
robertmyers | datsun180b: I think it is an issue with mockito | 16:07 |
datsun180b | i built a dummy test class that has no trove deps except the backup models, and got the same problem | 16:11 |
datsun180b | https://gist.github.com/ed-/094337a9c27738eb9d09 the assert in setUp fails | 16:11 |
datsun180b | though that doesn't rule out other tests ahead of that test running | 16:12 |
datsun180b | well that's interesting, importing testtools all at once causes the problem, BUT just importing testcase doesn't | 16:15 |
datsun180b | false alarm | 16:20 |
*** Barker has quit IRC | 16:31 | |
*** Barker has joined #openstack-trove | 16:34 | |
*** achampio1 has joined #openstack-trove | 16:34 | |
*** achampion has quit IRC | 16:36 | |
*** grapex has joined #openstack-trove | 16:37 | |
denis_makogon | guys, take a look at this one https://review.openstack.org/#/c/54900/ | 16:40 |
datsun180b | aw man i like doing that kind of thing | 16:42 |
openstackgerrit | Denis M. proposed a change to openstack/trove: Provide guest_info per datastore https://review.openstack.org/58574 | 16:42 |
denis_makogon | datsun180b, thanks))) | 16:43 |
datsun180b | well i mean i would have preferred to have made those updates but if you'll do it, then it saves me work! | 16:43 |
denis_makogon | datsun180b, sorry for stilling your bread | 16:44 |
denis_makogon | datsun180b, you could do the same with trove | 16:45 |
denis_makogon | datsun180b, not so long ago i tried to do that, but i've failed, pep8 took all my time =/ | 16:45 |
*** grapex has quit IRC | 16:48 | |
datsun180b | it's fine, we all have work to do | 16:48 |
*** grapex has joined #openstack-trove | 16:49 | |
datsun180b | and thanks for your work too | 16:49 |
denis_makogon | datsun180b, thanks, you also doing great stuff | 16:50 |
openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Adding ConnectionError class https://review.openstack.org/56930 | 16:51 |
openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Exception handle on unreachable trove-api https://review.openstack.org/56930 | 16:52 |
*** flaper87 is now known as flaper87|afk | 16:59 | |
*** yidclare has joined #openstack-trove | 17:00 | |
denis_makogon | robertmyers, ping | 17:06 |
robertmyers | denis_makogon: hello | 17:06 |
denis_makogon | robertmyers, hi, Robert | 17:06 |
denis_makogon | i saw your comment on my review | 17:06 |
denis_makogon | robertmyers, i'd like to explain if you don't mind | 17:07 |
denis_makogon | robertmyers, print was there, but also there is _pretty_print | 17:07 |
robertmyers | yeah, the print command is now a function in python 3 | 17:08 |
robertmyers | so you can call print('blah') | 17:08 |
denis_makogon | robertmyers, i'm not so deep with such things, but gate was failing on print statement, because Py3 syntax doesn't support such format | 17:08 |
robertmyers | instead of print 'blah' | 17:08 |
denis_makogon | yes | 17:08 |
denis_makogon | that is what i did | 17:08 |
robertmyers | I think it will work if you use is as a function the same way... I'd have to see, but I'm guessing that is what the failure is | 17:09 |
denis_makogon | robertmyers, ok, if it would be ok, could i get a +1 ? | 17:10 |
robertmyers | yes | 17:11 |
*** plodronio has joined #openstack-trove | 17:11 | |
denis_makogon | robertmyers, thanks | 17:12 |
*** SushilKM has joined #openstack-trove | 17:15 | |
robertmyers | denis_makogon: it looks like it can be either 'Okay: print(msg)' or 'Okay: print (msg)' | 17:17 |
robertmyers | personnally I like the former | 17:17 |
robertmyers | but I'll +1 it | 17:18 |
SushilKM | needed reviews followed by merge for https://review.openstack.org/#/c/58203/ :) | 17:18 |
SushilKM | being ecstatic | 17:18 |
SushilKM | :) | 17:18 |
hub_cap | SushilKM: a merge typically follows reviews :) | 17:19 |
SushilKM | news to me hub_cap, was thinking otherwise ;) | 17:19 |
SushilKM | hub_cap, please check that, though its a small change | 17:20 |
hub_cap | SushilKM: -2 :) | 17:20 |
SushilKM | what happen, any complaints, please | 17:20 |
*** david-lyle is now known as david-lyle_afk | 17:22 | |
datsun180b | oh so hub_cap / grapex / vipul / SlickNik guess what turned green yesterday: https://review.openstack.org/#/c/45116/ | 17:22 |
datsun180b | finally nailed that unittest problem, at least for now | 17:22 |
datsun180b | it bears deeper analysis but not at the cost of essential features | 17:23 |
*** grapex_ has joined #openstack-trove | 17:23 | |
*** grapex has quit IRC | 17:25 | |
*** Barker has quit IRC | 17:27 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: clean up the trailing spaces in xsd/dbaas.wadl https://review.openstack.org/58591 | 17:27 |
hub_cap | datsun180b: plz mark a bug for that issue, and make a comment in the code that says "check lp:XXXXXXX" w/ the bug id | 17:28 |
datsun180b | hub_cap: is the bug against trove, trove-int, or something else? | 17:28 |
datsun180b | oh you want another patchset don't you | 17:29 |
hub_cap | datsun180b: i do, they are free :) | 17:29 |
*** Barker has joined #openstack-trove | 17:29 | |
hub_cap | ummm.... i think trove is good to file it | 17:29 |
datsun180b | fine, i'll get on that | 17:30 |
SushilKM | http://upload.wikimedia.org/wikipedia/commons/5/55/Bug_blank.svg is a need everywhere ;) | 17:31 |
datsun180b | i don't want to lose sleep over conductor when i'm on vacation :c | 17:31 |
datsun180b | i want everyone happy about it so it gets merged and we can all move on | 17:32 |
datsun180b | And I want to quit pestering you all about it | 17:33 |
ikhudoshyn | hup_cap: ... the mount_point | 17:40 |
hub_cap | datsun180b: when do u go on vaca? | 17:40 |
datsun180b | hub_cap: taking tomorrow off, and of course thursday | 17:40 |
datsun180b | closer to crimbo probably a week or more, got floats to burn off | 17:40 |
*** achampion has joined #openstack-trove | 17:41 | |
hub_cap | good call.. i think i do too | 17:41 |
denis_makogon | datsun180b, agreed with your comments on PEP8 review, soring is our everything) | 17:41 |
datsun180b | denis_makogon: in vim, use V to grab the lines you want to sort, then type :%!sort and hit enter | 17:42 |
denis_makogon | datsun180b, i'm almost newbie with vim | 17:42 |
*** grapex_ has quit IRC | 17:42 | |
datsun180b | now it inserts '<'> for you before you hit ! since you used V (visual-line mode) but the result is what you highlighted gets sorted | 17:42 |
*** jasonb365 has joined #openstack-trove | 17:43 | |
datsun180b | with vim i'm not sorceror supreme but i know a handful of neat tricks | 17:44 |
*** achampio1 has quit IRC | 17:44 | |
ikhudoshyn | hub_cap: hi again, i guess i misspelled ur nick, blame me.. but what about mount_point | 17:44 |
*** rongze has quit IRC | 17:44 | |
hub_cap | ikhudoshyn: ask your questions | 17:44 |
hub_cap | we can speak about it | 17:44 |
ikhudoshyn | what's the whole idea, at first | 17:45 |
ikhudoshyn | or could u point me, if it was discussed | 17:45 |
hub_cap | ikhudoshyn: i dont remember it being discussed | 17:45 |
hub_cap | the last few wks have been quite hazy though | 17:45 |
ikhudoshyn | then what should i change | 17:46 |
hub_cap | ikhudoshyn: link your review | 17:46 |
ikhudoshyn | https://review.openstack.org/#/c/57189/ please | 17:46 |
ikhudoshyn | the main point is to remove mount_point from RPC. Den's idea (Den, fix me if i wrong) to get mount_point from db specific config | 17:47 |
ikhudoshyn | but for me these two not necessary directly linked | 17:47 |
ikhudoshyn | Den, was that u? | 17:48 |
hub_cap | ikhudoshyn: great... i didnt leave any context in what i said | 17:48 |
hub_cap | denis_makogon: ^ ^ | 17:48 |
denis_makogon | hub_cap, hello_ | 17:49 |
denis_makogon | )) | 17:49 |
ikhudoshyn | let me try, | 17:49 |
ikhudoshyn | the context was, its nice to get mount_point from, say, my.cnf, for mysql, or from mongo.conf | 17:50 |
*** grapex has joined #openstack-trove | 17:50 | |
denis_makogon | hub_cap, my point is to avoid multiple configuration for singe parameter: one comes in trove-taskmanager.conf, another one comes in config.template | 17:50 |
ikhudoshyn | but it shuold not be taken from RPC call anyway | 17:50 |
hub_cap | ah yes | 17:51 |
ikhudoshyn | this looks like two steps for me | 17:51 |
ikhudoshyn | isn't it? | 17:51 |
denis_makogon | nope | 17:51 |
denis_makogon | one | 17:51 |
denis_makogon | but 2 steps for one review | 17:51 |
openstackgerrit | Ed Cranford proposed a change to openstack/trove: Conductor proxies host db access for guests https://review.openstack.org/45116 | 17:52 |
datsun180b | spoilers: i added comments | 17:52 |
hub_cap | datsun180b: http://static3.wikia.nocookie.net/__cb20130921024344/tardis/images/e/e5/River_Looks_Left_LKH.jpg | 17:53 |
ikhudoshyn | well, that review mostly updates server-side. Using config.template mostly implies guest-agent side.. | 17:53 |
datsun180b | i'm so behind on dr who | 17:53 |
hub_cap | ikhudoshyn: denis_makogon brings up this point because you can do that "instead of" this code, rather than do both | 17:54 |
*** vipul is now known as vipul-away | 17:54 | |
hub_cap | datsun180b: she just says spoilers a lot cuz she knows the doctors timeline | 17:54 |
datsun180b | oh i know who River Song is | 17:54 |
denis_makogon | hub_cap, yeah | 17:54 |
datsun180b | I just left off early in season 7 I think | 17:54 |
hub_cap | datsun180b: as in, the year of the original doctor who year 7 calendar? | 17:55 |
*** SushilKM has quit IRC | 17:55 | |
kevinconway | hey there. easy with the Who spoilers | 17:55 |
hub_cap | or the newest season of the reboot datsun180b | 17:56 |
ikhudoshyn | hub_cap: got it, thanks | 17:56 |
hub_cap | kevinconway: dont worry, ill only spoil original doctor who | 17:56 |
hub_cap | his face changes, his dog is stupid and annoying, the reboot will most certainly be better | 17:56 |
hub_cap | ^ ^ kevinconway: spoilers | 17:57 |
kevinconway | YOU DON"T LIKE K-9? | 17:57 |
hub_cap | lol | 17:57 |
hub_cap | no | 17:57 |
hub_cap | ikhudoshyn: i think we left off not knowing if we should take that mount_point from the trove.conf and inject it in to the conf.template, or just assume itll be in the conf.template | 17:58 |
hub_cap | ikhudoshyn: either way, the guest can glean that info from the config file it gets sent, and i think thats a fine and dandy idea | 17:58 |
robertmyers | fine and dandy indeed | 17:59 |
datsun180b | i think you can figure out which season 7 i mean if i know who river song is | 18:00 |
hub_cap | well not exactly datsun180b, u could be reading up on your _own_ spoilers | 18:01 |
hub_cap | and be in classic who | 18:01 |
hub_cap | thiking, omg, why did i start it this way | 18:01 |
hub_cap | our son was the 11th doctor (matt smith) for halloween, now we have a baptism we need to go to for our friends baby in the area, so we are just gonna dress him back up, sans sonic screw, as the doctor for the day hehe | 18:03 |
ikhudoshyn | hub_cap: ic, and we could make the server-side code a lil'bit lighter by removing all the hustle around mount_point as well | 18:03 |
ikhudoshyn | i think i could love it | 18:04 |
hub_cap | most definitely ikhudoshyn | 18:04 |
hub_cap | ikhudoshyn: :) | 18:04 |
ikhudoshyn | ok, I'm done with my stuff for today, cu all | 18:04 |
hub_cap | thx ikhudoshyn cu | 18:05 |
denis_makogon | hub_cap, could we update delete instance method and let it delete instance in any kind state ? | 18:11 |
hub_cap | denis_makogon: do u have a problem w/ the way it works now? if so give reason for why you want to change it | 18:11 |
denis_makogon | hub_cap, suppose i want to break provisioning because i mentioned wrong parameter (bigger volume, flavor, name), but until it gets any status (excluding BUILDING) i need to wait | 18:13 |
denis_makogon | or exception happened even before poll until | 18:13 |
denis_makogon | in that case instance would stay in build status forever | 18:14 |
denis_makogon | hub_cap, btw i started to work on infra extension for us | 18:15 |
*** yogesh has joined #openstack-trove | 18:17 | |
hub_cap | denis_makogon: if exceptions are happening and its not moving to error, thats a bug | 18:20 |
hub_cap | id rather make sure we get the instance into error state | 18:20 |
denis_makogon | hub_cap, problems could come not from code | 18:21 |
denis_makogon | hub_cap, from code-free files, like templates | 18:21 |
denis_makogon | to be precise - missing templates | 18:21 |
hub_cap | denis_makogon: will a missing template not throw an exception? | 18:22 |
denis_makogon | missing template cause exception in taskmanager, instance stays in BUILD status | 18:22 |
denis_makogon | hub_cap, yes, that problem doesn't fixed | 18:22 |
openstackgerrit | Denis M. proposed a change to openstack/trove: Provide guest_info per datastore https://review.openstack.org/58574 | 18:23 |
hub_cap | denis_makogon: well lets fix that bug | 18:23 |
hub_cap | rather than let the bug stay around in the code and masking it by letting users delete "build" instances | 18:24 |
denis_makogon | sounds good | 18:24 |
hub_cap | also, letting a user delete somethign in build will, say, delete step 2 in a 4 or 5 step process, and then step 4 could fail on that | 18:24 |
hub_cap | causing it to go back in to build status | 18:24 |
hub_cap | and itll be even more cluttered system then | 18:24 |
denis_makogon | maybe | 18:24 |
denis_makogon | i mean, yes, it would bring us to more complicated handling system | 18:25 |
denis_makogon | hub_cap, could you please take a look at this https://review.openstack.org/#/c/56930/ | 18:25 |
denis_makogon | and maybe, this https://review.openstack.org/#/c/54900/ | 18:26 |
denis_makogon | hub_cap, with this change we could make secure access to instance specific ports, that is what we want but oslo didn't gave us more complex parameter that Dict | 18:29 |
denis_makogon | https://review.openstack.org/#/c/58574/ | 18:29 |
*** vipul-away is now known as vipul | 18:29 | |
openstackgerrit | Denis M. proposed a change to openstack/python-troveclient: Exception handle on unreachable trove-api https://review.openstack.org/56930 | 18:30 |
hub_cap | denis_makogon: yup i remember that | 18:31 |
datsun180b | when did we add a pypy gate to troveclient? | 18:32 |
denis_makogon | hub_cap, i'm on the way to make it real | 18:32 |
datsun180b | is that from test-requirements | 18:32 |
hub_cap | datsun180b: i think during the "rewrite" | 18:33 |
datsun180b | well whatever, apparently it's working | 18:34 |
hub_cap | datsun180b: those come from the -infra/config stuff | 18:34 |
datsun180b | that's what i thought | 18:34 |
denis_makogon | hub_cap, about your comment, i looked through all OS clients nobody need such class | 18:34 |
hub_cap | so what do the other OS clients do denis_makogon? | 18:35 |
denis_makogon | they reuse exception which comes from httplib or requests | 18:35 |
hub_cap | k | 18:35 |
hub_cap | can u link one in the review? | 18:35 |
denis_makogon | which review ? | 18:36 |
hub_cap | in the review that i -2'd, link the code that you viewed in the other OS client | 18:36 |
denis_makogon | hub_cap, i can try to add this class to oslo | 18:36 |
hub_cap | denis_makogon: you can, but if there is no reason, just prove there is no reason by linking the other OS client's code | 18:37 |
denis_makogon | https://github.com/openstack/python-novaclient/blob/935e1fe1674c7680dfdcfed53996d66530fd9ed3/novaclient/exceptions.py | 18:37 |
denis_makogon | #L54 | 18:37 |
denis_makogon | https://github.com/openstack/python-cinderclient/blob/0c2839414f999a8b9a0b085419de7a00d18e52ff/cinderclient/exceptions.py#L56 | 18:38 |
hub_cap | so they are both using some sort of ConnectionRefused/Error | 18:38 |
hub_cap | id say propose the connectionrefused or whatever u want to call it, to oslo | 18:38 |
denis_makogon | yeah, i can do that | 18:39 |
hub_cap | awesome | 18:39 |
denis_makogon | there's only one client where i cannot find any kind of such exception - heatclient | 18:39 |
denis_makogon | https://github.com/openstack/python-keystoneclient/blob/8c8ed5f74f36aad814ee8f4a23711f0acfe5ac0b/keystoneclient/httpclient.py#L116 | 18:40 |
SlickNik | robertmyers / vipul: Good comments on https://review.openstack.org/#/c/58062/5. Thanks! | 18:40 |
SlickNik | I like the idea of making it a dictopt and applicable across all backup types. | 18:40 |
SlickNik | Stay tuned for a new patchset. | 18:40 |
*** SnowDust has joined #openstack-trove | 18:41 | |
denis_makogon | SlickNik, what if you extract whole backup command from code into some kind of config or template or something else | 18:41 |
denis_makogon | personally i see no problems with doing it | 18:42 |
*** david-lyle_afk is now known as david-lyle | 18:42 | |
SlickNik | denis_makogon: You can only extract parts of it (and it's probably a good idea). Other parts like the subprocess code, the process supervisor, setting up the pipes, etc need to exist across all backup types. This still needs to be part of the common code. | 18:44 |
*** rongze has joined #openstack-trove | 18:44 | |
denis_makogon | SlickNik, maybe, but hardcoding all command is not so good idea, do you agree ? | 18:46 |
robertmyers | SlickNik: cool | 18:46 |
*** yogesh has quit IRC | 18:47 | |
*** yogesh has joined #openstack-trove | 18:47 | |
SlickNik | denis_makogon: The strategy class _is_ a template. | 18:49 |
SlickNik | There's nothing preventing you from defining and deploying another backup strategy with a different command, and switching out the backup strategy class in the config. | 18:50 |
*** yogesh has quit IRC | 18:52 | |
robertmyers | denis_makogon: I agree with SlickNik, it seems like you could easily place this in a config file but then you quickly run into an edge case which you need to add a function to the restore or backup class | 18:52 |
*** rongze has quit IRC | 18:53 | |
robertmyers | denis_makogon: see _pre_restore and _post_restore | 18:53 |
datsun180b | oh look the robots are still happy with conductor https://review.openstack.org/#/c/45116/ | 18:56 |
datsun180b | if only i could get some core approval too | 18:56 |
SlickNik | robertmyers / denis_makogon: However, something that needs some thought (and work) is how to support multiple backup types for multiple datastore types. | 18:56 |
robertmyers | SlickNik: denis_makogon: https://review.openstack.org/#/c/52905/ why was this adandoned? | 18:59 |
denis_makogon | SlickNik, i'm working on that | 19:01 |
denis_makogon | robertmyers, SlickNik https://review.openstack.org/#/c/58574/ | 19:01 |
datsun180b | SlickNik: <3 | 19:06 |
datsun180b | thought you needed more highlights | 19:06 |
*** yogesh has joined #openstack-trove | 19:12 | |
openstackgerrit | SlickNik proposed a change to openstack/trove: Replace uuidutils.generate_uuid() with str(uuid.uuid4()) https://review.openstack.org/58071 | 19:31 |
datsun180b | i seem to remember at least one point when generating uuid4s violated an api constraint for being too long but now i can't remember where | 19:41 |
datsun180b | i think i have in a test somewhere a generate_uuid(len=16), i'll have to look | 19:41 |
datsun180b | hah! it's in conductor's test_methods | 19:42 |
datsun180b | not that big a deal, that already uses uuid.uuid4(), hooray | 19:43 |
hub_cap | or u could use oslo uuid_utils.generate_uuid | 19:43 |
datsun180b | but i know we have uuid.uuid4(), that's probably not going away | 19:45 |
datsun180b | well this case is only used for generating bogus ids for bogus db tests and won't do any harm for now | 19:45 |
*** rongze has joined #openstack-trove | 19:49 | |
SlickNik | hub_cap: oslo uuid_utils.generate_uuid is going away. | 19:51 |
hub_cap | :o | 19:51 |
hub_cap | for just using uuid.uuid4 ? | 19:51 |
SlickNik | Yes, that's what they recommend. | 19:52 |
SlickNik | https://bugs.launchpad.net/nova/+bug/1253497 | 19:52 |
datsun180b | hence "hooray" earlier | 19:52 |
hub_cap | datsun180b: wins the prize | 19:53 |
datsun180b | i'll win the prize when zuul is happy | 19:53 |
datsun180b | then you ALL get to use conductor, it'll be so much fun oh boy | 19:54 |
*** rongze has quit IRC | 19:54 | |
hub_cap | datsun180b: i saw a typo, did u fix it? | 19:54 |
datsun180b | huh | 19:54 |
hub_cap | itll only be one more review | 19:54 |
datsun180b | don't even joke | 19:54 |
SlickNik | lol | 19:54 |
hub_cap | HAHAHA | 19:54 |
datsun180b | is it a semantic mistake | 19:54 |
hub_cap | yo i wants me some enchiladas today | 19:55 |
kevinconway | if the generate_uuid is going away then what gets used in all the SQLAlchemy models? | 19:57 |
*** NehaV has quit IRC | 19:57 | |
datsun180b | kevinconway: def generate_uuid(): return 9 | 19:57 |
datsun180b | # chosen fairly, guaranteed to be random | 19:57 |
hub_cap | HAH | 19:57 |
kevinconway | #link http://search.dilbert.com/comic/Random%20Number%20Generator | 19:58 |
datsun180b | exactly | 19:58 |
SlickNik | kevinconway: replace with trove.common.utils.generate_uuid which is a wrapper to str(uuid.uuid4()) https://review.openstack.org/58071 | 20:00 |
datsun180b | oh man, it's like someone's got their finger on the pulse about this kind of thing | 20:00 |
*** spredzy has joined #openstack-trove | 20:03 | |
*** jasonb365 has quit IRC | 20:06 | |
*** datsun180b is now known as datsun_F40PH | 20:12 | |
*** radez is now known as radez_g0n3 | 20:13 | |
*** vipul is now known as vipul-away | 20:14 | |
*** vipul-away is now known as vipul | 20:14 | |
datsun_F40PH | neat, freenode supports nick groups | 20:14 |
openstackgerrit | A change was merged to openstack/trove: Conductor proxies host db access for guests https://review.openstack.org/45116 | 20:15 |
datsun_F40PH | CHOO CHOO | 20:15 |
*** SnowDust has quit IRC | 20:17 | |
*** hub_cap is now known as pub_tap | 20:25 | |
pub_tap | i wanted to name change too | 20:25 |
pub_tap | datsun_F40PH: ^ ^ | 20:25 |
datsun_F40PH | yeah i'd say that's accurate | 20:25 |
*** jasonb365 has joined #openstack-trove | 20:27 | |
datsun_F40PH | so why didn't hudson go update https://blueprints.launchpad.net/trove/+spec/trove-conductor yet i wonder | 20:28 |
*** SnowDust has joined #openstack-trove | 20:32 | |
*** denis_makogon_ has joined #openstack-trove | 20:34 | |
*** denis_makogon has quit IRC | 20:41 | |
*** denis_makogon_ is now known as denis_makogon | 20:41 | |
*** vipul is now known as vipul-away | 20:41 | |
*** dmakogon_ has joined #openstack-trove | 20:41 | |
denis_makogon | guys, i'm back ,did i missed something ? | 20:42 |
*** yogesh has quit IRC | 20:43 | |
datsun_F40PH | denis_makogon: you know what got merged | 20:43 |
denis_makogon | datsun_F40PH, hah, cond ? | 20:43 |
denis_makogon | ))) | 20:43 |
*** yogesh has joined #openstack-trove | 20:43 | |
datsun_F40PH | oh yes | 20:43 |
denis_makogon | datsun_F40PH, we gonna have a party))) | 20:43 |
* denis_makogon wants party-hard | 20:44 | |
datsun_F40PH | no, we're going to finish our other work! | 20:44 |
denis_makogon | which one ? | 20:46 |
datsun_F40PH | work on our other guest agent | 20:47 |
*** yogesh has quit IRC | 20:48 | |
pub_tap | datsun_F40PH: the bp's do not get updated | 20:48 |
datsun_F40PH | plumbers_tape: then i'll leave it to our ptl to maintain | 20:49 |
*** rongze has joined #openstack-trove | 20:50 | |
denis_makogon | pub_tap, Michael))) | 20:50 |
pub_tap | denis_makogon: :) | 20:51 |
denis_makogon | looks like i missed huge party | 20:52 |
imsplitbit | pub_tap... | 20:52 |
denis_makogon | i've been away for 1h lol | 20:52 |
pub_tap | well datsun_F40PH changed his name and i wanted to tag along | 20:55 |
*** rongze has quit IRC | 20:56 | |
*** NehaV has joined #openstack-trove | 21:01 | |
*** vipul-away is now known as vipul | 21:05 | |
*** yogesh has joined #openstack-trove | 21:06 | |
pub_tap | konetzed: u suck! | 21:14 |
imsplitbit | what did he break now? | 21:15 |
*** ashestakov has quit IRC | 21:20 | |
pub_tap | he broke life | 21:28 |
pub_tap | cuz he sucks at it | 21:28 |
*** rnirmal has quit IRC | 21:31 | |
*** rongze has joined #openstack-trove | 21:52 | |
*** grapex has quit IRC | 21:53 | |
*** rongze has quit IRC | 21:59 | |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove: setting fake mode to print startup errors https://review.openstack.org/58440 | 21:59 |
openstackgerrit | Craig Vyvial proposed a change to openstack/trove-integration: fake mode script needs to be updated after changed were merged in trove https://review.openstack.org/58441 | 22:01 |
*** pub_tap is now known as hub_cap | 22:01 | |
*** pdmars has quit IRC | 22:05 | |
*** NehaV has quit IRC | 22:07 | |
*** NehaV has joined #openstack-trove | 22:07 | |
spredzy | Hi guys, I am getting started with trove, can someone points me to a doc/wiki, where I could find how to create a datastore please. | 22:21 |
spredzy | I've came accross https://wiki.openstack.org/wiki/Trove/trove-versions-types but it provided me on how to get a list of them, not how to create them | 22:21 |
spredzy | and trove help, didn't provide a datastore-create | 22:22 |
spredzy | so I am uncertain on what's the way to do so | 22:22 |
*** vipul is now known as vipul-away | 22:22 | |
*** vipul-away is now known as vipul | 22:22 | |
*** Barker has quit IRC | 22:26 | |
datsun_F40PH | No, we don't appear to have a create call in datastores just yet | 22:28 |
spredzy | so if I want to create a datastore, I should do it directly with an HTTP call atm? | 22:29 |
hub_cap | victory. got my stupid trackpad working | 22:29 |
hub_cap | UGH | 22:29 |
hub_cap | hey spredzy. as the operator you would use the trove-manage command | 22:30 |
datsun_F40PH | there you go | 22:30 |
hub_cap | https://github.com/openstack/trove-integration/commit/002ce73a07aca29fdc0b31257f5b5e0f1f5ec032 | 22:30 |
hub_cap | that command is where 2 dummy ones were created, so it shows usage | 22:30 |
*** amcrn has joined #openstack-trove | 22:32 | |
spredzy | hub_cap: thanks for the link, I do see the cmd_set_datastore(), will look into it closely | 22:33 |
robertmyers | hub_cap: trove-manage does not create new records | 22:33 |
robertmyers | only updates existing ones | 22:33 |
robertmyers | It is a bunch of fail | 22:33 |
robertmyers | see line 537 | 22:34 |
hub_cap | spredzy: awesome, and welcome | 22:34 |
hub_cap | robertmyers: huuuuuuuuuuuuuuuuuuh | 22:34 |
hub_cap | youre joking right robertmyers?? | 22:35 |
robertmyers | nope | 22:35 |
robertmyers | look at the code you just linked | 22:35 |
kevinconway | robertmyers: that's what rd-sql is for | 22:35 |
robertmyers | kevinconway: lol | 22:35 |
SlickNik | lol @ rd-sql | 22:36 |
robertmyers | hub_cap: https://github.com/openstack/trove/blob/master/bin/trove-manage#L79 | 22:36 |
robertmyers | the update will fail if it is not found | 22:36 |
robertmyers | no create method | 22:36 |
robertmyers | did we even review this ;) | 22:37 |
kevinconway | robertmyers: except it won't actually fail because it prints instead | 22:37 |
robertmyers | hahaha | 22:37 |
hub_cap | robertmyers: this is the problem w/ huge patches | 22:38 |
cp16net | yeah | 22:39 |
hub_cap | totally yet _another_ place we couldve broken this patch up | 22:39 |
robertmyers | agreed | 22:39 |
hub_cap | im tellin ya, i couldve broke that into like ~10 patches | 22:39 |
hub_cap | IM SO FREAKIN HAPPY | 22:39 |
hub_cap | ^ ^ happy | 22:39 |
hub_cap | <--- happy | 22:39 |
kevinconway | hub_cap: you should revert that commit | 22:39 |
robertmyers | hahaha | 22:39 |
hub_cap | kevinconway: lol | 22:39 |
spredzy | I'm so freaking lost ;) | 22:39 |
hub_cap | spredzy: easy to do in this oddball channel | 22:40 |
robertmyers | spredzy: so are we | 22:40 |
hub_cap | im happy for orthoginal reasons fwiw | 22:40 |
spredzy | thanks guys, I do feel better now | 22:40 |
kevinconway | does it involve food? | 22:40 |
hub_cap | kevinconway: are you saying the only thing that makes me happy is food? | 22:40 |
hub_cap | kevinconway: CUZ IF U ARE, OH BOY ARE U RIGHT | 22:40 |
robertmyers | nice | 22:40 |
hub_cap | but srsly, im happy cuz my trackpad now works on my c720 in jessie | 22:41 |
* cp16net larrups hub_cap with a concrete christmas tree | 22:41 | |
hub_cap | Bender: I've personalized each of your meals. Amy, you're cute, so I baked you a pony! | 22:41 |
cp16net | very festive | 22:41 |
* hub_cap spills some coffee over cp16net's keyboard | 22:41 | |
robertmyers | well, I have to run, laters! | 22:41 |
hub_cap | cu robertmyers | 22:41 |
*** robertmyers has quit IRC | 22:41 | |
cp16net | i think its about time for me to split out as well | 22:42 |
hub_cap | so spredzy might be just easiest to "insert into ..." | 22:42 |
cp16net | Bender: Pardon me, brother. Care to donate to the anti-mugging you fund? | 22:42 |
hub_cap | since the -manage call is 1/2 baked | 22:42 |
cp16net | Bender: Cheese it! | 22:42 |
cp16net | LOL | 22:43 |
hub_cap | woah | 22:43 |
hub_cap | that one is the BEST | 22:43 |
cp16net | Bender: Court’s kinda fun when it’s not my ass on the line! | 22:43 |
hub_cap | nope | 22:43 |
hub_cap | u sholdve quit | 22:43 |
cp16net | Homer: Getting out of jury duty is easy. The trick is to say you're prejudiced against all races. | 22:43 |
hub_cap | NOPE | 22:43 |
hub_cap | Bender: Cheese it! | 22:43 |
hub_cap | best ever | 22:43 |
*** kevinconway has quit IRC | 22:43 | |
hub_cap | having 2 laptops + a external mouse (cuz stupid c720 trackpad was borked) made me want to bust out synergy | 22:44 |
cp16net | yeah thats a nice proggy | 22:45 |
cp16net | alright i'm out | 22:45 |
cp16net | be on later | 22:45 |
spredzy | I am out for the day, ttyl. Thanks for the pointers | 22:49 |
*** spredzy has quit IRC | 22:49 | |
*** SnowDust has quit IRC | 22:49 | |
*** jasonb365 has quit IRC | 22:50 | |
*** rongze has joined #openstack-trove | 22:55 | |
*** denis_makogon has quit IRC | 22:56 | |
*** jcooley_ has joined #openstack-trove | 22:58 | |
*** jmontemayor has quit IRC | 22:59 | |
*** rongze has quit IRC | 22:59 | |
*** plodronio has left #openstack-trove | 23:07 | |
*** yogesh has quit IRC | 23:08 | |
*** yogesh has joined #openstack-trove | 23:08 | |
*** yogesh has quit IRC | 23:13 | |
*** datsun_F40PH has quit IRC | 23:15 | |
*** amytron has quit IRC | 23:16 | |
*** NehaV has quit IRC | 23:40 | |
*** vipul has quit IRC | 23:43 | |
*** vipul has joined #openstack-trove | 23:44 | |
*** rongze has joined #openstack-trove | 23:56 | |
*** jcooley_ has quit IRC | 23:56 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!