*** csoukup has joined #openstack-zaqar | 00:00 | |
*** Qiming has joined #openstack-zaqar | 00:20 | |
*** achanda_ has quit IRC | 00:31 | |
*** achanda has joined #openstack-zaqar | 00:33 | |
*** mdnadeem has joined #openstack-zaqar | 00:35 | |
*** achanda has quit IRC | 00:38 | |
*** itisha has joined #openstack-zaqar | 00:46 | |
*** achanda has joined #openstack-zaqar | 01:22 | |
*** achanda has quit IRC | 01:41 | |
*** Qiming has quit IRC | 01:42 | |
cpallares | flwang1: o/ | 01:42 |
---|---|---|
vkmc | flwang, o/ | 01:42 |
flwang1 | ? | 01:42 |
flwang1 | hi girls, what's up | 01:43 |
vkmc | flwang1, so we were talking about the Horizon use case for the websocket demo | 01:43 |
cpallares | flwang1: We're discussing the demo for the zaqar presentation and vkmc mentioned a poc for horizon | 01:43 |
vkmc | that yes ^ | 01:43 |
cpallares | lol jinxy | 01:43 |
flwang1 | haha | 01:43 |
flwang1 | what i can help? | 01:43 |
vkmc | and we would like to hear yours and flaper87's feedback | 01:43 |
cpallares | /s/jinxy/jinx | 01:43 |
flwang1 | yes, listening... | 01:43 |
vkmc | flwang1, cpallares found this etherpad from last summit https://etherpad.openstack.org/p/YVR-horizon-liberty-messaging | 01:44 |
cpallares | So we're looking over the blueprints from 2012 | 01:44 |
cpallares | and that link that vkmc just posted | 01:44 |
cpallares | lol | 01:44 |
vkmc | aahahaha | 01:44 |
vkmc | ok I'll stay back | 01:44 |
vkmc | cpallares, you tell him | 01:44 |
cpallares | vkmc: no no continue | 01:44 |
vkmc | so we were designing the poc, brainstorming on how it should look like | 01:45 |
*** c_soukup has joined #openstack-zaqar | 01:45 | |
flwang1 | reading the etherpad... | 01:45 |
flwang1 | may i get a summary? | 01:46 |
cpallares | Horizon wants to use zaqar for messaging | 01:46 |
cpallares | flwang1: such as deliving continous notifications to users, admins | 01:47 |
flwang1 | yes | 01:47 |
cpallares | and other things like standardizing error messages | 01:47 |
cpallares | which is what vkmc and I were discussing | 01:47 |
cpallares | Would the poc be creating one these features? | 01:48 |
cpallares | There seems to be so many things horizon could use with zaqar | 01:48 |
flwang1 | so which user case are you trying to focus? | 01:48 |
flwang1 | notifications to end user/admin or the error message stuff | 01:48 |
vkmc | they are kinda related | 01:48 |
*** csoukup has quit IRC | 01:49 | |
flwang1 | i would suggest focusing one | 01:49 |
vkmc | let's say errors | 01:49 |
flwang1 | ok | 01:49 |
cpallares | Well, they both need the same code to be built on top of it, no? | 01:49 |
flwang1 | cool | 01:50 |
flwang1 | then my question is how about the sizing/workload for the poc? | 01:50 |
flwang1 | there are about 40 days before the summit | 01:50 |
cpallares | flwang1: Do you think it's not realistic? | 01:51 |
flwang1 | no, i just want to figure out the scope and make sure it can be done before summit | 01:51 |
flwang1 | because it could be a really good demo | 01:52 |
vkmc | yes... its mine concern to | 01:52 |
vkmc | o | 01:52 |
flwang1 | to accelerate the integration with horizon | 01:52 |
cpallares | Well standardizing the error messages from different projects seems more ambitious than just sending messages within horizon | 01:52 |
flwang1 | so i think we need to define the user case clearly and confirm with horizon guys to get a reasonable scope | 01:53 |
flwang1 | cpallares: +1 | 01:53 |
vkmc | the thing is... the queue should be useful for processing them | 01:54 |
vkmc | and display useful information accordingly | 01:54 |
vkmc | also, I'm reading that a problem is that things are confusing because calls to underlying services are asynchronous | 01:55 |
vkmc | so maybe you are launching a nova instance and then you go to the admin menu | 01:55 |
vkmc | and you don't see the nova instance failure reason | 01:55 |
vkmc | so Horizon should be able to pull out from the queue according to the service | 01:56 |
flwang1 | vkmc: but does that mean horizon need to pool the queue? | 01:56 |
flwang1 | or we will leverage the notifications? | 01:56 |
vkmc | I dunno, that is on us I guess | 01:57 |
vkmc | one of the sides has to do the effort | 01:57 |
flwang1 | i have an meeting after 3 mins, will you be online after 45 mins? sorry about that | 01:57 |
*** njohnston is now known as nate_gone | 01:57 | |
vkmc | Horizon polls Zaqar for changes or we notify Horizon that they have a message for service X | 01:57 |
vkmc | np | 01:57 |
vkmc | yes I think so :) | 01:57 |
flwang1 | awesome | 01:57 |
flwang1 | thanks for following up this, vkmc | 01:58 |
vkmc | np, cpallares pinged me first | 01:58 |
vkmc | haha | 01:58 |
flwang1 | haha | 01:58 |
vkmc | team work \o/ | 01:58 |
cpallares | \o/ | 01:58 |
flwang1 | cpallares will be award a nz candy | 01:58 |
cpallares | :D | 01:59 |
*** flwang1 has quit IRC | 02:03 | |
*** flwang1 has joined #openstack-zaqar | 02:22 | |
*** david-lyle has joined #openstack-zaqar | 02:34 | |
*** achanda has joined #openstack-zaqar | 02:41 | |
*** achanda has quit IRC | 02:45 | |
flwang1 | back | 03:00 |
flwang1 | vkmc: cpallares: are you around? | 03:00 |
vkmc | flwang1, here! | 03:00 |
flwang1 | vkmc: so I would suggest have a chat with horizon guys and define the scope | 03:02 |
flwang1 | how do you think? | 03:02 |
cpallares | o/ | 03:02 |
* vkmc approves | 03:02 | |
cpallares | flwang1: in a meeting, their channel, mailing list? | 03:02 |
vkmc | seems that mrunge was working on that | 03:02 |
vkmc | according to YVR etherpad | 03:03 |
flwang1 | i prefer IRC | 03:03 |
vkmc | me too | 03:03 |
cpallares | seems more quickly to do it that way too | 03:03 |
*** achanda has joined #openstack-zaqar | 03:03 | |
flwang1 | cool, vkmc do know mrunge? | 03:04 |
vkmc | we better catch them asap | 03:04 |
vkmc | I do know him | 03:04 |
flwang1 | awesome | 03:04 |
cpallares | vkmc: awesome | 03:04 |
vkmc | so I'll ping him tomorrow | 03:05 |
vkmc | and maybe arrange a meeting | 03:05 |
flwang1 | it would be excellent | 03:05 |
flwang1 | i will try to ping him/her either | 03:05 |
flwang1 | and i will go through the etherpad to understand the user case | 03:05 |
vkmc | cool | 03:06 |
vkmc | I think we have some highlights at least | 03:06 |
vkmc | 1. operations being asynchronous and for that reason error message display is confusing | 03:07 |
vkmc | 2. information displayed in error messages is not user friendly (hence the need of a processing before display) | 03:07 |
vkmc | 3. order or events (a big ? here... since I don't think we can enforce any ordering there... FIFO) | 03:08 |
vkmc | no | 03:08 |
vkmc | yeah is FIFO | 03:08 |
vkmc | o.o | 03:08 |
flwang1 | you need some coffee :) | 03:09 |
cpallares | lol | 03:09 |
cpallares | or a nap | 03:09 |
cpallares | an 8 hour nap | 03:09 |
vkmc | hahaha | 03:10 |
vkmc | both O.O | 03:10 |
cpallares | haha | 03:11 |
flwang1 | the highlights looks good for me, and personally, i would like to focus on the asynchronous issue | 03:12 |
flwang1 | but i need some test | 03:12 |
vkmc | yeah | 03:13 |
flwang1 | vkmc: cpallares: can you correct me? if i create a instance on horizon and switch the pages to another panel, then as an end user, i won't be notified if the instance is created successfully or not, right? | 03:13 |
vkmc | flwang1, exactly yes | 03:13 |
vkmc | and if an error occurs likewise | 03:13 |
flwang1 | cool | 03:14 |
flwang1 | and i may need to keep another thing in mind, to resolve this, if zaqar is too heavy. my point is, it would be nice if the poc code could be wrapper | 03:15 |
flwang1 | then zaqar could be a driver | 03:15 |
flwang1 | we may get the challenge from the audience | 03:15 |
flwang1 | it's exactly like the notification middleware of swfit | 03:16 |
flwang1 | does that make any sense? | 03:17 |
cpallares | flwang1: what challenge from the audience? | 03:19 |
vkmc | it does yes | 03:19 |
cpallares | flwang1: on the design? | 03:19 |
flwang1 | cpallares: i mean, to avoid somebody say: no, we don't need zaqar, we need a more lightweight lib/script to do that instead of deploying a zaqar service | 03:20 |
flwang1 | or something like that | 03:20 |
flwang1 | oh, my English :( | 03:20 |
cpallares | flwang1: No, no I understand you, but I didn't understand what challenge they would put up. | 03:21 |
flwang1 | that's maybe the challenge from Horizon team, just think loud | 03:22 |
flwang1 | s/challenge/question | 03:22 |
cpallares | flwang1: I guess that would be good to discuss with Horizon people | 03:24 |
cpallares | We can do that tomorrow | 03:24 |
cpallares | Or when they're online. | 03:24 |
flwang1 | sure | 03:25 |
cpallares | vkmc, flwang1: Ok, I'm heading out o/ | 03:25 |
cpallares | vkmc: We now have an action plan. | 03:26 |
cpallares | :D | 03:26 |
vkmc | sure | 03:28 |
vkmc | :) | 03:28 |
vkmc | thanks guys | 03:28 |
vkmc | ttyt o/ | 03:28 |
flwang1 | thank you so much, girls | 03:28 |
*** achanda has quit IRC | 03:35 | |
*** achanda has joined #openstack-zaqar | 03:35 | |
*** boris-42 has quit IRC | 03:50 | |
*** boris-42 has joined #openstack-zaqar | 03:51 | |
*** achanda has quit IRC | 04:39 | |
*** flwang1 has quit IRC | 05:04 | |
*** achanda has joined #openstack-zaqar | 05:20 | |
openstackgerrit | MD NADEEM proposed openstack/zaqar: Correct "log_file" option in zaqar documents https://review.openstack.org/224430 | 05:30 |
*** c_soukup has quit IRC | 05:31 | |
*** mdnadeem has quit IRC | 05:43 | |
*** mdnadeem has joined #openstack-zaqar | 06:06 | |
openstackgerrit | OpenStack Proposal Bot proposed openstack/zaqar: Imported Translations from Zanata https://review.openstack.org/224260 | 06:08 |
*** Qiming has joined #openstack-zaqar | 06:26 | |
zigo | therve: Did you find out how to fix the never closed connections to mongo? | 07:21 |
therve | zigo, Yeah, in progress :) | 07:21 |
zigo | Ok, will wait then. | 07:21 |
openstackgerrit | Flavio Percoco proposed openstack/zaqar: Allow running zaqarclient for gate tests https://review.openstack.org/221068 | 07:26 |
flaper87 | zigo: therve you guys rock | 07:26 |
flaper87 | promise to get beers from you | 07:26 |
flaper87 | therve: i'll send yours to your place | 07:26 |
flaper87 | no worries | 07:26 |
therve | flaper87, Just visit the Paris office already | 07:28 |
flaper87 | therve: I can do that, I can totally do that. :D | 07:29 |
openstackgerrit | Thomas Herve proposed openstack/zaqar: Close backend connections in tests https://review.openstack.org/224466 | 07:34 |
*** achanda has quit IRC | 07:38 | |
*** achanda has joined #openstack-zaqar | 07:47 | |
*** achanda has quit IRC | 07:50 | |
therve | flaper87, Closing seems to make tests much slower :/ | 08:01 |
flaper87 | therve: closing connections ? | 08:01 |
flaper87 | ah yeah | 08:02 |
flaper87 | just saw the patch | 08:02 |
flaper87 | mmh | 08:02 |
-openstackstatus- NOTICE: Gate is currently stuck, failing grenade upgrade tests due the release of oslo.utils 1.4.1 for Juno. | 08:02 | |
zigo | therve: Should I try this patch now? | 08:05 |
zigo | therve: YOUR MY HERO !!! | 08:10 |
zigo | \o/ | 08:10 |
zigo | (ie: it's now working...) | 08:10 |
zigo | And ... no ... | 08:11 |
zigo | UNIT TEST FAILURE ! :) | 08:11 |
flaper87 | zigo: but just the tests, not the connection thingy | 08:14 |
flaper87 | right? | 08:14 |
flaper87 | what's the unittest failure? | 08:14 |
flaper87 | zigo: thanks for your patience | 08:14 |
zigo | flaper87: Well, I could never run many unit tests. | 08:14 |
zigo | Now, all of them pass without issue. | 08:14 |
zigo | So everything's green ! :) | 08:14 |
zigo | Now, got to test zaqar a bit ... | 08:14 |
flaper87 | zigo: aaaaaaaah, you said and no unittest failure | 08:14 |
zigo | Make sure all daemons are up and running. | 08:14 |
zigo | Just before, I wrote "And ... no ..." | 08:15 |
flaper87 | yup | 08:15 |
flaper87 | just got it | 08:15 |
* flaper87 slow | 08:15 | |
flaper87 | zigo: glad to hear that | 08:15 |
flaper87 | I'll declare therve as the Liberty Zaqar hero | 08:15 |
zigo | What is the zaqar-gc for? | 08:16 |
zigo | I don't have a keystone_authtoken when generating zaqar.conf ... :( | 08:19 |
zigo | It should be --namespace keystonemiddleware.auth_token and not --namespace keystoneclient.middleware.auth_token. | 08:24 |
zigo | IMO, etc/oslo-config-generator/zaqar.conf should be fixed. | 08:25 |
zigo | Is zaqar/transport/wsgi/app.py what is used for zaqar-api ? | 08:26 |
zigo | flaper87: ^ | 08:26 |
flaper87 | zigo: I'll fix the config-gen thing | 08:28 |
flaper87 | zigo: that's the wsgi app used to run zaqar under different wsgi containers (mod_wsgi, uwsgi, etc) | 08:28 |
zigo | flaper87: Right, but is this for the zaqar-api thing? | 08:42 |
zigo | Or does the zaqar-server includes the API too? | 08:42 |
zigo | What I already did for Barbican was using uwsgi in a startup script, so I'm trying to figure out what to do here. | 08:43 |
flaper87 | zigo: there's no zaqar-api, there's just zaqar-server and it includes the API | 08:43 |
zigo | Ok, thanks. | 08:43 |
flaper87 | zigo: you can use uwsgi for zaqar | 08:43 |
flaper87 | and skip running zaqar-server | 08:43 |
flaper87 | zigo: in devstack/plugin.sh you'll find how to run zaqar w/ uwsgi | 08:43 |
flaper87 | but you probably know how to do that already | 08:44 |
zigo | I'm fine with zaqar-server, those who want to implement wsgi can do it in Apache. | 08:44 |
zigo | Yup! :) | 08:44 |
* zigo thinks typing zaqar is to much effort for the left hand little finger... | 08:46 | |
zigo | Is there a howto install somewhere? | 08:47 |
zigo | (so I can make it useless ... :)) | 08:47 |
zigo | flaper87: What's the endpoint type, port, and URL for Zaqar? | 08:52 |
flaper87 | zigo: this is probably the best we have as to how run zaqar-server https://github.com/openstack/zaqar#running-a-local-zaqar-server-with-mongodb | 08:54 |
flaper87 | zigo: port is 8888, endpoint type is messaging | 08:54 |
flaper87 | http://$HOST:8888/ | 08:54 |
*** flwang has quit IRC | 09:27 | |
*** flwang has joined #openstack-zaqar | 09:41 | |
*** flwang1 has joined #openstack-zaqar | 10:03 | |
*** Qiming has quit IRC | 10:12 | |
*** flwang1 has quit IRC | 10:15 | |
*** flwang1 has joined #openstack-zaqar | 10:15 | |
-openstackstatus- NOTICE: Gate back to normal, thanks to the backlisting of the problematic version | 10:16 | |
flwang1 | vkmc: ping | 11:10 |
*** Qiming has joined #openstack-zaqar | 11:10 | |
*** malini has joined #openstack-zaqar | 12:15 | |
*** nate_gone is now known as njohnston | 12:23 | |
*** malini has quit IRC | 12:34 | |
*** achanda has joined #openstack-zaqar | 12:35 | |
*** malini has joined #openstack-zaqar | 12:52 | |
*** sriram has joined #openstack-zaqar | 13:03 | |
*** dynarro has joined #openstack-zaqar | 13:18 | |
*** malini has quit IRC | 13:20 | |
*** openstackgerrit has quit IRC | 13:31 | |
*** openstackgerrit has joined #openstack-zaqar | 13:31 | |
flaper87 | flwang1: congrats, you're now Zaqar's PTL | 13:39 |
*** achanda has quit IRC | 13:40 | |
vkmc | we have a new PTL? | 13:40 |
vkmc | white smoke can be seen in Zaqar's chimney | 13:40 |
flaper87 | vkmc: LOOOOOL | 13:41 |
flaper87 | good one | 13:41 |
flaper87 | but hey, I'm not dead | 13:41 |
flaper87 | I'm just going to party | 13:41 |
vkmc | haha we don't need you dead! | 13:41 |
*** JAHoagie has joined #openstack-zaqar | 13:57 | |
*** kgriffs|afk is now known as kgriffs | 13:57 | |
*** JAHoagie has quit IRC | 14:02 | |
openstackgerrit | Thomas Herve proposed openstack/zaqar: Switch to oslo_cache https://review.openstack.org/224745 | 14:11 |
*** JAHoagie has joined #openstack-zaqar | 14:13 | |
*** JAHoagie has quit IRC | 14:19 | |
*** achanda has joined #openstack-zaqar | 14:22 | |
*** achanda has quit IRC | 14:25 | |
*** kgriffs is now known as kgriffs|afk | 14:40 | |
*** kgriffs|afk is now known as kgriffs | 14:40 | |
sriram | flwang: congrats! | 14:44 |
*** achanda has joined #openstack-zaqar | 14:52 | |
*** flwang1 has quit IRC | 15:01 | |
*** Qiming has quit IRC | 15:07 | |
*** achanda has quit IRC | 15:09 | |
*** kgriffs is now known as kgriffs|afk | 15:14 | |
*** amitgandhinz has joined #openstack-zaqar | 15:24 | |
*** csoukup has joined #openstack-zaqar | 15:25 | |
*** ryansb has quit IRC | 15:46 | |
*** ryansb has joined #openstack-zaqar | 15:48 | |
*** ryansb has quit IRC | 15:48 | |
*** ryansb has joined #openstack-zaqar | 15:48 | |
*** kgriffs|afk is now known as kgriffs | 15:59 | |
*** dynarro has quit IRC | 16:05 | |
*** mpanetta_ has joined #openstack-zaqar | 16:12 | |
*** kgriffs is now known as kgriffs|afk | 16:27 | |
*** stanchan has joined #openstack-zaqar | 16:33 | |
*** david-lyle has quit IRC | 16:36 | |
*** david-lyle has joined #openstack-zaqar | 16:47 | |
*** david-lyle has quit IRC | 17:03 | |
*** achanda has joined #openstack-zaqar | 17:39 | |
*** kgriffs|afk is now known as kgriffs | 18:07 | |
*** itisha has quit IRC | 18:09 | |
cpallares | flwang: Congrats PTL! | 18:53 |
*** david-lyle has joined #openstack-zaqar | 19:11 | |
*** mpanett__ has joined #openstack-zaqar | 19:14 | |
*** mpanett__ has quit IRC | 19:14 | |
*** david-lyle has quit IRC | 19:25 | |
*** flwang1 has joined #openstack-zaqar | 19:28 | |
*** mpanetta_ has quit IRC | 19:42 | |
*** flwang1 has quit IRC | 19:51 | |
*** mpanetta_ has joined #openstack-zaqar | 19:57 | |
openstackgerrit | Nate Johnston proposed openstack/zaqar: Use decorator instead of request logging in responders https://review.openstack.org/224882 | 19:59 |
njohnston | congrats fwlang1 | 19:59 |
*** stanchan has quit IRC | 20:08 | |
*** stanchan has joined #openstack-zaqar | 20:09 | |
zigo | What build dependency did I forget? https://liberty-jessie.pkgs.mirantis.com/job/zaqar/2/console | 20:11 |
zigo | Oh, it's building the wrong branch, sorry. | 20:12 |
zigo | :) | 20:12 |
*** fifieldt has quit IRC | 20:19 | |
*** fifieldt has joined #openstack-zaqar | 20:20 | |
openstackgerrit | Nate Johnston proposed openstack/zaqar: Use decorator instead of request logging in responders https://review.openstack.org/224882 | 20:30 |
*** flwang1 has joined #openstack-zaqar | 21:05 | |
*** david-lyle has joined #openstack-zaqar | 21:06 | |
*** david-lyle has quit IRC | 21:09 | |
flwang | vkmc: ping | 21:16 |
vkmc | flwang, pong | 21:16 |
flwang | i just had a chat with mrunge last night | 21:16 |
vkmc | sweet! | 21:16 |
flwang | and got some useful information | 21:16 |
vkmc | I couldn't catch him today | 21:16 |
vkmc | cpallares, ^ | 21:16 |
* vkmc is all eyes | 21:16 | |
flwang | i think what we can work on the is async user case | 21:17 |
flwang | currently, the problem of horizon is it has always polling to get the status change for instance, image and volume | 21:18 |
flwang | for example, if you just create an image, and keep the page opening, horizon has to poll per 2.5 sec to send image-get request to glance to get the latest status | 21:19 |
vkmc | aha... | 21:20 |
flwang | based on my understanding, it would be really nice if we can use notification, since if we just use the queue/message of zaqar, then horizon still has to poll to get the message, am i missing anything? | 21:20 |
vkmc | it sounds good | 21:21 |
flwang | but as for notification, we only support webhook and email | 21:21 |
vkmc | and therve left in the queue websocket | 21:21 |
flwang | so we may need an endpoint in horizon as the webhook so that zaqar can notifiy/push the message to it | 21:22 |
flwang | vkmc: this one https://review.openstack.org/#/c/206080/8 ? | 21:22 |
vkmc | yeah | 21:22 |
vkmc | and yeah | 21:22 |
flwang | so we may need a new publisher driver in ceilometer to forward the notification to zaqar to avoid changing code in nova, cinder and glance | 21:24 |
vkmc | right now status changes are managed by ceilometer in horizon? | 21:24 |
flwang | and horizon should be able to use current way if there is no ceilometer and zaqar | 21:24 |
vkmc | I thought it directly connected to the original services | 21:24 |
flwang | vkmc: no, currently horizon is polling the status to the service directly, by client | 21:25 |
flwang | like call glance client to get the image status | 21:25 |
flwang | we can directly let glance post message to zaqar | 21:25 |
*** sriram has quit IRC | 21:25 | |
flwang | but with that way, we may need similar change in glance, nova, cinder, etc.... | 21:26 |
vkmc | exactly, I was thinking on that | 21:26 |
vkmc | yeah | 21:26 |
flwang | now all the services are publishing message/notification to ceilometer, so we could be able to add a new driver in ceilometer to publish/forward them into zaqar | 21:28 |
*** kgriffs is now known as kgriffs|afk | 21:30 | |
vkmc | cool | 21:32 |
vkmc | thing is... what kind of messages/notifications the services send to ceilo? | 21:32 |
flwang | good pint | 21:37 |
flwang1 | i need to review the glance code, i don't think glance will send notifications for each status change | 21:47 |
flwang1 | it's a good question | 21:47 |
flwang1 | but you got my point, the basic workflow should be like i mentioned above | 21:48 |
*** amitgandhinz has quit IRC | 21:48 | |
*** mpanetta_ has quit IRC | 22:02 | |
*** achanda has quit IRC | 22:11 | |
*** csoukup has quit IRC | 22:15 | |
*** achanda has joined #openstack-zaqar | 22:26 | |
*** csoukup has joined #openstack-zaqar | 22:55 | |
*** david-lyle has joined #openstack-zaqar | 22:59 | |
*** stanchan has quit IRC | 23:18 | |
*** csoukup has quit IRC | 23:23 | |
*** Qiming has joined #openstack-zaqar | 23:38 | |
*** flwang2 has joined #openstack-zaqar | 23:46 | |
*** flwang1 has quit IRC | 23:48 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!