/srv/irclogs.ubuntu.com/2015/09/17/#ubuntu-ci-eng.txt

bzoltanis here anybody who knows (remembers) how to convince mako to keep adb available even when the screen is locked? Flashing the latest rc-proposed + touching /userdata/.adb_onlock does not make it.04:32
bzoltanrobru: ^04:32
robruUhhh04:34
robrubzoltan: i dunno. Maybe jibel or fginther might know more about the device side.04:35
bzoltanrobru: thank you.. hopefully someone can help.. it would be nice to put my mako back to work as contnious testing agent ...04:37
bzoltanI remember ogra mentioned that we might need a special firmware for this... but not sure04:38
robrubzoltan: maybe ask cihelp if they can share the secret of how they test with devices.04:39
bzoltanrobru: for Mirv it works04:40
Mirvyeah for some reason I don't have that problem, and I've been flashing a lot again now this week.04:41
robrubzoltan: Mirv should be around soon04:41
bzoltanrobru:  he is :) already04:43
iahmadevanwang, ping05:13
evanwangiahmad, pong05:13
evanwangiahmad, failed to connect canonical irc server from our site, how is the connection from you site?05:20
iahmadevanwang, working for me05:20
evanwangiahmad, seems something wrong at our site :(05:21
bzoltanrobru: Mirv: this failure seems to be serious -> https://launchpadlibrarian.net/218108553/buildlog_ubuntu-wily-amd64.ubuntu-ui-toolkit_1.3.1639%2B15.10.20150916.2-0ubuntu1_BUILDING.txt.gz  "/usr/bin/ld.gold: --push-state: unknown option"06:09
Mirvbzoltan: FYI robert is on vacation06:10
Mirvbzoltan: eh..06:11
bzoltanMirv:  ohh... i am sorry robru06:11
Mirvbzoltan: I have never seen that06:12
bzoltanMirv:  me neither... it is new to me06:12
Mirvlet's see if anyone else has that on devel06:16
dbarthmardy: hi07:20
dbarthmardy: about facebook, i think the solution is to get a pro account07:21
mardydbarth: hi!07:21
mardydbarth: let's wait a bit, I'm discussing the issue in the forums07:21
dbarthmardy: i saw a link that offers to go pro, and then you get an "account manager", ie a human, to help with the review07:21
dbarthah, do you?07:21
dbarthdo you have a pointer, are they replying ?07:21
mardydbarth: some users are replying, but I believe that one might actually be a facebook employee07:22
mardydbarth: https://www.facebook.com/groups/fbdevelopers/07:22
dbarthmardy: i can't find the link to that pro contact, as it was linked from the failed review page07:40
mardydbarth: yes, it disappeared. But let's wait for the results of this review07:41
sil2100pete-woods: hey! Did you manage to test the package from the silo I pasted yesterday? :)07:42
dbarthmardy: but if next time it fails please keep it as is, and we will check that path07:43
pete-woodssil2100: sorry, I didn't read my scrollback07:44
pete-woodswill have a look now07:44
mardydbarth: ok07:44
pete-woodssil2100: just snapshotting the VM now07:45
sil2100Thanks :) I doubt this will work, but at least we'll know that the latest is needed07:45
pete-woodsasdas07:46
pete-woodswhoops07:46
pete-woodsthat's fine07:46
pete-woodshappy to do exploratory testing of it07:46
pete-woodssil2100: the PPA seems to fix the problem! :)07:58
sil2100pete-woods: uh oh!08:00
sil2100Who would have thought, a no-change rebuild08:00
pete-woodssil2100: is that a bad thing?08:00
sil2100pete-woods: it's good, just didn't expect that ;p08:00
sil2100Let me copy it to the overlay then08:01
pete-woodsthanks08:03
bzoltanMirv: ehh... 2015-09-17 08:19:47,420 ERROR dch: fatal error at line 1141:08:22
bzoltanNew version specified (1.3.1639+15.04.20150917-0ubuntu1) is less than08:22
bzoltanthe current version number (1.3.1639+15.10.20150916-0ubuntu1)!  Use -b to force.08:22
sil2100bzoltan: yeah, that happens when you want to turn a wily trunk into a vivid one - what do you want to release exactly?08:23
bzoltansil2100:  I want a dual landing like normally.. but wily is blocked so I could settle with a vivid only08:23
Saviqcihelp, hey, it looks like unity8's autopilot on wily/mako is busted, ENOSPC when installing dependencies http://s-jenkins.ubuntu-ci:8080/job/generic-deb-autopilot-runner-wily-mako/380/console08:24
Mirvsil2100: there's a bug #1496743 introduced by new GCC 5 in wily, not sure how it should be fixed08:27
ubot5bug 1496743 in qtbase-opensource-src (Ubuntu) "/usr/bin/ld.gold: --push-state: unknown option" [Undecided,New] https://launchpad.net/bugs/149674308:27
psivaaSaviq: I dont think i saw this on Mako before,  would need to take a closer look. How urgent is this?08:29
psivaai.e. can this wait until late this afternoon?08:29
Saviqpsivaa, yeah08:29
psivaaSaviq: ack, thanks08:29
Saviqpsivaa, apparently an issue for quite some time now, but we didn't pay close enough attention to wily http://s-jenkins.ubuntu-ci:8080/job/generic-deb-autopilot-runner-wily-mako/376/console08:30
psivaaSaviq: right, i see08:30
psivaaSaviq: will add this to the list. We'll look into it08:31
Saviqpsivaa, thanks08:31
Saviqpsivaa, likely the wily image just grew too much08:31
Saviqtrainguards, can we please have publish on https://requests.ci-train.ubuntu.com/#/ticket/359 ?08:42
sil2100Saviq: on it08:43
sil2100dobey: hey! Looking at silo 29 now - did those changes land in wily already, or are there any reasons for it not to work there?08:53
sil2100ogra_: hey! :) I want to upload the new ubuntu-touch-meta source to wily that includes the new pocket-desktop seed - can I have you looking at the debdiff before I upload it?09:04
ogra_sil2100, after the mtng i'm currently in09:05
sil2100ogra_: thanks, let me prepare teh diff09:05
sil2100ogra_: the debdiff - the changes are rather straightforward: http://paste.ubuntu.com/12436400/09:06
sil2100The generated meta package has proper deps09:06
ogra_sil2100, the diff looks ok to me09:20
sil2100ogra_: thanks :)09:20
Laneyis there such a thing as deleting a request?09:37
Laney(not assigned)09:37
Laneyor do they just stay foreverrrrrrrr09:37
sil2100Damn, using SUBPROJECT in livecd-rootfs is soo tempting, making ubuntu-pd a separate project would mean sooo much copy & paste09:37
sil2100Laney: yeah, there's a manual method for that09:37
sil2100Laney: you edit the request and change its status manually to 'Abandoned'09:38
sil2100(that's the official way to do it)09:38
Laneywhich drop down has that?09:38
* Laney can't find it09:38
sil2100Laney: it's the 'Status' field, no drop-down there09:39
sil2100Laney: if you hover over it it says 'Do not touch, auto updated', but that's just for already assigned silos09:39
Laneyok, I did that, thanks!09:39
sil2100yw!09:39
Laneyno silos :(09:39
sil2100Oh! Let me free up mine09:39
sil2100Laney: silo 35 should be free now09:41
Laneythanks09:41
sil2100Ugh, that's a bug that needs to be fixed ^09:41
Laneyit's like building more roads eh09:41
Laneyseb128: on the no free silos, the libusermetrics one failed to clean, do you know why?09:42
Saviqsil2100, I remember reading something about boot failures in wily, will the boottest failure prevent migration? http://people.canonical.com/~ubuntu-archive/proposed-migration/wily/update_excuses.html#unity809:45
seb128Laney, no, I don't09:50
sil2100Saviq: I think it will... boottests currently are a bit a pain in the ass10:52
Saviqsil2100, how do we deal with that?10:53
SaviqFWIW the unity-scope-click tests seem to take a long time, too10:53
Mirvsil2100: any idea about the part of 029 that it claims it requires qtmir from 027 but davmor2 approved 029 anyway so apparently nothing hugely crashy?10:53
sil2100Mirv: not sure, I was waiting for dobey to comment on this landing, since I didn't know if the same things landed in wily already or not10:54
sil2100Mirv: btw. do you have power to re-run boottests?10:54
sil2100Saviq: I think this requires cihelp, but from what I know currently boottests are unmaintained really ;/10:54
psivaasil2100: Saviq: Which job is this?10:57
Mirvsil2100: yes, just give me boottest urls10:57
psivaaor package10:57
Saviqpsivaa, unity810:57
Saviqpsivaa, http://people.canonical.com/~ubuntu-archive/proposed-migration/wily/update_excuses.html#unity810:57
Mirvsil2100: but some of them are pretty permanently failing (lately) so needs psivaa anyway10:57
psivaaSaviq: i've kicked this off again, and watch it10:59
Saviqpsivaa, thanks10:59
psivaaSaviq: Sorry for this long txt, regarding previous failure:11:01
psivaa"Tested this manually and found that the UI does not come up when testing unity8. The device is accessible via 'adb shell' so it does pass the basic boottest requirement, it's just failing on the adt-run check for a working UI.11:01
psivaaThe device did come up fully when the rest of the silo containing unity8 was installed. This points to a missing dependency. I discussed this with the unity8 team and this was expected and the silo will all be promoted together."11:01
davmor2Mirv: I can look at 27 now but everything worked like it has before.11:02
Saviqpsivaa, I think this was resolved by now11:04
psivaaSaviq: ack, let me see if the rerun works11:04
Saviqpsivaa, btw, (where) can I monitor adt runs in progress? like the unity-scope-click ones from http://people.canonical.com/~ubuntu-archive/proposed-migration/wily/update_excuses.html#unity811:07
Mirvdavmor2: ok. it's not yet set ready for QA though.11:08
Mirvdavmor2: oh, it is, sorry11:08
psivaaSaviq: http://d-jenkins.ubuntu-ci:8080/view/Wily/view/BootTest/job/wily-boottest-unity8/23/console, but this would not show anything until the job actually finishes11:08
rvrabeato: ping11:08
davmor2Mirv: its' in our queue :P11:09
abeatorvr, pong11:09
rvrabeato: Regarding the failures...11:09
rvrabeato: Can you confirm they work well in your case?11:09
abeatorvr, call forwarding works fine for pepephone11:10
rvrabeato: I also had problems with USSD11:10
abeatorvr, hmm, I tried some ussd strings for vodafone and all looked one11:11
abeato*good11:11
abeatorvr, what failed?11:11
rvrabeato: The first case of the test plan11:12
rvrabeato: I got an unsupported error11:13
abeatorvr, which ussd string and which operator?11:13
rvrabeato: Yoigo11:15
rvrabeato: “*118#”11:15
abeatorvr, ussd strings are operator-specific, “*118#” works only for vodafone (ES), I'll make that more explicit in the test11:17
rvrabeato: Ahh11:17
rvrabeato: Ok, I see... googling for yoigo's ones11:18
Saviqpsivaa, that's only boottest, how about the package ones?11:19
psivaaSaviq: you mean 'autopkgtest' ?11:19
Saviqpsivaa, yeah, like unity8 is waiting for unity-scope-click results for a few hours now11:20
psivaaif that's the case that's now handled by the release team, I think.11:20
Saviqwanted to check what's going on11:20
Saviqright11:20
abeatorvr, if you find something for yoigo please add that to the test case11:20
abeatorvr, note that ussd codes are for services like asking your balance, data consumed,etc. some time operators use different means for that (say, sms), so maybe yoigo does not have ussd strings11:21
* sil2100 jumps out for some lunch groceries11:39
rvrabeato: send-ussd "*111#" dbus.exceptions.DBusException: org.freedesktop.DBus.Error.UnknownMethod: Method "GetProperties" with signature "" on interface "org.ofono.SupplementaryServices" doesn't exist11:49
abeatorvr, is that krillin11:50
abeato?11:50
abeatorvr, if that's the case try "send-ussd  /ril_0  "*111#""11:50
rvrabeato: Yes, krillin11:52
rvrIt worked... somehow USSD: Yoigo Info: Este servicio no esta disponible para clientes de contrato.11:52
=== alan_g is now known as alan_g|lunch
rvrGoogle results are crappy lately12:02
rvrabeato: I can't check the USSD menu test, and "3. Receive a USSD notification" fails for me12:17
rvrabeato: I don't see "{SupplementaryServices} [/ril_0] NotificationReceived" in the logs12:17
=== alan_g|lunch is now known as alan_g
abeatorvr, in krillin you need to specify which modem slot to use when executing the script, that's why you need either /ril_0 or /ril_1 as argument13:14
dobeysil2100: hmm?13:15
abeatorvr, for the third test you need an operator that provides those notifications, again, that's operator specific13:15
abeatorvr, and that's not very common, I know only Tesco UK as an operator that does that13:15
dobeysil2100: what's the problem with 29?13:16
dobeysil2100: anyway, yes, the changes there are already in wily13:19
sil2100dobey: excellent13:20
sil2100That's what I wanted to know13:20
Saviqpsivaa, any word on the boottest? and any idea who can I ask about autopkgtests?13:21
sil2100ogra_: heeeey, I know you're probably super busy as always, but - could you maybe take a look at this merge? :) https://code.launchpad.net/~sil2100/livecd-rootfs/pocket-desktop/+merge/27144813:23
sil2100ogra_: I know I'm pushing on this subproject idea a bit too much, I can seriously change that to a separate project, just want to know if there are any objections to do it this way - as the change now is only 3 lines13:24
psivaaSaviq: still looking at the boottest issue13:31
psivaaafaik, pitti was handling the autopkgtests13:32
pstolowskihey trainguards, any idea what's wrong with silo 35 - https://ci-train.ubuntu.com/job/ubuntu-landing-035-1-build/55/console ?13:37
sil2100pstolowski: hmm, strange, looks like there's a problem with dh, even though I see it in the Build-Depends13:39
sil2100pstolowski: ah, maybe the change wasn't there when the silo was built?13:41
sil2100pstolowski: since I don't see the train installing debhelper13:41
pstolowskisil2100, no, debhelper was definately there. note however that i'm testing some crazy stuff in this silo (control file is re-generated from control.in)13:44
pstolowskilet me try to rebuild just in case13:45
sil2100kenvandine: hey! Could you take care of https://ci-train.ubuntu.com/job/ubuntu-landing-029-2-publish/45/ ? I have no power here ;)13:49
pstolowskisil2100, hmm, failed, debhelper should be installed, see the MP https://code.launchpad.net/~stolowski/unity-scopes-shell/singletree/+merge/27131413:51
kenvandinesil2100, sure13:51
sil2100kenvandine: thanks :)13:51
sil2100pstolowski: interesting... it seems as if the train is unable to find any build deps13:53
sil2100Let me check this locally here13:53
pstolowskisil2100, we did same trick in singletree branch of unity-scopes-api, and it worked13:58
kenvandinesil2100, exception while publishing14:05
sil2100pstolowski: indeed, so far everything looks fine, but somehow it's not resolving the missing build-deps14:11
sil2100kenvandine: ouch14:11
rvrawe_: abeato: Approving the silo14:13
awe_thanks rvr!14:14
abeatorvr, \o/ thanks14:14
kenvandinesil2100, let me know if you need me to try to publish that again14:19
sil2100kenvandine: thanks, I see the packages have reached the destination so I just merge&cleaned it14:23
kenvandineoh, cool14:23
dbarthhey there, i can't find that in the faq, but is there a quick way to force rebuild a package in a silo from bileto?14:28
dbarthie, i would do a null merge-proposal or something, but is there something cleaner14:28
dbarththis is to get a new build of a webapp-container, with the newer platform-api lib we have in a silo14:28
sil2100dbarth: I think the no-change MP is the fastest, not counting poking someone to prepare the package manually ;)14:36
dbarthaye aye sir14:37
dbarthgnn14:49
pstolowskisil2100, is "debconf: delaying package configuration, since apt-utils is not installed" relevant for that failure in silo 35?15:00
pstolowskisil2100, btw, any news on silo 10? can we get somebody to push the magic button? ;) we need to that in asap, it's blocking other stuff...15:25
sil2100pstolowski: poking!15:26
fgintherSaviq, hello, any suggestions on the unity8 wily phone test running out of space?15:59
Saviqfginther, only thing I can think of is that the wily image grew too much for the unity8 autopilot dependencies to fit on top15:59
fgintherSaviq, that's what I thought too. I don't know if there is anything we can do in CI to make this work... I think we just need a resized or cleaned up image16:01
Saviqfginther, yeah most likely, we need a bug and some investigation to happen16:01
fgintherSaviq, If the test is of no use, can we disable it for now to free up the resources?16:02
Saviqfginther, let me just confirm this is what's happening16:02
fgintherSaviq, thanks16:03
jibelsil2100, is there a landing meeting?16:03
davmor2sil2100: popey meeting????16:04
sil2100Oh crap!16:04
sil2100Be right there16:04
SaviqI've 500MB free on a freshly installed wily (dual booted, though)16:04
psivaaSaviq: btw, http://people.canonical.com/~ubuntu-archive/proposed-migration/wily/update_excuses.html#unity8 is not having boottest PASS.16:07
=== nuclearbob is now known as nuclearbob_lunch
Saviqpsivaa, "now" you meant? thanks, anything we should do or is this going to be better now?16:15
Saviqnow, in absence of pitti, who can help with autopkgtests being stuck in "Test in progress" for the whole day? :/16:16
psivaaSaviq: yes, 'now'. Sorry finger memory.16:17
fgintherSaviq, I think psivaa meant "now" (as it does show a PASS)16:17
psivaaSaviq: this unity8 failure is tricky, as we're testing with an older image (191). With the latest one the provisioning is not very complete16:18
Saviqfginther, "75 upgraded, 127 newly installed, 6 to remove and 412 not upgraded.16:19
SaviqNeed to get 150 MB/168 MB of archives.16:19
SaviqAfter this operation, 577 MB of additional disk space will be used.", there's less than .5GB free on wily16:19
Saviqfginther, I wonder why that list is so big...16:19
Saviqlike g++, really?16:20
Saviqlooks like unity-scope-tool has excessive deps16:20
Saviqlet's see if we can drop that, and/or fix it16:21
sil2100seb128: hey! Did you get any news regarding the libusermetrics regression?16:21
greyback_trainguards: any reason silo27 hasn't been landed? QA was granted.16:22
sil2100greyback_: will land in a minute if possible, just still working on something16:24
greyback_sil2100: no biggie, just curious16:24
seb128sil2100, there is a fix approved on https://code.launchpad.net/~mterry/unity8/handle-empty-infographic-label/+merge/27147916:35
sil2100seb128: would it be able to land today?16:35
seb128sil2100, don't ask me, I'm not maintaining unity816:40
seb128Saviq & co probably know?16:40
sil2100seb128: ok, thanks for the info ;)16:40
sil2100Saviq: hey!16:40
seb128yw16:41
Saviqsil2100, here, whatup?16:41
sil2100Saviq: any ETA on mterry's branch ^ to land? :)16:41
sil2100The one seb128 pasted above16:42
Saviqsil2100, we've an approved branch for that16:42
Saviqsil2100, we'll land soon16:42
sil2100Saviq: you know if that's possible today still?16:43
Saviqsil2100, no, we're stuck in proposed since this morning16:43
Saviqbecause unity-scope-click autopkgtests are endlessly In Progress16:43
sil2100Ok, so I'll soft-revert the libusermetrics landing in the overlay and then re-release once unity8 is out16:43
sil2100Uuugh16:43
Saviqfginther, ok seems we can drop unity-scope-tool, which is the reason all the packages are pulled in, from the job conf, let me try a run without it16:47
Saviqwe need to fix that in any case16:47
Saviqor is there a reason why all the packages are explicitly installed?16:50
Saviqfginther, this is the stripped-down run, let's see how it goes http://s-jenkins.ubuntu-ci:8080/job/unity8-ci/6290/console16:51
=== pstolowski is now known as pstolowski|eod
=== alan_g is now known as alan_g|EOD
cjwatsonSaviq: pitti was saying earlier today that it's just processing a long queue17:02
Saviqcjwatson, anywhere we can see that?17:03
cjwatsondunno sorry17:03
Saviqtx anyway, let's hope it sorts itself out then17:06
sil2100kenvandine: could you publish the two silos that are ready for release? Those are dual silos so with th e current train bug I can't publish them...17:07
sil2100kenvandine: https://requests.ci-train.ubuntu.com/#/trainguards17:07
kenvandinesil2100, sure17:08
sil2100kenvandine: thanks :)17:08
fgintherSaviq, AIRC, those were needed to make sure all of the latest test dependencies were installed. The list could easily be stale now17:25
pmcgowansilo 55 is being tested now or is it blocked still?17:36
pmcgowansilo 55 is being tested now or is it blocked still?18:00
pmcgowandavmor2, ^^18:00
davmor2pmcgowan: I will have a look at it shortly18:01
pmcgowandavmor2, cool thanks18:02
=== nuclearbob_lunch is now known as nuclearbob
pmcgowanpopey, so is there a new music app to go with that mpris silo?18:10
popeypmcgowan: there is a merge that's being worked on18:17
popeypmcgowan: issues list -> https://docs.google.com/document/d/1xzI4D6knVJ6VmXxqJ2gyGSMowbtVVoNPgQl1fywim_U/edit18:18
pmcgowanpopey, those seem to be all over the stack, so wondering how silo 55 can be ready for testing18:19
popeyindeed.18:19
popeythis doc was collaboration between jim and ahayzen18:19
popeyahayzen: that's right, right?18:20
davmor2pmcgowan, popey: ouch, I wonder why Jim was pushing for the silo then.  Aiui this couldn't land till the ui fix was in place, that has been added to the silo this afternoon so is due a retest and this unblocked everything else then, I was not made aware of a previous issues list18:23
pmcgowandavmor2, yeah the indicator with fixes is now in the silo but does not sound readt18:23
pmcgowanready18:23
davmor2pmcgowan: also the bulk of stuff you can't actually test due to the fact that nothing uses the playlist feature yet :(18:25
pmcgowandavmor2, should be a way to at least test the indicator and see no regressions, plus the music app18:26
pmcgowanthen we can land inline playback stuff after18:27
davmor2pmcgowan: yes so the indicator works for play pause but skip back and forward just end the track/video because there is only one item in the playlist18:28
pmcgowandavmor2, seems we wait for abeato to give the all clear18:28
ahayzen_popey, yup that doc is a list tracking issues that me and jim know of between music-app/media-hub/indicator-sound19:21
ahayzen_popey, silo 55 adds MPRIS support in media-hub... and had some other improves bundled19:22
=== ahayzen_ is now known as ahayzen
popeypmcgowan: ^19:35
pmcgowanpopey, ahayzen right but not clear to me when they get addressed and we can land everything19:39
ahayzeni think Jim has been trying to iteratively land things and keep them all contained in separate silos19:40
ahayzenfrom our point of view the music-app is pretty close, most of the issues are with media-hub itself19:40
ahayzenor indicator-sound or whatever19:41
pmcgowanahayzen, we have al the latest in silo 55, but it cannot land if it breaks music=app19:41
ahayzenhow does it break the music-app?19:42
pmcgowandavmor2, ?19:42
ahayzenremember the current music-app in the store isn't using the bg playlists, so assuming the indicator-sound doesn't show the MPRIS controls when running with that, i don't think anything in the existing app would break?19:43
pmcgowanahayzen, right but the goal is to enable them in the indicator19:44
pmcgowanso we can do inline playback in scopes19:44
ahayzencan it not detect when you are using playlists?19:44
ahayzenand i assume the scopes will probably have the same issues we are having in our WIP branch19:45
ahayzenIIRC 047 has just the MPRIS stuff in it as well https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-04719:48
fgintherSaviq, your change to get unity8 on wily running worked, at least it was able to install and run the tests19:55
fgintherSaviq, Also, for dual landing of unity-api, should we drop CI on lp:unity-api/trunk-15.04 ?19:57
Saviqfginther, yes, and yes20:26
Saviqfginther, we want dual-landing for lp:qtmir as well, so ci for it on both wily and vivid please20:27
Saviqfginther, can you please drop unity-scope-tool and indicator-client from unity8's test_packages param?20:29
fgintherSaviq, yes as well. I don't see any other branches for qtmir except lp:qtmir, so nothing to remove there (which is why I only asked about unity-api)20:42
fgintherSaviq, will also have the unity8 job updated shortly20:43
TrevinhoWe're still out of free silos?21:07
Saviqfginther, nothing to remove, but we'd also like both wily and vivid testing on them21:09
fgintherSaviq, I'm setting up qtmir now. See http://s-jenkins.ubuntu-ci:8080/job/qtmir-ci/426/console for the results of testing with lp:qtmir21:10
Saviqfginther, great, thanks a lot21:10

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!