[00:26] cihelp, jenkins isn't building music-app [00:26] https://code.launchpad.net/~vthompson/music-app/fixes-1259962/+merge/198601 [05:05] cwayne: sorry, didn't get a good chance to reconfigure the jobs before, but I merged the change already. I just pushed and merged it all now, and re-started the custom jobs here: http://q-jenkins:8080/view/Touch/view/Ubuntu%20Touch%20Master%20Jobs/job/trusty-touch_custom-mako-smoke-master/ [05:05] cwayne: we should check on them tomorrow morning, but they will all run now anytime there's a new custom image [07:57] Morning [09:13] Mirv: hello! Did you notice any infra failures today maybe? [09:14] sil2100: intel failing right now there running apps stack ap:s [09:14] nothing else [09:15] I got one good result before that === ev changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cihelp (EU CI members sprinting this week) | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - [09:33] ev: joining hangout? [09:43] psivaa: hello! [09:43] psivaa: heya [09:44] psivaa: could you retry terminal and unity8 on maguro after recording hte URLs of failing tests? [09:44] lool: will do [09:44] psivaa: also would need webbrowser on mako [09:44] lool: will do [09:45] psivaa: cool, could you ping us when done? [09:45] psivaa: would like to kick an image build once it's done [09:45] lool: sure [09:45] psivaa: seems we duplicate here :) [09:45] err in -touch [09:46] :) [09:46] so yeah. let us know when done so we can swiftly kick [09:50] lool: sorry, couldn't make it. Running a sprint on juju and ci in bluefin today [09:52] ok [09:53] sil2100: building it now first. I think we had the rule for this week that cu2d desktop tests need to pass as well before releasing anything, so let's see them first and I'll continue on my Qt stuff [09:56] Mirv: thanks! [10:03] Mirv: regarding testing the bug, from the merge request I see that it's indeed just switching between 12h and 24h locale, looking at some PM time and checking if the indicator changes [10:19] lool: asac: the three tests have been run and passed. sending you the earlier failed links in a mail [10:20] psivaa: \o/ [10:21] awesome [10:21] sil2100: good to start a build? [10:22] lool: fire it up! [10:22] 99.5% [10:22] that's pretty good [10:23] building [10:23] sil2100: Do we want to promote image 70? [10:24] sil2100: maybe check with popey and others [10:24] sil2100: just let me know here if you want it promoted [10:24] I can check on mako, but you need someone (omer) with maguro to test too [10:25] hmm [10:25] lool: how can we get those final tests fixed? [10:25] om26er: ^^ [10:25] om26er: we have like 3 AP tests... can you just crush them? :) [10:26] popey: if you have a free moment then check, if you can - I would personally prefer promoting an image with the datetime fix, but any promotion doesn't hurt ;) [10:26] om26er: http://ci.ubuntu.com/smokeng/trusty/touch/mako/70:20131217:20131211.2/5503/ubuntu-weather-app-autopilot/ [10:26] om26er: if you can fix those i will give you whatever you want :) [10:26] 2!!! [10:34] asac, sure. get me a pay raise ;) [10:35] om26er: how much? :) [10:35] what is needed for a new ubuntu-ui-toolkit release? It has been in the landing asks page for a while now [10:36] asac, i'll settle for 10% but before april. [10:36] working on the failures now [10:37] om26er: seems its just a DB not being created [10:37] not sure whats going on with the .ini files [10:37] and who should create those [10:38] in create_blank_db [10:38] something is missing [10:38] :) [10:38] feels all it wants is to figure what location to use for the sqlite file [10:38] not sure why the algo is so weird and goes through .ini files in Databases directory === xnox_ is now known as xnox [10:50] sil2100: 70 looks good to me on mako [10:51] popey: thanks! [10:52] om26er: could you give image 70 a spin on maguro as well, just like popey ? [10:52] sil2100, Ok, updating. maguro is too slow to be fun :) [10:52] ;D [10:56] sil2100, how to run autopilot tests for click packages ? [10:56] om26er: http://paste.ubuntu.com/6588007/ i use that script [10:57] ./testing.sh lp:ubuntu-rssreader-app ubuntu_rssreader_app [10:57] like that [10:58] popey, thanks [10:58] om26er: https://wiki.ubuntu.com/Touch/Testing <- this guide works for me as well [10:58] So just phablet-config autopilot --dbus-probe enable [10:58] And then: phablet-click-test-setup --click com.ubuntu.calculator; phablet-test-run ubuntu_calculator_app [10:58] etc. [10:59] Damn, my neighbour again started drilling [10:59] Doesn't help with concentrating [11:07] bug 1192711 is a little irritating [11:07] bug 1192711 in mediaplayer-app "video player won't play a second video without closing completely" [High,Confirmed] https://launchpad.net/bugs/1192711 [11:07] not a regression, seems to have been there a while, but I'd not noticed [11:21] sil2100: https://code.launchpad.net/~timo-jyrinki/cupstream2distro-config/indicators_missing_deps/+merge/199265 [11:22] Mirv: do you know if both libupstart-app-launch1 and libupstart-app-launch2 are required to be installed right now? [11:24] om26er: any clue yet? [11:24] om26er: sergiusens knows how to run AP test for click [11:24] oh already answerd :) [11:27] Can someone help me figure out why most of the autopilot tests for ubuntu-system-settings are borked (the ones which try to select anything, it seems), please? [11:27] asac, so the issue is the test suite expects a certain directory to be there but its not able to find it because that directory is only created once the weather app is started [11:28] om26er: i got so far... but it also expects .ini files in that directory :) [11:29] so i dont know where that .ini is supposed to come from [11:29] i think this is super bogus code to be honest :) [11:29] asac, works fine for me after I started the weather app manually [11:29] om26er: where is that directory created? [11:29] borked on the device only, that is --- they work on the desktop... [11:29] om26er: i didnt see anythign in code creating that directory ... nor the .ini file [11:29] sil2100: in the previous cases it seemed to be, at least earlier, that if the older one was removed it did not work [11:29] asac, ~/.local/share/com.ubuntu.weather/ [11:30] om26er: where in code i mean :) [11:30] oh wait [11:30] yeah .../Databases [11:30] i still dont see anywhjere in code that this dir is created [11:30] also the code checks for .ini files [11:30] asac, exactly that directory is not created by the test suite, rather it is automatically created when the app is started [11:30] anyway. seems you already have it running, so debugging shoudl be simple [11:31] om26er: which code is doing that? [11:31] om26er: is that something our sdk does? does it also create the .ini files? [11:31] asac, the app is using QtLocalStorage for that [11:31] asac, the dir is created when using LocalStorage [11:31] ok seems the test doesnt use it [11:31] kk [11:32] so the tests should be using that :) [11:32] sil2100: anyhow, the cu2d tests are not green so if I should test on device or not [11:32] instead of raw mkdirs etc. [11:32] err raw os. stuff [11:32] interesting that i get no direct hit when googling for QtLocalStorage [11:32] is that no public API? [11:33] sergiusens, I am thinking to check if that directory exists and if it does not just start the app and close it in the suite. sounds hacky ? [11:33] sil2100: I understand that we won't release unity stack until it's green, but what about indicators? do you think I can go ahead and test on device with those results currently in cu2d and publish if device ok? [11:33] sounds hacky :) [11:33] but makdirs that directory would make sense [11:33] :) [11:33] om26er, that was there before and actually removed from the test suite to make it more stable [11:33] hehe... just dont know about the .ini and if that info in ther ematters [11:33] om26er, I don' t get it; I tested by deleting the dir completely when first accepting [11:34] so why cant the test use qt local storage as well? [11:34] or call an app func to do the setup [11:34] hmm. guess its not nicely available in python :) [11:34] asac, QtQuick.LocalStorage [11:35] right. so if we would have qmltests rather than AP things would be easier? [11:35] anyway. i feel fro the test all that matters is that we give it SOME db filepath [11:35] that is valid [11:35] but qmltests are not integration tests [11:36] so whatever we give too it its good and if its even in the right .local/ directory owned by the app its even better [11:36] asac, the app is supposed to create everything it needs on it's own, and it does [11:37] sergiusens: right, so its a race? [11:37] sergiusens: where does the test start the app? [11:37] om26er, asac, the test code just needs to contemplate the inexistence of a prior database, unless it mangles with it before launch [11:37] sergiusens: so we might want to change the find_db into a poll that waits for 60 seconds? [11:37] and only then gives up? [11:37] :) [11:37] asac, that may be the case; not sure if it waits for everything [11:38] asac, yeah, that's what's been happening with the tests [11:38] Mirv: let's test it on the device, since I guess not-green tests on indicators right now basically might be related to some desktop issues [11:38] Mirv: let's make this one exception [11:42] btw ci, need a release for autopilot to get some emulator love http://www.youtube.com/watch?v=n2pVpjBFiaI [11:42] cihelp that is [11:43] nice one [11:43] sergiusens: lets get one 100% image [11:43] then we land autopilot... hwoever, i wouldprefer if we could pick just what you need [11:43] and leave the rest out [11:43] There's a lot of autopilot changes pending still [11:43] check much stuff people have accumulated http://people.canonical.com/~platform/cu2d/results [11:43] i would prefer if we only land exactly what we understand out of those [11:43] I guess we could release this, but I would like to get all the other pending bits out separately first [11:44] at best all, but only after looking at each change [11:44] :) [11:44] sil2100: lets get 100% green before pulling in AP [11:44] AP folks should be testing the hell out of their AP [11:44] so once we take it its clean [11:45] asac, sil2100 would a package patch do? It's a one line add to a map, that's about it for now [11:45] robru: ping [11:45] I can prepare a debdiff [11:45] sergiusens: not before we have 100% :) [11:45] get us the fix for this test :) [11:46] then we cna land whatever you want [11:46] hehe [11:46] asac, ack, I can wait [11:46] sergiusens: help fixing the test [11:46] its just this DB thing [11:46] at best next image would have the fix [11:46] sergiusens: that one-liner, would it help in getting us green, or is it related to something else? [11:46] sergiusens: check with om26er if he needs help etc. [11:46] sil2100, not related [11:46] om26er, do you need help? [11:47] om26er, I can review whaever you've done so far [11:47] sergiusens: then let's wait and release it with all the AP changes then - we need to release those ASAP anyway ;) [11:47] sil2100: please send a mail to thomi,sergiusenss etc. that they should start running APs all day long against their AP package from daily-build [11:47] and give us a list of APs that are green [11:47] after that ... at best with logs :) [11:49] sil2100: alright [11:52] sergiusens, I only figured what was the problem. now working on workarounds/solutions [11:52] sergiusens, will get help from you if i run out of ideas [11:54] ack [11:54] Mirv, sil2100 do you know what is blocking the UITK release? the flaky test from https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0Au6idq7TkpUUdGNWb0tTVmJLVzFZd0doV3dVOGpWemc&usp=drive_web#gid=0 was fixed a while ago [11:55] it was this one https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1259476 [11:55] Ubuntu bug 1259476 in ubuntu-ui-toolkit (Ubuntu) "Flaky gallery.test_textinput.TextInputTests.test_textfield_numbers, fix or disable" [Undecided,New] [11:55] timp: hi! [11:55] sil2100: hello :) [11:55] timp: from what I see we required an ACK from you guys and a re-addition to the Landing Asks once the flacky test is gone :) [11:56] timp: so, if you can, please add it and we'll do everything we can to release it ASAP [11:56] sil2100: I don't have edit rights for that doc [11:56] sil2100: can I get rights or should I ask zoltan to add it tot he list? [11:56] *to the list [11:57] om26er: did I see sil2100 bribe you to look at the weather test failures? Are you looking at those? [11:58] popey, yes, I am working on it [11:58] timp: it's usually the manager that adds it [11:58] om26er: great. let me know if you need any testing of it done [11:58] Mirv: okay. [11:58] timp: if you can, please ask Zoltan :) Thanks! [12:00] okay, asking. thanks [12:00] robru: unping :) [12:03] sil2100: it is there nwo :) [12:03] *now [12:03] I seem to have a typing problem today [12:15] sergiusens, summary is: the test code adds certain fixed locations to the database (hamburg and London in this case) before starting the app so that the tests have some data in the app to interact with. for injecting these pre-defined db entries it needs to have an already created database. [12:15] sergiusens, the notes-app does exactly the same thing. it check if the db directory is there and if its not it starts and quits the app [12:18] om26er, ok, there's old code in there that calls a launch_and_close_app === gatox is now known as gatox_dr [12:40] sil2100: can you please keep me informed about the uitk release? If there are any issues please let me know so we can fix them :) thanks [12:40] timp: sure, thanks! :) We'll add it to our todays release list [12:40] great [12:41] Mirv: how's the progress on indicator-datetime ? [12:41] popey: hmm, you said you have some bluetooth audio device for Ubuntu Touch testing? [12:41] sil2100: it's starting to be done, I'm wondering though who to bug with packaging changes [12:42] lool: did we ever build an image :)? [12:43] ah seems we have 71 [12:43] nice [12:43] psivaa: something was going wrong on mako it seems [12:43] guess you are on it? [12:44] Mirv: that's a valid question! Maybe try poking Laney or lool I guess? [12:44] Later we can always pester Ken or Mathieu [12:45] ack. pick any friendly core-dev :) [12:46] There's loads in #ubuntu-devel, but sure [12:46] Laney: http://pastebin.ubuntu.com/6588664/ (to fix http://pad.lv/1256061) [12:46] Launchpad bug 1256061 in indicator-datetime (Ubuntu) "clock forcefully switches from 24h to 12h AM/PM format once the panel clock is loaded" [Undecided,New] [12:46] maybe you could look up my question from earlier that went unanswered to thank me back :-) [12:47] oh it's this [12:47] the merge request does not explain the details (small fixes) https://code.launchpad.net/~charlesk/indicator-datetime/lp-1256061/+merge/198968 but seem sok [12:47] I told him it's better to use locale-gen directly [12:47] but fine [12:47] sil2100: no [12:48] sil2100: oh, actually, my daughter has a speaker, let me see if that works [12:48] popey: since someone mentioned on the ML that bluetooth devices still don't work with the latest image [12:48] Laney: thanks. ooh, autopilot test failures.. did you see the logs already? [12:48] which logs? [12:49] Laney: settings failures, there are some tests failing in the CI [12:49] show me? [12:49] it's probably the same thing [12:49] http://q-jenkins.ubuntu-ci:8080/job/autopilot-trusty-daily_release/957/label=autopilot-nvidia/ + http://q-jenkins.ubuntu-ci:8080/job/autopilot-trusty-daily_release/957/label=qa-intel-4000/ [12:49] for some reason on intel test_about.StorageTestCase.test_installed_apps and on nvidia test_about.LicenseTestCase.test_licenses_page [12:50] /var/local/autopilot/autopilot.log: AssertionError: The following apps were started during the test and not closed: [] [12:50] is it that? [12:50] sil2100: it doesn't detect my bluetooth speaker [12:50] but I don't know if it ever did, because I have never tested this [12:50] CraB [12:50] Laney: looks like that. a) it seems like apport, b) it's different from your "most tests failing" [12:51] it is [12:51] yep, http://q-jenkins.ubuntu-ci:8080/job/autopilot-trusty-daily_release/957/label=autopilot-nvidia/artifact/results/autopilot/videos/ubuntu_system_settings.tests.test_about.LicenseTestCase.test_licenses_page.ogv [12:51] system-image-dbus crash [12:52] psivaa: hello! Do you have a Ubuntu Touch compatible bluetooth device by any chance? [12:52] seems like it's not our fault [12:52] also that runs on desktop ... [12:52] the problem I had was only on the phone [12:53] Laney: bypassing those, are the selection problems using selection via UI Toolkit somehow? just wondering if SDK people could help debugging the errors [12:53] it uses select_single from autopilot, don't know what that does [12:54] that's at least used also by ubuntu-ui-toolkit which has all tests passing, so maybe it'd be worth asking AP people like thomi directly [12:56] ok === alan_g is now known as alan_g|lunch [12:59] popey, here: https://code.launchpad.net/~om26er/ubuntu-weather-app/fix_db_tests/+merge/199278 [12:59] popey, will the CI run on that branch ? === cjohnston changed the topic of #ubuntu-ci-eng to: Ubuntu CI Engineering Team | Vanguard: cjohnston | Landing instructions: http://paste.ubuntu.com/6292280/ | Known issues: - [13:06] om26er: lemme test it first [13:07] it wont if nobody approves it [13:09] testing now [13:17] sil2100, Mirv: Feel free to bug me for packaging reviews if I'm around [13:18] lool: thanks :) [13:20] om26er: sergiusens do we have a breakthrough on those tests yet :)? [13:20] asac, I gave something to popey to test [13:20] WFM [13:20] wfm too [13:20] cool [13:20] :) [13:20] nice [13:20] * popey commented on the mr [13:20] who need to approve so this can land? [13:21] did something else land as well on trunk after we updated in image last time? [13:21] lets get that removed :) [13:21] mhall119: can probably approve https://code.launchpad.net/~om26er/ubuntu-weather-app/fix_db_tests/+merge/199278 for us [13:21] when he awakens shortly. [13:23] popey, om26er: with this fix in, the weather app AP tests are fixed? [13:23] thats the plan [13:23] sil2100, atleast the one's that I was given link to [13:23] om26er: +beer for you! [13:25] sil2100, ;) [13:26] asac, nothing; the apps with the most failures in testing have been apps that only hd test updates for the past two months [13:32] sil2100: i dont have any devices with me at the moment :/ [13:33] om26er: sorry to bother you again, but maybe you by some lucky chance have some UT compatible bluetooth devices? ;) [13:33] sil2100, I don't have. the one in my car is not supported by Ubuntu touch [13:35] sergiusens: hi! I had some questions related to LP: #1234361 [13:35] Launchpad bug 1234361 in lxc-android-config (Ubuntu) "Bluetooth panel not getting scan results" [High,Fix released] https://launchpad.net/bugs/1234361 [13:35] sergiusens: someone reported on the ML that it's not fixed - and I can't find anyone that would have the equipment for re-testing this [13:37] sergiusens: do you have the tools to check if this is fixed with the latest image? [13:38] sergiusens: ok... please help the fix get into the store and then lets build an image [13:38] psivaa: mako still has no resutls for 71 [13:39] http://ci.ubuntu.com/smokeng/trusty/touch/maguro/71:20131217.1:20131211.2/5521/ [13:39] (sorry if you answered) [13:39] plars: ^^ [13:39] asac: looking at that, sorry dint notice that [13:40] sil2100: did we promote 70 or do we have that in the bank and already well tested, but hope for 71 to be at least as good and push that? [13:42] sil2100, I don't have bluetooth headsets, but I can check from the cli and see if I find anything [13:46] asac: from what I see, 71 doesn't have anything more than 70 (no packages updated) [13:47] asac: so let's promote 70 anyway, since 72 will only be ready for tomorrow morning [13:47] asac: if there is of course no cost of promoting an image that is [13:47] It doesn't matter if we promote 70 or 71 basically [13:48] But just to have a clear concious we can promote 70, since that's what popey and om26er used for testing [13:48] lool: ^ could you promote 70 ? [13:49] sil2100: yup [13:49] Thanks! [13:50] asac: just got to my desk, looking now, unless psivaa has already [13:50] plars: the devices is shown as offline soon after flashing [13:51] probably need to assign another device. mako-06 looks ok to do that [13:52] psivaa: wow, it even shows in adb but shows offline [13:53] psivaa: yeah, you want me to switch it, or do you want to? [13:53] plars: i am in the sprint,so if you could. that would help [13:53] psivaa: ack [13:53] thanks plars [13:56] rfowler: when you are around, could you take a look at mako-02? It's marked offline for some reason, so nothing can run on it. I've moved all the jobs from it to 06 for now [13:57] asac: jobs are restarting now for mako on image 71 [13:59] om26er, popey where's the MR [13:59] ? [13:59] https://code.launchpad.net/~om26er/ubuntu-weather-app/fix_db_tests/+merge/199278 [14:00] morning [14:01] FYI, image promoted [14:01] (70) [14:01] lool: my phone can't see it [14:01] pfft ... gimme 72 ... with 24h clock [14:02] popey, om26er ironically, the tests fail on jenkins [14:02] sergiusens, that's not related to my code [14:02] falky tests are flaky :/ [14:04] ogra_: patience! [14:04] :D [14:05] popey: something is wrong with the push mirrors [14:07] indeed [14:08] hmm maybe it's a cron and I had never realized [14:09] it's usually immediately after ogra_ says it's published [14:11] lool, nusakan seems dead here [14:11] or at least inaccessible [14:11] * ogra_ cant ssh [14:11] popey: it's being fixed [14:11] ogra_: you need to use sshebang [14:11] ogra_: or update your ssh config to use FQDN [14:12] ogra_: works fine for me [14:12] since when ? [14:12] I don't use sshebang) [14:12] ogra_: since some hours ago [14:12] it worked on monday [14:12] it worked wihtout .canonical.com up til this morning :-) [14:12] ah [14:12] my config has used nusakan.canonical.com for ages [14:12] I raised it with #is, but it's intended [14:12] * ogra_ fixes /etc/hosts [14:13] hmm, nothing to fix there it seems [14:13] 91.189.89.127 nusakan nusakan.canonical.com [14:13] but indeed, using the FQDN works [14:13] cjwatson: ssh nusakan used to work without .canonical.com from the jump host, allowing to have a single .ssh/config entry for *.canonical.com + nusakan; now I have two [14:14] ogra_: this is in your .ssh/config [14:14] I doubt that /etc/hosts matters. Fix ~/.ssh/config [14:14] ogra_: your proxycommand needs a FQDN [14:14] ogra_: nc -q0 nusakan.canonical.com or nc -q0 %h.canonical.com [14:14] The host name is looked up on the proxy jumphost, not on your machine [14:14] And I'm pretty sure you can't write to chinstrap's /etc/hosts :) [14:14] oh, ok [14:15] thanks, that fixes it === alan_g|lunch is now known as alan_g [14:21] #70 on the way \o/ [14:22] \o\ [14:22] It's a bird, it's a plane! No, it's #70! [14:23] IS should have fixed it entirely now [14:23] I see #70 on trusty [14:26] om26er, check my comment on the MR [14:27] plars: I'm going to 1ss in a bit... my car was completely frozen and had to thaw it before the next storm hits [14:29] rfowler: ugh, be safe [14:30] rfowler: it's no rush on our end, its up and running for now === gatox_dr is now known as gatox [14:46] sergiusens, pushed. looks better now ? [14:47] sergiusens: did you have a moment to do that cli bluetooth sweep? [14:47] om26er, let me check; let's also get balloons on now that he's online [14:47] sil2100, no [14:48] om26er, looks better, let's wait for ci [14:48] om26er, are you looking @ weather or calendar? [14:48] balloons, weather [14:48] balloons, weather, and its cold. [14:48] balloons, https://code.launchpad.net/~om26er/ubuntu-weather-app/fix_db_tests/+merge/199278 [14:48] sergiusens: om26er: are we close :)? [14:48] om26er, it's 35 C [14:48] the next image will be cut in 2h approx [14:48] would be nice to have it in :) [14:49] asac, I can't approve MRs for community apps, but if it works, I'm fine with the solution [14:49] balloons: can you plesae check that out? [14:50] balloons: we have the fix pending for 100% dashboard gren on mako [14:50] and i want to get this in opportunistically for next image [14:50] om26er, why are you readding launch and quit? [14:51] balloons, these tests assume ~/.local/share/com.ubuntu.weather/Databases/ to be there. It is only created once the app is started [14:52] balloons, since we manipulate an already existent database, and if the suite is not able to find one it fails without telling why it failed [14:53] om26er, we have the ability to create a db if one doesn't exist in the test [14:53] I don't believe this changes anything [14:53] launch and quit would be a regression imho [14:54] balloons, remove ~/.local/share/com.ubuntu.weather/Databases/ on your phone and run those tests, they fail with the traceback that's in the dashboard [14:54] balloons, then start the app with hands and run those tests again and they will pass [14:55] balloons, from what I saw from the test code is that it tries to add a few database entries to an already existent database, it does not create a whole new database [14:55] balloons, line 165 in __init__.py [14:56] om26er, I don't doubt your assessment. However, the fix imho is not to go back to the launch and quit thing. Let's just create a db from scratch if we need to [14:59] balloons, not really sure if that's going to be a simple task ? these are Qt LocalStorage databases, and there aren't nice APIs for python [15:00] om26er, well I understand the desire to push this in, but it's going to be tough to review beyond I would really like to not see launch and quite come back into this test [15:02] balloons, another nasty thought is: clear the db, launch the app, copy the db and ship that with weather app tests. In the test suite make sure to move any old db, copy the shipped one over to the default location, run the tests. Profit! [15:04] om26er, I would be happier with that kind of approach [15:04] if we can't build it on the fly, just ship it with the tests and copy it it to set it up that way [15:05] balloons, ok, I can work on that. btw how do these tests get in in the first place ? [15:07]