Log of the #duraspace-ff channel on chat.freenode.net

Using timezone: Eastern Standard Time
* nbanks joins00:08
* nbanks leaves00:13
* eddies1 joins00:14
* eddies1 leaves
* eddies1 joins
* eddies leaves00:16
* nbanks joins01:07
<bljenkins>Project fcrepo-kitchen-sink build #136: FAILURE in 6 hr 57 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/136/01:15
Project fcrepo-fixity build #130: FAILURE in 6 hr 56 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/130/
* kaarefc joins01:53
* kaarefc leaves02:06
* kaarefc joins03:41
* kaarefc leaves04:00
* kaarefc joins04:02
* eddies1 leaves04:58
* eddies joins05:29
* eddies leaves
* eddies joins
* eddies leaves06:18
* eddies joins06:32
* eddies leaves
* eddies joins
* ap2972 joins07:02
* ap2972 leaves07:19
* eddies leaves08:16
* eddies joins08:21
* eddies leaves
* eddies joins
* ap2972 joins08:24
* ap2972 leaves08:50
* ap2972 joins08:54
* ap2972 leaves
* ap2972 joins08:55
* VincentNG joins09:05
* ajs6f joins09:06
* github-ff joins09:11
[fcrepo4] nigelgbanks opened pull request #45: Ignore .DS_Store files and ignore emacs backup files *~ (master...ignore-osx-files) http://git.io/X5ubVA
* github-ff leaves
* kaarefc leaves09:19
* escowles joins09:33
* github-ff joins09:41
[fcrepo4] ajs6f pushed 2 new commits to master: http://git.io/D-nO7A
fcrepo4/master 4455d09 Nigel Banks: Ignore .DS_Store files and ignore emacs backup files *~
fcrepo4/master 055e6ad ajs6f: Merge pull request #45 from nigelgbanks/ignore-osx-files...
* github-ff leaves
* fasseg joins09:53
eddies: the problem is that failsafe tries to start a jetty/modeshape instance for each submodule with integration tests. But since modeshape is not completely shut down before a new instance is spun up by the next module's integration test it fails...any tips?09:54
so cxf migration is complete, but i have this problem with tests...
i could move all ITs to one submodule like in fedora3....
<eddies>please let's not do that09:57
so you've followed: http://maven.apache.org/plugins-archives/maven-failsafe-plugin-2.11/usage.html
i.e., bound jetty to pre & post-integration lifecycle?09:58
<ajs6f>No i-test modules for more than one module!
If we can't i-test a module from inside it, we need to refactor at a higher level.
<fasseg>yes but jetty is not the problem
jetty is down before modeshape
so modeshape contents for some resource and fails to start the second instance
<ajs6f>Is it Spring that is keeping MODE alive?
* travis-ci joins09:59
[travis-ci] futures/fcrepo4#331 (master - 055e6ad : ajs6f): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/de7df59ef906...055e6add957d
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6050373
* travis-ci leaves
<fasseg>i thik it just takes some time for housekeeping
<ajs6f>Cause that doesn't happen for the main codebase, which has tons of itests in each module.
<bljenkins>Yippie, build fixed!
Project fcrepo-kitchen-sink build #137: FIXED in 4 min 41 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/137/
<ajs6f>Are you using the Spring test framework, or manually starting jetty?
<eddies>was just going to say, if you checkout modeshape, it has plenty of ITs that start/stop container10:00
although they're using the maven cargo plugin rather than the jetty plugin
<ajs6f>Are they?
No they aren't. I just looked. They're doing it by Spring starting CXF which starts Jetty.10:01
<fasseg>pom example for legacy api is here: https://github.com/futures/fcrepo4/blob/resteasy/fcrepo-legacy-api/pom.xml
<ajs6f>Or at least some are.
<eddies>ah, well there are at least *some* ITs that use cargo
<ajs6f>Like the generator modules.
(They use Spring -> CXF -> Jetty)
<bljenkins>Yippie, build fixed!
Project fcrepo-fixity build #131: FIXED in 4 min 14 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/131/
* kaarefc joins
<fasseg>ajs6f: i bound jetty-maven-plguinn to pr-integration phase10:02
for each submodule
<ajs6f>Oh, this is in your branch?
<ajs6f>Cause I was looking at master.
I see.
In master, we don't use cargo.
<eddies>ajs6f: you're referring to modeshape/modeshape right?10:03
<fasseg>neither does the reasteasy branch...or i the jetty plugin wrapping cargo?
<ajs6f>No, I'm talking about the master branch of our code.
The jjetty pluging does not wrap cargo. It's independent. You want to use one or the other.10:04
<eddies>*i* was referring to the use of cargo in modeshape/modeshape
<ajs6f>I always choose for cargo, because it offers to possibility of changing containers.
<eddies>we don't use cargo in fcrepo4
Or, not yet. :)
But frank is writing for fcrepo4, right?10:05
<eddies>yes, i sent an email about this quite some time ago
<ajs6f>frank, are you using spring with resteasy?
<eddies>yes. yes. yes. i just want to make sure folks are talking about the same thing :-P
<ajs6f>I thinkk we now are. :)
<fasseg>In what way? for dependency injection or for the servlet dispatcher?10:06
* kaarefc leaves
<ajs6f>Servlet dispatcher/resteasy
<eddies>back in 5 mins
* kaarefc joins
<fasseg>servlet is: org.jboss.resteasy.plugins.server.servlet.HttpServletDispatcher10:07
so yeah
with resteasy spring bootstrap
<ajs6f>Then I thikn you ought to be letting spring start and stop the container via the spring test framework.10:08
Otherwise the lifecycles aren't going to mesh as is happening now.
<fasseg>okay ill try that...thanks for the hint!
hmm i thought i ran the tests with SpringJunti4Runner10:09
that's what were talking about right?10:10
or is there another spring test framework im unaware off?
<ajs6f>That's what I mean. Are you trying to use package scanning?10:12
<fasseg>for the resteasy resources, yes
<ajs6f>That may be a problem.
* kaarefc leaves
"It's crucial that you do not configure resteasy.scan or resteasy.resources.  This will result in strange behavior; sometimes the beans created by Spring will be used by RESTEasy and sometimes RESTEasy will create its own.  So, don't configure your RESTEasy-annotated classes via RESTEasy's context params (I would assume implementing javax.ws.rs.core.Application to expose your classes would also cause these weird problems)"
<fasseg>hmm nah i never had this problem10:14
<ajs6f>That may be the problem you are seeing.
<fasseg>but now i remember that i removed the RunWith annotation for some reason, ill have to check tha
Otherwise you may get REST Resource instances(UserResource) from RestEasy instead of Spring container10:15
^^^ I dind not have this problem
<ajs6f>But you are having a problem where the liefcycles of different compoents/beans are not aligning.
Or rather, their containers.10:16
<fasseg>nah the container seems to shut down nicely (i.e. jetty) just modeshape needs a couple of more seconds to shutdown i think...
so ican restart jetty in between the integration tests, but modeshape will fail..10:17
<ajs6f>But we don't have that problem in the i-tests where the web container and the bean cotainer are controlled by the same lifecycle machinery.
You're controlling them seperately.
So MODE _can_ shutdown cleanly between tests.
<fasseg>and jetty waits for it before it shuts down?10:18
<fasseg>ah ok..
<ajs6f>It waits because the same lifecycle is controlling both.
<fasseg>so ill try again with the springrunner and come back to you..
<ajs6f>You've got independent lifecycles and they're not meshing, so they get out of sync.10:19
yeah, i'd try that.
* barmintor joins10:21
<fasseg>but doesnt modeshape get started by jetty so they are by default in the same spring context? well nevermind ill run the tests
<ajs6f>In the i-tests in the master branch (where this problem doesn't occur) then no, both MODE and Jetty get started by Spring. That's why they stay in sync.10:22
<fasseg>so jetty does not control modeshapes lifecycle....10:23
<ajs6f>No. Spring controls both.
Well, Spring -> CXF -> Jetty.10:24
and Spring -> MODE.
They both have the same root container.
<fasseg>i see
<ajs6f>Anytime you have two or more containers and they _don't_ have the same root container, you are liable to this problem.
Slipping gears.10:25
it's time :-P
fasseg, escowles, nbanks, vincentng: weekly call now10:32
<ajs6f>I have a pull request for my ticket.10:33
* escowles_ joins10:34
* github-ff joins10:48
[fcrepo4] nigelgbanks opened pull request #46: Updated the pom.xml to include Nigel Banks. (master...added-nigel-to-pom.xml) http://git.io/A8elrw
* github-ff leaves
* bljenkins leaves10:50
* bljenkins joins10:52
<barmintor>seems difficult to find a place w/ more than 4 BR on airbnb10:59
<jonathangee>hangout is full :(11:00
<pivotal-bot____>Edwin Shin edited "Create BagIt bag-info.txt -based ExtraPropertiesStore" https://www.pivotaltracker.com/story/show/4738331511:01
Edwin Shin edited "Create BagIt bag-info.txt -based ExtraPropertiesStore" https://www.pivotaltracker.com/story/show/47383315
Edwin Shin edited "Switch unit tests to integration tests where applicable" https://www.pivotaltracker.com/story/show/4653743711:02
Edwin Shin edited "Switch unit tests to integration tests where applicable" https://www.pivotaltracker.com/story/show/46537437
<barmintor>eddies: if we were 6 in the house: https://www.airbnb.com.sg/rooms/734748
<eddies>i saw that one. and they actually own a couple properties11:03
<cbeer>yeah, we're at 10, i guess11:04
eddies: can you ask Ed Fugikawa to leave?
* kaarefc joins11:07
<pivotal-bot____>Chris Beer added "Document new module creation procedure/archetype" https://www.pivotaltracker.com/story/show/4745077711:08
Edwin Shin accepted "Unit test fcrepo-kernel/LowLevelStorageService" https://www.pivotaltracker.com/story/show/4682666711:11
Edwin Shin accepted "Unit test fcrepo-kernel/RepositoryService" https://www.pivotaltracker.com/story/show/46826735
Edwin Shin accepted "Unit test the kernel's Function subclasses" https://www.pivotaltracker.com/story/show/4733739911:12
Edwin Shin edited "Create a REST endpoint for dynamically projecting files from a directory from a federated content as datastreams on a..." https://www.pivotaltracker.com/story/show/4737377711:23
* VincentNG leaves
* VincentNG joins11:24
lost connection :-(11:25
anyone have a link to the modeshape federation documentation handy? if not, i'll look it up11:26
<escowles_>VincentNG: i lost my connection a couple times -- anyone else having problems staying connected?
<VincentNG>thanks cbeer!!
* github-ff joins11:28
[fcrepo4] cbeer pushed 2 new commits to master: http://git.io/ghhHkw
fcrepo4/master 972bfac Nigel Banks: Updated the pom.xml to include Nigel Banks.
fcrepo4/master d72a792 Chris Beer: Merge pull request #46 from nigelgbanks/added-nigel-to-pom.xml...
* github-ff leaves
* escowles_ leaves11:33
* travis-ci joins11:36
[travis-ci] futures/fcrepo4#333 (master - d72a792 : Chris Beer): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/055e6add957d...d72a7928632b
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6053097
* travis-ci leaves
* github-ff joins11:37
[fcrepo-bagit-modeshape-federation-connector] ajs6f pushed 1 new commit to master: http://git.io/J4lBOA
fcrepo-bagit-modeshape-federation-connector/master 6fe9331 ajs6f: Added failing i-test to make pass
* github-ff leaves
<ajs6f>Going into the office. Be back in a few.11:38
* ajs6f leaves
<bljenkins>Project fcrepo-kitchen-sink build #138: SUCCESS in 3 min 10 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/138/11:46
Project fcrepo-fixity build #132: SUCCESS in 3 min 18 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/132/11:47
<pivotal-bot____>Chris Beer finished "Jenkins deploys ModeShape builds to Nexus" https://www.pivotaltracker.com/story/show/4737515311:50
Chris Beer deleted "Deploy something to the fcrepo nexus repository to prove that it works" https://www.pivotaltracker.com/story/show/47376369
<cbeer>ok, i've made a branch in futures/modeshape called fcrepo-snapshot11:54
which should be what gets deployed to our nexus repo11:55
<pivotal-bot____>Chris Beer deleted "merge MODE-1836 into our fork of modeshape" https://www.pivotaltracker.com/story/show/47375005
Chris Beer deleted "fork modeshape into futures so we can merge arbitrary pull requests" https://www.pivotaltracker.com/story/show/47374975
<nbanks>Hey Chris, could you create a user for me on the oxford vm's? I prefer the user name nbanks12:08
<eddies>cbeer: the nexus deploy works now?12:09
what changed?
<cbeer>eddies: i fiddled some settings. we told it to use the workspace's settings.xml, which I guess overrode the global settings.xml12:10
but there's another place in the build config to give it a global settings path
<eddies>ah ok12:11
<cbeer>now i'm rebuilding from the current modeshape/modeshape master
(in our fcrepo-snapshot branch)
and we'll see how that goes..
running tests this time too
* VincentNG leaves12:12
<eddies>barmintor: to follow up on that cambridge airbnb unit...
as i understand, that listing covers 2 separate apartment units, each w/ 2 bedrooms12:13
and each unit's living room also has a pull-out sofa
* VincentNG joins12:14
<eddies>but that owner also has another property entirely, which can sleep another 2 ppl
i can't tell how far it is exactly, but looks to be within a block
so i think it's going to be less-than-ideal for a large, collaborative workspace12:15
although the location (harvard sq) is great
<pivotal-bot____>Edwin Shin added "Update jmeter test plan to support fcrepo4" https://www.pivotaltracker.com/story/show/4745752312:17
Edwin Shin added "JMeter delete objects return spurious 404s" https://www.pivotaltracker.com/story/show/4745771312:18
Chris Beer finished "Unit test the fcrepo-kernel service proxies and node wrappers" https://www.pivotaltracker.com/story/show/4682629312:24
<cbeer>ok, got a recent version of the modeshape snapshot deployed12:26
eddies: if we stay around harvard, we can get space at 'gbh for working12:27
or the harvard i-lab, maybe
or tufts
or MIT12:30
<barmintor>eddies: I guess it depends on how many people are collaborating on a thing at once, too12:34
in chicago, we tended to be working in groups of 2-3 at a time12:35
<barmintor>we're the week of the 21st, right? Show up on Sunday?12:40
get in by sunday night
leave friday night/sat morn12:41
i'm not too crazy about those cambridge units on airbnb. i think we'd end up needing three units, at which point we might as well be considering hotel rooms12:43
and that's getting away from part of the point of the hackfest12:44
also. the salem house is *cheap*
and to some degree, being away from the distractions of boston/cambridge is a good thing12:45
but we could make a day trip out there for meeting w/ folks (unless they want to come up to salem)
50 mins by train to harvard sq12:46
* ajs6f joins13:15
* fasseg leaves13:20
* ajs6f leaves13:29
* ap2972 leaves13:33
* ajs6f joins
<cbeer>ajs6f/eddies: i think i need some maven help. i'm trying to make fcrepo4 use the 3.2 snapshot builds13:39
so i added a <repository> entry in the parent pom.xml
and changed the modeshape.version property to 3.2-SNAPSHOT
<eddies>cbeer: did you enableSnapshots true?
or i forget the exact xml element name13:40
<ajs6f>I think it's just "snapshots", but the Eclipse editor will tell you.
<cbeer>and.. i first need to remember how to get maven to stop using its cached stuff so i can get the error back
<ajs6f>-U often helps.
<eddies> <enabled>true</enabled>
<cbeer>-U, thanks
<ajs6f>That's minus three votes for U. Poor U.
<cbeer>and i've apparently fixed it myself
so, all is good
(there was a jenkins setting for appending some crap to the artifact names)
<eddies>so happy to be of service ;-)
cbeer: just emailed the karens about setting up a meeting while we're in town.13:42
i floated the idea at the partners meeting last week
so she's prepared for it
* github-ff joins13:46
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/JSmXOQ
fcrepo4/master 6327173 Chris Beer: build against our MODE 3.2-SNAPSHOTs.
* github-ff leaves
<cbeer>supposedly the clustering fix is in those snapshots13:47
so i'll spend today trying to remember how to set that up on the oxford vms
* ap2972 joins13:48
<bljenkins>Project fcrepo-kitchen-sink build #139: SUCCESS in 2 min 24 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/139/13:57
Project fcrepo-fixity build #133: SUCCESS in 2 min 42 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/133/13:58
* ajs6f leaves14:00
* ajs6f joins14:04
<barmintor>ajs6f: could you do me a favor and link me a gist with the maven settings for the snapshot repo?14:05
<cbeer>barmintor: which repo?14:06
the fcrepo snapshots?
<barmintor>the one w the modeshape snapshot
<cbeer>(sorry, the sonatype one, or the one with modeshape in it)
are you looking for that?
or the block in settings.xml?
(so you can push to it)14:07
<barmintor>Hmm… maybe I just need that. I thought it was settings.xml
* barmintor tries it out
<cbeer>settings.xml just has the username/password stuff for pushing to it
* travis-ci joins14:08
[travis-ci] futures/fcrepo4#334 (master - 6327173 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/d72a7928632b...6327173cc30b
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6056453
* travis-ci leaves
<barmintor>ok, I thnik I just need to reinstall the parent pom
<pivotal-bot____>Benjamin Armintor edited "Watcher for federated filesystem that triggers events" https://www.pivotaltracker.com/story/show/4738061914:23
Benjamin Armintor started "Watcher for federated filesystem that triggers events" https://www.pivotaltracker.com/story/show/4738061914:24
<barmintor>FYI: http://junit.sourceforge.net/javadoc/org/junit/Ignore.html14:37
<eddies>and notably: @Ignore takes an optional default parameter if you want to record why a test is being ignored, e.g. @Ignore("not ready yet")14:40
cause i hate going back trying to figure out why various tests are being ignored =)
<barmintor>I assume, fo a BagIt connector, that we're not watching the directory so much as we're watching the mnifest file14:43
since files not on the manifest aren't "in" the bag
<escowles>barmintor: we might also want to watch the bag-info.txt if we're watching for property updates too14:44
<barmintor>escowles: fair enough!
* ajs6f leaves15:10
* ajs6f joins15:12
* ajs6f leaves15:28
* ap2972 leaves15:31
<pivotal-bot____>Esme Cowles started "Create BagIt bag-info.txt -based ExtraPropertiesStore" https://www.pivotaltracker.com/story/show/4738331515:33
* github-ff joins15:42
[fcrepo4] escowles pushed 1 new commit to master: http://git.io/BPX9Eg
fcrepo4/master 1d08282 Esmé Cowles: Adding my name to pom.xml contributor list, adding bag-info.txt ExtraPropertyStore implementation
* github-ff leaves
* nbanks leaves15:45
<bljenkins>Project fcrepo-kitchen-sink build #140: SUCCESS in 3 min 25 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/140/15:52
* travis-ci joins16:02
[travis-ci] futures/fcrepo4#335 (master - 1d08282 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/6327173cc30b...1d0828227742
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6059616
* travis-ci leaves
* kaarefc leaves16:09
* ap2972 joins16:15
<bljenkins>Project fcrepo-fixity build #134: FAILURE in 28 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/134/16:23
Project fcrepo-kitchen-sink build #141: FAILURE in 29 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/141/
<barmintor>just saw the Northwestern news, that's awesome16:28
<cbeer>barmintor: what news?16:35
* github-ff joins
[fcrepo-kitchen-sink] cbeer pushed 1 new commit to master: http://git.io/o5bOrg
fcrepo-kitchen-sink/master d5d8373 Chris Beer: fix kitchen sink build
* github-ff leaves
<barmintor>cbeer: they're joining FF steering as a funding partner16:36
<cbeer>just $$ or developers too?
<barmintor>I only heard about $$ in this email
<barmintor>but maybe we can hijack mbklein for a while :)
<cbeer>that's what I was hoping :)16:38
* escowles leaves16:39
* ap2972 leaves16:42
<bljenkins>Yippie, build fixed!16:45
Project fcrepo-kitchen-sink build #142: FIXED in 3 min 29 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/142/
Project fcrepo-kitchen-sink build #143: SUCCESS in 3 min 51 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/143/16:49
Yippie, build fixed!16:51
Project fcrepo-fixity build #135: FIXED in 3 min 52 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/135/
* VincentNG leaves16:52
* nbanks joins17:05
* ajs6f joins17:09
* ajs6f leaves
<cbeer>replication works!17:29
(but there's still a massive delay..)
* nbanks leaves17:33
* barmintor leaves17:53
<cbeer>and a NPE. bleh17:54
oh, we're doing distribution with n=117:55
makes sense.
* github-ff joins17:59
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/NPcaUA
fcrepo4/master 9e1d071 Chris Beer: do synchronous replication in the infinispan chained-cache scenario.. we really care about the safety of the data (or it makes it easier to test that clustering is working)
* github-ff leaves
<bljenkins>Project fcrepo-kitchen-sink build #144: SUCCESS in 3 min 21 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/144/18:11
<pivotal-bot____>Chris Beer edited "JMeter delete objects return spurious 404sw" https://www.pivotaltracker.com/story/show/4745771318:12
Chris Beer edited "JMeter delete objects return spurious 404s" https://www.pivotaltracker.com/story/show/4745771318:13
<bljenkins>Project fcrepo-fixity build #136: SUCCESS in 3 min 47 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/136/
* travis-ci joins18:18
[travis-ci] futures/fcrepo4#336 (master - 9e1d071 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/1d0828227742...9e1d07174d6d
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/6064021
* travis-ci leaves
* ap2972 joins18:26
* nbanks joins18:40
* nbanks leaves18:44
* ap2972 leaves19:10
* ap2972 joins19:38
* nbanks joins19:41
* nbanks leaves19:53
* ap2972 leaves21:14
<pivotal-bot____>Vincent Nguyen edited "Create a REST endpoint for dynamically projecting files from a directory from a federated content as datastreams ..." https://www.pivotaltracker.com/story/show/4737377721:46
Vincent Nguyen started "Create a REST endpoint for dynamically projecting files from a directory from a federated content as datastreams..." https://www.pivotaltracker.com/story/show/47373777