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

Using timezone: Eastern Standard Time
<eddies>hello01:19
<pivotal-bot>Hello, eddies
* eddies leaves01:48
* eddies joins02:21
* eddies leaves
* eddies joins
* eddies1 joins02:24
* eddies leaves
* eddies1 leaves
* eddies1 joins
* kaarefc joins02:36
* eddies1 leaves04:29
* eddies joins04:30
* eddies leaves
* eddies joins
* kaarefc leaves05:53
* eddies leaves06:52
* kaarefc joins07:33
* fasseg joins07:50
* eddies joins07:53
* eddies leaves
* eddies joins
* eddies leaves07:59
* eddies joins08:03
* eddies leaves
* eddies joins
* eddies1 joins08:07
* eddies1 leaves
* eddies1 joins
* eddies leaves08:08
* eddies1 leaves08:17
* VincentNG joins08:19
* eddies joins08:21
* eddies leaves
* eddies joins
* VincentNG leaves08:22
* VincentNG joins08:25
<fasseg>eddies, cbeer: so the unit tests in fcrepo4 spin up jetty instances...would it make sense to change these into integration tests using failsafe?09:08
<eddies>yep
<fasseg>this kinda blocks me atm for changeing to resteasy...09:09
since the unit tests use an special cxf dependency to spin up jetty instances before the junit tests
and that's not available in resteasy...
sooo shall I?09:10
create and grab a ticket for that?
sec brb09:12
* fasseg leaves
* fasseg joins09:14
* fasseg leaves09:16
* fasseg joins09:17
grmbl gdm was hogging one cpu core...
hurray for slim!09:18
soo eddies: shall i create a ticket to change the integartion tests into the failsafe lifecycle?
and start it?
<eddies>yes pls09:20
<fasseg>k
<pivotal-bot>Frank Asseg added "Switch unit tests to integration tests where applicable" https://www.pivotaltracker.com/story/show/4653743709:22
<fasseg>so the integration tests should prbably be moved to a separated module...right?09:26
eddies ^^
otherwise it quite hacky runing failsafe and surefire on the same build cycle
<eddies>as i recall, the default behavior for failsafe is to run classes that end in IT, e.g. FooIT09:28
<fasseg>and surefire will skip those by default?
<eddies>whereas surefire runs everything that ends in Test, e.g. FooTest
if you run mvn test, surefire runs. mvn verify runs *Test and *IT
that's my recollection. so no need for separate module09:29
<fasseg>ok cool, so ill just keep them in the same moduke
hmpf but the eclipse junit thingie runs all the tests including the ones ending in *IT09:35
* jcoyne joins09:41
* github-ff joins09:42
[fcrepo4] fasseg created failsafe-migration (+1 new commit): http://git.io/8sNrBg
fcrepo4/failsafe-migration adf2d28 fasseg: added failsafe renamed integration tests to *IT for legacy-api
* github-ff leaves
* github-ff joins09:46
[fcrepo4] fasseg pushed 1 new commit to failsafe-migration: http://git.io/RujvRg
fcrepo4/failsafe-migration 2195be1 fasseg: added failsafe plugin and moved integration tests to *IT for geenrator-dc
* github-ff leaves
* jcoyne leaves09:47
* jcoyne joins
<pivotal-bot>Edwin Shin edited "Update islandora to point at new, tomcat location for fcrepo4" https://www.pivotaltracker.com/story/show/4614227509:48
Edwin Shin added "Cost analysis of Glacier vs traditional storage" https://www.pivotaltracker.com/story/show/4653942309:49
Edwin Shin edited "Cost analysis of Glacier vs traditional storage" https://www.pivotaltracker.com/story/show/46539423
<fasseg>eddies: what about tests where a Modeshape instance gets spun up? integration or unit? strictly speaking I'd say those are integration tests as well..09:51
<eddies>fasseg: those are integration tests09:52
<fasseg>k
<cbeer>waiting for barmintor?10:00
<eddies>i guess so10:01
and fasseg
i think
<fasseg>hm?
<cbeer>is https://wiki.duraspace.org/display/FF/Sprint+Schedule right?
<eddies>fasseg: you're on this sprint "unofficially" right?
cbeer: it's mostly right ;-)10:02
<fasseg>not really, i have a lot to do on another project...
* github-ff joins10:05
[fcrepo4] fasseg pushed 1 new commit to failsafe-migration: http://git.io/Hyjs2Q
fcrepo4/failsafe-migration 4fffd1b fasseg: moved unit test to integration test for he rest of the modules
* github-ff leaves
<cbeer>what is all this renaming doing for us?10:07
<eddies>cbeer: let's us run unit tests separately from integration tests
<cbeer>ok, so.. blahIT is an integration test?
<eddies>yes
<cbeer>ok
<eddies>so mvn test by default won't run that10:08
but mvn verify will run *Test and *IT
<cbeer>do we have to do anything different on jenkins/travis then?
<eddies>hmm. dunno
<cbeer>(or, does install call verify?)
<eddies>on jenkins, it's running mvn install right?
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to failsafe-migration: http://git.io/fCzxGw
fcrepo4/failsafe-migration 366ece2 fasseg: added failsafe dependencies to submodules
* github-ff leaves
<eddies>yes, install runs verify
<bljenkins>Project fcrepo4 build #216: UNSTABLE in 4 min 9 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/216/10:09
Yippie, build fixed!10:15
Project fcrepo4 build #217: FIXED in 5 min 29 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/217/
<pivotal-bot>Chris Beer added "Iteration Shortlist" https://www.pivotaltracker.com/story/show/4654184310:19
Chris Beer edited "Iteration Shortlist" https://www.pivotaltracker.com/story/show/46541843
Chris Beer added "Demonstrate Metrics + Graphite integration" https://www.pivotaltracker.com/story/show/4654201710:21
Chris Beer edited "Demonstrate Metrics + Graphite integration" https://www.pivotaltracker.com/story/show/46542017
Chris Beer estimated "Demonstrate Metrics + Graphite integration" as 1 point https://www.pivotaltracker.com/story/show/4654201710:23
Edwin Shin added "Demonstrate Metrics + Nagios integration" https://www.pivotaltracker.com/story/show/46542175
Edwin Shin edited "Demonstrate Metrics + Nagios integration" https://www.pivotaltracker.com/story/show/46542175
* jcoyne leaves10:28
<cbeer>https://gist.github.com/cbeer/ca43f0f71a79fc3bceea10:30
* jcoyne joins10:31
<pivotal-bot>Edwin Shin added "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/46542837
<cbeer>https://www.hostedgraphite.com/
<pivotal-bot>Edwin Shin edited "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/4654283710:32
<bljenkins>Project fcrepo-kitchen-sink build #86: FAILURE in 16 hr: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/86/10:35
* barmintor joins10:36
<bljenkins>Project fcrepo-fixity build #79: FAILURE in 16 hr: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/79/
<pivotal-bot>Edwin Shin added "Tomcat 7 on futures6 Permgen Exception on re-deploy" https://www.pivotaltracker.com/story/show/4654330110:38
Edwin Shin edited "Tomcat 7 on futures6 Permgen Exception on re-deploy" https://www.pivotaltracker.com/story/show/46543301
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to failsafe-migration: http://git.io/eHWcLg
fcrepo4/failsafe-migration 8fd8ddb fasseg: moved integration test to org.fcrepo.integration
* github-ff leaves
<bljenkins>Yippie, build fixed!
Project fcrepo-kitchen-sink build #87: FIXED in 2 min 52 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/87/
Yippie, build fixed!
Project fcrepo-fixity build #80: FIXED in 2 min 8 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/80/
<pivotal-bot>Edwin Shin added "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/46543367
Edwin Shin edited "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/46543367
Edwin Shin edited "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/4654336710:39
* github-ff joins
[fcrepo4] fasseg merged failsafe-migration into master: http://git.io/n9Zwug
* github-ff leaves
<cbeer>barmintor: !?10:40
<eddies>barmintor?!
<barmintor>I made it.
<eddies>Barmintor the Absent has returned
<barmintor>I only had to walk 20 blocks at the end!
<eddies>come hang out with us =)
<barmintor>ok- should I also invite Alekesy?10:41
<eddies>sure
* travis-ci joins10:45
[travis-ci] futures/fcrepo4#260 (master - 8fd8ddb : fasseg): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/7dc98bf8e3ac...8fd8ddb49192
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5658375
* travis-ci leaves
<cbeer>booo10:47
<fasseg>check the error logs first, the error was introduced in build 25910:48
and i get the same error when building master on my local machine and i told eddies about that...
but nobody else seems to have those, maybe a race condition?
No node exists at path '/objects/testObject' in workspace "fedora"?
i get these sporadically10:49
and travis ci als well it seems
<cbeer>https://github.com/futures/fcrepo4/blob/master/fcrepo-kernel/src/main/java/org/fcrepo/spring/ModeShapeRepositoryFactoryBean.java#L39
that's where /objects should be created
i'm not sure how you're creating the repository object now10:50
<fasseg>no change
<bljenkins>Project fcrepo4 build #219: UNSTABLE in 5 min 19 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/219/
* frank asseg: added failsafe renamed integration tests to *IT for legacy-api
* frank asseg: added failsafe plugin and moved integration tests to *IT for geenrator-dc
* frank asseg: moved unit test to integration test for he rest of the modules
* frank asseg: added failsafe dependencies to submodules
* frank asseg: moved integration test to org.fcrepo.integration
<fasseg>that's just renames...
dont be so quick to judge, i got these errors 4 days ago the first time...10:51
<cbeer>weird.
<pivotal-bot>Chris Beer added "Fix JCR bug" https://www.pivotaltracker.com/story/show/4654478310:54
Chris Beer edited "Fix JCR bug" https://www.pivotaltracker.com/story/show/46544783
<fasseg>there is some stuff left in fcrepo-kernel after mvn clean and i thought that was the cause but removing this before a cluen did nothing
<cbeer>the travis builds are on a brand-new VM each time, so definitely not left over stuff
<pivotal-bot>Chris Beer started "Fix JCR bug" https://www.pivotaltracker.com/story/show/4654478310:55
<fasseg>but i introduced a bug where the application context could not be loaded by jenkins in the last build
checking this out atm
i think it's just @runwith missing in the webhook test though10:57
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to master: http://git.io/ELnTrQ
fcrepo4/master f69b070 fasseg: added runwih annotation for spring context
* github-ff leaves
<cbeer>hm. i'm not getting failures locally.11:00
that's annoying.
<pivotal-bot>Chris Beer edited "Demonstrate pointing fcrepo4 on top of an external data source" https://www.pivotaltracker.com/story/show/4603960111:05
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to master: http://git.io/ZSfLNw
fcrepo4/master d1637d9 fasseg: fixed package name in jaxrs bean
* github-ff leaves
* travis-ci joins11:11
[travis-ci] futures/fcrepo4#262 (master - d1637d9 : fasseg): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/f69b07095254...d1637d936cfa
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5659116
* travis-ci leaves
* github-ff joins
[fcrepo4] fasseg pushed 1 new commit to master: http://git.io/RwEbKA
fcrepo4/master bcf7c28 fasseg: typo in jaxrs xml, wargl
* github-ff leaves
<cbeer>at least it consistently fails on travis11:12
hm. who is supposed to create that testObject anyway?
oh, i see it there11:13
* travis-ci joins
[travis-ci] futures/fcrepo4#261 (master - f69b070 : fasseg): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/8fd8ddb49192...f69b07095254
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5658915
* travis-ci leaves
<fasseg>a lot fater though than jenkins .)
*faster
<pivotal-bot>Edwin Shin edited "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/4654283711:14
<fasseg>hmm now it passed locally w/o the errors
but no change in fcrepo-kernel11:15
* github-ff joins11:17
[fcrepo4] fasseg pushed 1 new commit to master: http://git.io/93YSuQ
fcrepo4/master 1c5eb5a fasseg: missing point in package name for webhook
* github-ff leaves
* travis-ci joins11:19
[travis-ci] futures/fcrepo4#263 (master - bcf7c28 : fasseg): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/d1637d936cfa...bcf7c28b1adc
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5659374
* travis-ci leaves
<pivotal-bot>Benjamin Armintor added "Unit test fcrepo-http-api" https://www.pivotaltracker.com/story/show/46547091
<fasseg>cbeer: it passes in eclipse using the junit plugin but fails when i run it on the console...maybe this helps?11:20
<cbeer>fasseg: i'm going to try giving the different classes different fedora objects to use11:21
because, as long as i'm not misreading things, the tests that fail change
so, i'm suspicious that the junit rollback is.. funny.
<pivotal-bot>Edwin Shin edited "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/46543367
Edwin Shin edited "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/4654336711:22
<fasseg>cbeer: jenkins build did also pass, very strange indeed...11:23
<pivotal-bot>Edwin Shin edited "Automated code coverage reporting of futures projects" https://www.pivotaltracker.com/story/show/46274571
<bljenkins>Yippie, build fixed!
Project fcrepo4 build #223: FIXED in 5 min 50 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/223/
frank asseg: missing point in package name for webhook
<pivotal-bot>Edwin Shin started "GSoC brainstorming" https://www.pivotaltracker.com/story/show/44476621
Edwin Shin edited "Automated code coverage reporting of futures projects" https://www.pivotaltracker.com/story/show/46274571
<cbeer>fasseg: hey, did you get the SelfHealing test to pass when run in the suite?11:24
<fasseg>yeah
<cbeer>fasseg++
<fasseg>wait no changes....
<cbeer>but it passes now.
so maybe that's a side-effect of the failsafe work?
<fasseg>so i didnt fix any broken tests or anything
<pivotal-bot>Edwin Shin added comment: "See: http://mojo.codehaus.org/cobertura-maven-plugin/" https://www.pivotaltracker.com/story/show/4627457111:25
<fasseg>err no idea
<pivotal-bot>Chris Beer edited "Make it possible to run JUnit tests with per-test or per-class modeshape configurations" https://www.pivotaltracker.com/story/show/46484445
Edwin Shin added comment: "Also see: http://www.sonarsource.org/" https://www.pivotaltracker.com/story/show/46274571
<bljenkins>Project fcrepo-fixity build #81: SUCCESS in 1 min 9 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/81/
Project fcrepo-kitchen-sink build #88: SUCCESS in 2 min 25 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/88/11:26
* github-ff joins
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/rechQg
fcrepo4/master 2b4bfc3 Chris Beer: use unique node names for the integration tests
* github-ff leaves
<fasseg>same error for me...:/11:28
<pivotal-bot>Edwin Shin added "Add Aleksey to futures6" https://www.pivotaltracker.com/story/show/4654823111:31
Edwin Shin edited "Add Aleksey to futures6" https://www.pivotaltracker.com/story/show/46548231
<bljenkins>Project fcrepo4 build #224: UNSTABLE in 5 min 0 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/224/
Chris Beer: use unique node names for the integration tests
* travis-ci joins11:32
[travis-ci] futures/fcrepo4#265 (master - 2b4bfc3 : Chris Beer): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/1c5eb5ab7db3...2b4bfc3d533b
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5659910
* travis-ci leaves
<cbeer>i thought java was supposed to make these cross-platform issues a thing of the past.
* travis-ci joins11:33
[travis-ci] futures/fcrepo4#264 (master - 1c5eb5a : fasseg): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/bcf7c28b1adc...1c5eb5ab7db3
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5659575
* travis-ci leaves
<fasseg>err?!
<cbeer>i really broke it11:34
* github-ff joins
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/eyjQSw
fcrepo4/master 95b721f Chris Beer: fix object name in self-healing test
* github-ff leaves
<fasseg>ahh! i tried changing the contextconfiguration to master.xml of DatastreamIT and now only SelfHEalingIT fails...
cbeer ^^
with a constrainviolation...rest passes
<cbeer>all master does is also pull in eventing11:35
and we don't need eventing in these tests, i hope.
<pivotal-bot>Edwin Shin added comment: "I'd appreciate at least a cursory overview of pros/cons of the three tools listed above and why we selected w..." https://www.pivotaltracker.com/story/show/46543367
Edwin Shin edited "Add automated code-quality checking" https://www.pivotaltracker.com/story/show/46543367
<cbeer>but, hey, whatever works, i guess
<pivotal-bot>Edwin Shin edited "Demonstrate Metrics + Nagios integration" https://www.pivotaltracker.com/story/show/46542175
<barmintor>is it possible I'm not an owner of ff-tech?
eddies ^^
<eddies>who's j. donovan11:36
<cbeer>barmintor: checking
<eddies>barmintor: sec, let me check ff-tech
<cbeer>eddies: jack.
i assume
<barmintor>cbeer++ // correct
<cbeer>oh, eddies is the only owner of ff-tech11:37
<eddies>ok, just made cbeer and barmintor managers
<barmintor>thanks eddies11:38
<eddies>and you guys were already managers of ff-internal
* github-ff joins11:40
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/WEfngg
fcrepo4/master 5f84e3e Chris Beer: trivial examples of the metrics plugin, exposed through the DatastreamTest
* github-ff leaves
<bljenkins>Yippie, build fixed!11:41
Project fcrepo4 build #225: FIXED in 6 min 42 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/225/
Chris Beer: fix object name in self-healing test
<fasseg>cbeer: so the evnting bus does nothing when persisting nodes? I'm a bit intrigued that now everything passes the selfhealing test fails..11:43
<cbeer>i broke the self-healing test.11:45
and fixed in "fcrepo4/master 95b721f Chris Beer: fix object name in self-healing test"
<fasseg>yeah and I saw the event bus is just used in the webhook, the rsspublisher and the simpleobserver, so scratch that11:46
still fails for me11:48
* ap2972 joins11:50
* travis-ci joins11:51
[travis-ci] futures/fcrepo4#266 (master - 95b721f : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/2b4bfc3d533b...95b721fcc6b9
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5660114
* travis-ci leaves
<bljenkins>Project fcrepo-kitchen-sink build #89: SUCCESS in 3 min 17 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/89/11:52
Project fcrepo-fixity build #82: SUCCESS in 3 min 7 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/82/11:53
* travis-ci joins11:58
[travis-ci] futures/fcrepo4#267 (master - 5f84e3e : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/95b721fcc6b9...5f84e3ea2886
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5660380
* travis-ci leaves
* VincentNG leaves12:08
<cbeer>i wonder if it'll pass again..12:09
guess we're using a snapshot release of metrics.12:11
grr. and they took away the ability to dump metrics12:20
<eddies>cbeer: eh? what's been taken away?
that dump method you were using in your test class?
<cbeer>eddies: well, or is a whole lot uglier12:21
i was passing a handful of arguments
and now it looks like i have to pass about 8
and do a lot of stuff manually
considering filing a feature request
<eddies>that's from the "more arguments == better" school of thought, i guess12:22
<cbeer>https://github.com/codahale/metrics/blob/master/metrics-core/src/main/java/com/yammer/metrics/ConsoleReporter.java#L35
that's a little.. excessive
at least give me a method with reasonable defaults
<eddies>ok. i have another call coming up. afk for ~1hr12:24
<pivotal-bot>A. "Schwarmerei" Soroka edited "Demonstrate pointing fcrepo4 on top of an external data source" https://www.pivotaltracker.com/story/show/4603960112:31
* ajs6f joins12:35
* VincentNG joins
* github-ff joins13:00
[fcrepo4] cbeer pushed 1 new commit to master: http://git.io/lzc71A
fcrepo4/master 67316f1 Chris Beer: add metrics for some fcrepo-kernel methods
* github-ff leaves
<bljenkins>Project fcrepo4 build #227: UNSTABLE in 3 min 37 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/227/13:04
Chris Beer: add metrics for some fcrepo-kernel methods
* ap2972 leaves13:05
* travis-ci joins13:22
[travis-ci] futures/fcrepo4#268 (master - 67316f1 : Chris Beer): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/5f84e3ea2886...67316f1044b1
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/5663065
* travis-ci leaves
<pivotal-bot>Vincent Nguyen started "Automated code coverage reporting of futures projects" https://www.pivotaltracker.com/story/show/4627457113:58
Vincent Nguyen added comment: "Multi-module builds aren't supported by Cobertura but there's a patch according to http://jira.codehaus.o..." https://www.pivotaltracker.com/story/show/4627457114:01
* ap2972 joins14:11
* github-ff joins14:13
[fcrepo4] fasseg created resteasy (+1 new commit): http://git.io/GU1UPg
fcrepo4/resteasy ff5e0a6 fasseg: resteasy for first part of legacy api
* github-ff leaves
<bljenkins>Yippie, build fixed!14:18
Project fcrepo4 build #229: FIXED in 4 min 58 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/229/
* ap2972 leaves14:28
* eddies leaves14:30
* eddies joins14:35
* eddies leaves
* eddies joins
<pivotal-bot>Chris Beer edited "Add Aleksey to futures6" https://www.pivotaltracker.com/story/show/4654823114:37
Chris Beer finished "Add Aleksey to futures6" https://www.pivotaltracker.com/story/show/46548231
Edwin Shin added comment: "MCOBERTURA-65 is marked fixed against 2.5. What's still not supported is coverage of moduleB by tests in modu..." https://www.pivotaltracker.com/story/show/4627457114:40
Vincent Nguyen added comment: "Which VM should we use to run Sonar?" https://www.pivotaltracker.com/story/show/46274571
Edwin Shin added comment: "I don't know. futures6 is feeling a bit over-utilized compared to our other vms. I'd ask on irc if anyone has..." https://www.pivotaltracker.com/story/show/4627457114:43
<VincentNG>Anyone have a preference as to which VM we use to run Sonar?14:49
maybe future1? i believe it's running jmeter, so maybe this is our "tools" server?14:50
<pivotal-bot>A. "Schwarmerei" Soroka added comment: "On hold waiting for MODE 3.2. Ref: this conversation: ""14:56
rhauch
2:50 ajs6f: can you gist the error ..." https://www.pivotaltracker.com/story/show/46039601
* ap2972 joins14:57
* ap2972 leaves14:59
* ap2972 joins
<eddies>vincentng: in the absence of objections, i say go for futures115:14
<VincentNG>eddies: thanks :-)15:15
* jcoyne leaves15:19
<pivotal-bot>A. "Schwarmerei" Soroka added comment: "See also: https://issues.jboss.org/browse/MODE-1862" https://www.pivotaltracker.com/story/show/46039601
<cbeer>on the cutting edge, i guess.15:20
<barmintor>ajs6f: is there a way to get on the DPN mailing list that's not asking Mr. Morales to become a member institution?15:21
<ajs6f>barmonster: Don't know, but I will find out.15:22
<barmintor>thank you
<eddies>ajs6f++ nice discussion on #modeshape15:23
<pivotal-bot>A. "Schwarmerei" Soroka added comment: "MODE 3.2 will be released around the second week of April, per rhauch: ""
ajs6f: fwiw, it looks li..." https://www.pivotaltracker.com/story/show/46039601
<ajs6f>What was really awesome about it was how the resolution created no new work for me. :)
<eddies>i was wondering if they were going to push a release out as originally scheduled today given the as-yet-unresolved blockers/critical issues15:24
<ajs6f>barmintor: Are you talking about the general DPN announce list?
<cbeer>"except for all the things broken about the release, it's in good shape"
<ajs6f>eddies: And what's more, rhauch is about to start a week of vacation. I remember what vacation felt like.
<barmintor>ajs6f: I'm talking about whatever list is going to be notified of the location of the recording of that 2pm webex today :P
<cbeer>we need to remember that
<ajs6f>barmintor: Oh, yeah— the one I couldn't get into.15:25
<barmintor>yay!
<ajs6f>barmintor: If you just want the recording, I can ask for the link itself. It ought to be up-ish.
<barmintor>ajs6f: one of my co-workers did, it sounds like a re-hash of earlier stuff, but my boss will want me to have an opinion
<eddies>vacation? i think that's what i call the extra hours of sleep i get on sunday
<cbeer>DPN... is deep. en?15:26
<ajs6f>Deep fried preservation goodness.
<cbeer>if only they supported fcrepo 4.x!
<ajs6f>APTrust is expecting to use it/15:27
<eddies>all the cool kids are
<ajs6f>_And_ APTrust.
<eddies>heh
<cbeer>(clearly our esteemed steering group hasn't set expectations suitable low yet.)15:28
a small, sleep deprived group of developers has created software that passes most of its tests most of the time.15:29
what could go wrong
(but, hey, you have fixity checks and self-healing.. so i guess, really, what could go wrong?015:30
hello
<pivotal-bot>Hello, cbeer
<cbeer>hm
<eddies>goodbye15:31
bye
<cbeer>i'm stealing this ticket: https://www.pivotaltracker.com/story/show/46542837
<pivotal-bot>feature: Add Metrics Servlet to kitchen-sink (unstarted) / owner: Edwin Shin
Chris Beer edited "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/46542837
<cbeer>there we go
<eddies>i support this sort of thievery
* jcoyne joins15:38
Did you guys see this: http://basho.com/riak-cloud-storage/
Just open sourced today
<eddies>hadn't even heard of it before. that's a lot cheaper than s3, assuming there aren't any hidden data transfer rates15:43
<jcoyne>eddies: you're looking at the "Enterprise" offering.15:44
I was more excited about the stuff that preceded that.
<eddies>guilty as charged. i was skimming ahead15:45
<jcoyne>eddies: Biggest issue is "content up to 5GB"15:46
<eddies>jcoyne: where's that stated?
<jcoyne>http://docs.basho.com/riakcs/latest/
<eddies>chunk it!15:47
<jcoyne>That's an option.
<eddies>wonder where that 5gb limit comes from15:48
<jcoyne>eddies: I'm not certain that the "Enterprise" is a hosted service. I think it might be a licensing fee.
eddies: Startup opportunity!15:49
<eddies>oi
<jcoyne>You can be in charge of our Asia data center.
<eddies>oh boy15:51
<jcoyne>Can you relocate to HK?
<eddies>my g6 is in the shop15:52
<cbeer>eddies/ajs6f: ok, i'm lost. I have to define a servlet context parameter, and i guess it needs to be the result of a static method of a class
i gather there's some bean magic i can do?15:53
<eddies>oh look. a lost thief ;-)
<cbeer>turns out the answer must be so obvious, it wasn't worth documenting anywhere
<ajs6f>cbeer: Back me up. What are we trying to do?15:54
<eddies>can you point me to a class or docs?
<ajs6f>cbeer: What's the context.
?
<cbeer>https://github.com/codahale/metrics/blob/master/metrics-servlets/src/main/java/com/yammer/metrics/servlets/MetricsServlet.java#L32
<eddies>ajs6f, i think he's trying to set up: http://metrics.codahale.com/manual/servlet/
<cbeer>that METRICS_REGISTRY attribute needs to be the value of our RepositoryService.getMetrics()15:55
(because they helpfully kill the default registry in 3.0.0-SNAPSHOT :/ )
<ajs6f>So the MetricRegistry needs to be in the ServletContext, but that can arise by several different means— why do we need to do it with a static param?15:57
or method, rather.
<cbeer>i guess we don't. that's just where I know to get it15:58
<ajs6f>Oh, there's a static method that supplies a MR?
From their code, I mean.
<cbeer>no, from our code.
they provide a factory, though
so, as far as i know, i need to get either one of these values into that servletcontext:15:59
<ajs6f>Okay… so we have to get the result of that into the ServletContext, eh?
<cbeer>- the result of RepositoryService.getMetrics()
- new MetricRegistry("fcrepo");
<ajs6f>Right. Crossed wires.
<cbeer>and i've never understood web.xml in the first place.16:00
<ajs6f>Okay, here's a weird thought— how about a Servlet Filter that just injects that attribute into the Context on the way in...?16:02
<eddies>that sounds cleverer. i was going to say just create a subclass of MetricRegistry16:05
<ajs6f>Or (instead of a filter) a ServletContextListener which could inject the attribute once at app startup.
eddie: How complex is MetricRegistry? Maybe we want our own?16:06
<eddies>https://github.com/codahale/metrics/blob/master/metrics-core/src/main/java/com/yammer/metrics/MetricRegistry.java
<cbeer>so, i think i understand the filter suggestion
what is subclassing metricregistry getting us? it's the servlet that's the problem, isn't it?
<eddies>but metricservlet (as i understand) just wants something that is instanceOf MetricRegistry
<cbeer> final Object registryAttr = config.getServletContext().getAttribute(METRICS_REGISTRY);16:07
if (registryAttr instanceof MetricRegistry) {
this.registry = (MetricRegistry) registryAttr;
} else {
throw new ServletException("Couldn't find a MetricRegistry instance.");
}
<eddies>oh. i thought you were just needing to pass in a class as a web.xml param
<cbeer>that.
<eddies>and that that class needed to basically super("fcrepo")
i've been drinking cognac as of 15mins ago so i may not be thinking as clearly as i ought =)16:08
quitting time and all here in singapore ;-)16:09
<ajs6f>barmintor: Still trying to get you that link. Even the DPN team can't get it from WebEx yet, but when they do I'll get it from them.
<barmintor>ajs6f: :P
thanks
<ajs6f>Wow, millions of people all over SIngapore, drinking cognac all at the same time.
<cbeer>eddies: oh, come on! it's only 4! slacker.
<eddies>i woke up at 9:30am :P
<ajs6f>barmintor:16:10
https://meetings.webex.com/collabs/files/playRecording?encryptedDocID=0_F91505970EFE83957CE09F6686FFD1C56AA99B4968D7E04EAF2B2380858F54E19BBC2C176FD2A470ACB5D46717285DD5_8DCEA27959AFB6F84A0579983BA8D96B95D1059F
<eddies>just for that i'm pouring myself another
<ajs6f>I wasn't there, but I bet it was the BEST PRESENTATION EVER!
<barmintor>ajs6f: no it wasn't16:11
<eddies>fedora4: driving you to drink
ack. i should remember that this channel is logged
<ajs6f>Fedora4: Ending careers.
(He said, before getting on his bicycle to go to the local brewery to fill his growler.)16:12
<barmintor>we're just putting your career in a dark archive
ajs6f: thanks so much for that, btw
<ajs6f>b: no prob. If you like slides, you're in for a TREAT!16:13
Now that I've managed to cold cock filesystem federation into the next version, I'm going to go back to the shiny thing: reifying queries to provide dynamic collections.16:14
Wow. This presentation used the phrase "dissolution or divestment of depositors". We need more alliteration in fcrepo.16:15
<eddies>federation or force-feeding of fcrepo16:17
<ajs6f>freaky fun for foolish fffff… archivists.16:18
afk
<eddies>ok. off for a bit. may be back later.16:19
ajs6f: if you come to boston, i want more apple brandy!
<ajs6f>eddies: That makes me happy!
* github-ff joins16:20
[fcrepo-kitchen-sink] cbeer pushed 1 new commit to master: http://git.io/cDXCsQ
fcrepo-kitchen-sink/master 468fb95 Chris Beer: hack in our own version of the MetricsServlet that has our registry hard-coded
* github-ff leaves
<cbeer>so, that works
<ajs6f>Drinking Laird's next to calvados explains more about the New and Old World than any number of Henry James novels.
<cbeer>i couldn't follow your suggestions i'm afraid.
<ajs6f>But you got something going?16:21
<cbeer>https://github.com/codahale/metrics/blob/master/metrics-servlets/src/main/java/com/yammer/metrics/servlets/MetricsServlet.java
theirs ^
ours v
https://github.com/futures/fcrepo-kitchen-sink/blob/468fb95044ba3167b743cc1658e1b1430679391d/src/main/java/org/fcrepo/metrics/MetricsServlet.java
<ajs6f>So we kinda "forked" their deal?
<cbeer>their lines 32-37 -> our line 33
i hope there's an obvious way to do that that i was missing16:22
<ajs6f>ell, you just bound their types right into our kernel, no?
<cbeer>maybe they're just doing it wrong.. wouldn't be the first time
<ajs6f>I think we ideally want some indirection, like through JNDI or something like that.
Are metrics really a kernel-level concern?16:23
<cbeer>i think eddies thinks they are.
<ajs6f>Not that I'm beefing at you…
<cbeer>(i'd note, there's also some spring aop annotation things that'd maybe make it better.. but i couldn't figure them out.. and they are tied to spring-aop)
<ajs6f>Well, okay. Fair enough. If you want, I can take a look at breaking it back out into a listener.
<cbeer>although i did find a metrics-cdi thing too
<ajs6f>Yeah, Spring-aop is pretty powerful, but it's a over-the-head dive into Spring.16:24
I don't know of many vendor-agnostic approaches to the same problems (as opposed, say, to DI).
Well, I'm not going to make a big stink about about it. If metrics are a core concern, than by gum they belong in the kernel.16:25
<cbeer>it might be nice to encapsulate them in our own deal, but that's a job for later, imo16:26
let's prove that the metrics we can gather are actually useful
<ajs6f>Right on, metric meister.
And while we're at it, let's switch the US to metric.16:27
Okay, I got to head out. That last line was really silly.
See y'all soon.
* ajs6f leaves
<pivotal-bot>Chris Beer finished "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/4654283716:31
Chris Beer delivered "Add Metrics Servlet to kitchen-sink" https://www.pivotaltracker.com/story/show/46542837
Chris Beer edited "Instrument fcrepo-kernel API methods with metrics" https://www.pivotaltracker.com/story/show/4647380316:32
Chris Beer finished "Instrument fcrepo-kernel API methods with metrics" https://www.pivotaltracker.com/story/show/46473803
Chris Beer added comment: "re-scoped to just the kernel... it looks like other modules may be more or less tricky.. there's a module to ..." https://www.pivotaltracker.com/story/show/4647380316:33
<cbeer>ok, heading out too16:37
* github-ff joins16:49
[fcrepo4] fasseg pushed 3 new commits to resteasy: http://git.io/tpdn9Q
fcrepo4/resteasy 9bad234 fasseg: removed rest of CXF; added some Resteasy stuff
fcrepo4/resteasy 268001d fasseg: added exception mapper to create 404s
fcrepo4/resteasy 16eb6ad fasseg: using commons-exception handlers in resteasy, annotations added
* github-ff leaves
* jcoyne leaves16:50
<bljenkins>Project fcrepo4 build #231: UNSTABLE in 3 min 10 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/231/16:52
* github-ff joins17:09
[fcrepo4] fasseg pushed 1 new commit to resteasy: http://git.io/o6BWOg
fcrepo4/resteasy 1227758 fasseg: moved http-api to resteasy
* github-ff leaves
* VincentNG leaves17:12
<bljenkins>Yippie, build fixed!17:13
Project fcrepo4 build #232: FIXED in 4 min 21 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/232/
* jcoyne joins17:17
* ap2972 leaves17:19
* kaarefc leaves17:26
* jcoyne leaves18:46
* github-ff joins18:48
[fcrepo4] barmintor deleted unit-testing at 6beaf28: http://git.io/jmR8xw
* github-ff leaves
<cbeer>barmintor: can you help ap2972 get onto futures6 sometime? i gave him an account and password, and i assume he's not doing something right18:53
<barmintor>cbeer: sure, I'll look at ti tomorrow
* ap2972 joins19:19
* github-ff joins19:46
[fcrepo4] barmintor created unit-testing (+1 new commit): http://git.io/ddRwSQ
fcrepo4/unit-testing 5f6fadc Benjamin Armintor: moving away from static imports to mockable/testable service objects
* github-ff leaves
* barmintor leaves19:47
<cbeer>+119:48
<bljenkins>Project fcrepo-kitchen-sink build #90: FAILURE in 6 hr 11 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/90/20:31
Project fcrepo-fixity build #83: FAILURE in 6 hr 11 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/83/
Yippie, build fixed!20:32
Project fcrepo-fixity build #84: FIXED in 1 min 43 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/84/
Yippie, build fixed!20:34
Project fcrepo-kitchen-sink build #91: FIXED in 3 min 23 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/91/
* jcoyne joins
<pivotal-bot>Chris Beer added comment: "http://fcrepo4.fcrepo.org/fcrepo/metrics" https://www.pivotaltracker.com/story/show/4654283720:35
* ap2972 leaves21:04
* jcoyne leaves21:36
* ap2972 joins21:47
* ap2972 leaves21:48
* VincentNG joins22:17
* VincentNG leaves22:19
* jcoyne joins22:58
* barmintor joins23:41
* github-ff joins23:50
[fcrepo4] barmintor pushed 1 new commit to unit-testing: http://git.io/-mBkhA
fcrepo4/unit-testing 618c811 Benjamin Armintor: fixing a bug in Datastream not to assign mixins to existing nodes; porting ObjectService changes to remaining modules; IT passing
* github-ff leaves
* github-ff joins23:53
[fcrepo4] barmintor opened pull request #41: Unit testing (master...unit-testing) http://git.io/SvIUwg
* github-ff leaves
<cbeer>+1 ship it.
<barmintor>if it's not all-caps I don't think you mean it.23:54
also, I want to look again after a night's sleep.
<bljenkins>Project fcrepo4 build #235: UNSTABLE in 4 min 16 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/235/23:55
* github-ff joins23:56
[fcrepo4] fasseg pushed 1 new commit to resteasy: http://git.io/I1lQuQ
fcrepo4/resteasy 1ec2c77 fasseg: fixed the rest of the single tests, but still the complete suite fails, since the jetty server does not get stopped somehow
* github-ff leaves
<fasseg>eddies: I got a problem with the resteasy migration: I was able to migrate all of the resources and the test run when started individually from the maven submodule directoy, but when the tests are run from the parent dir, it seems jetty does not get stopped correctly and some spring beans are left over, so that a new jetty instance can not be started.00:00
eddies: im quite finished now and will go to bed, id be happy if you can take a look at that branch: https://github.com/futures/fcrepo4/tree/resteasy00:02
<eddies>fasseg: i'll give it a whirl
<fasseg>thanks! and good night00:03
<eddies>night!
* fasseg leaves
<bljenkins>Yippie, build fixed!
Project fcrepo4 build #236: FIXED in 7 min 25 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo4/236/
Project fcrepo-fixity build #85: SUCCESS in 2 min 8 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/85/00:07
Project fcrepo-kitchen-sink build #92: SUCCESS in 4 min 6 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/92/00:08

Generated by Sualtam