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

Using timezone: Eastern Standard Time
* kaarefc joins03:08
* kaarefc leaves03:10
* kaarefc joins03:12
* eddies leaves04:46
* eddies joins05:09
* eddies leaves
* eddies joins
* eddies leaves05:58
* eddies joins05:59
* eddies leaves
* eddies joins
* eddies leaves06:18
* eddies joins06:39
* eddies leaves
* eddies joins
* eddies leaves06:50
* eddies joins06:58
* eddies leaves
* eddies joins
* fasseg joins07:03
eddies: git time for a hangout for 5 mins?07:04
<eddies>give me 5mins
<fasseg>sure thx
<pivotal-bot_>Edwin Shin added "HTML response for /rest/ endpoint is missing node creation forms at /rest/objects" https://www.pivotaltracker.com/story/show/5098248108:52
Edwin Shin edited "HTML response for /rest/ endpoint is missing node creation forms at /rest/objects" https://www.pivotaltracker.com/story/show/50982481
* github-ff joins08:55
[fcrepo-kitchen-sink] eddies pushed 1 new commit to master: http://git.io/r9yoDg
fcrepo-kitchen-sink/master a76c8c1 Edwin Shin: Disables Metrics reporting to Graphite by default unless explicitly enabled via Spring profile
* github-ff leaves
<fasseg>eddies: I added perfromance increase factors for nodes and threads and it looks like this now: https://docs.google.com/spreadsheet/ccc?key=0Apnd_qlYdcqydDQ4UFVmdmY3ZHhOWEN0bXFENEpPREE#gid=009:04
* kaarefc leaves09:06
<pivotal-bot_>Frank Asseg added "Create a simple spreadsheet for calculating required times for ingest" https://www.pivotaltracker.com/story/show/5098364509:11
Frank Asseg started "Create a simple spreadsheet for calculating required times for ingest" https://www.pivotaltracker.com/story/show/50983645
* ajs6f joins09:12
* gregjansen joins09:14
<fasseg>eddies: no event model in HDFS but it's a new feature request for HDFS from 2011 :/ https://issues.apache.org/jira/browse/HDFS-174209:22
<ajs6f>no events from HDFS -> no events in MODE connector
<bljenkins>Project fcrepo-kitchen-sink build #341: SUCCESS in 30 min: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/341/09:26
* ajs6f leaves09:31
* kaarefc joins09:32
<gregjansen>I was going to add more events to the bagit connector, to support sequencing of new bags..09:35
* kaarefc leaves09:37
* jcoyne joins09:42
* kaarefc joins09:43
<fasseg>i see decreasing durations over the course of 1000 create object requests..thats kinda nice...but feels strange...as if the repo could be faster ;)09:47
<awoods>how are you measuring the process, fasseg?
<fasseg>jmeter atm09:48
ill publish the graph in a minute when the 100 objects are done
<awoods>that would be great.
fasseg: Hanging results as sub-pages of this page makes sense: https://wiki.duraspace.org/display/FF/Fedora+4+test+results09:51
<pivotal-bot_>Chris Beer added "Re-enable the kitchen sink deploy after 6/3" https://www.pivotaltracker.com/story/show/5098697509:52
Chris Beer edited "Re-enable the kitchen sink deploy after 6/3" https://www.pivotaltracker.com/story/show/50986975
<fasseg>awoods: a single threaded run looks like this atm: https://docs.google.com/file/d/0B5nd_qlYdcqyTWg1eTV0M3F0ZTQ/edit?usp=sharing09:54
Linux 64 / JDK7 / Tomcat7 with 16gb09:56
hmm maybe native tomcat APR will increase perf a bit09:59
<awoods>It does appear to warm up over time. Can you send me the ./log/jmeter-fedora4-xxxmean-xxxstddev-1threads.csv?10:00
<cbeer>and this is with random content?
<fasseg>sorry awwods still testing the tests, already did mvn clean ;)10:01
but i will upload the data to the wiki once finished
cbeer: this is only fcr:new
* ajs6f joins
<fasseg>no noctent at all
<fasseg>but it still seem to be IO bound...iostat gives me 98%10:03
so no increase when adding threads...
let's try the ramdisk :)10:04
* ajs6f leaves10:05
<fasseg>oh yes -> 22ms per request on a ramdisk instead of ~500ms on my spinning disk, that's def io bound10:07
* github-ff joins10:09
[fcrepo-kitchen-sink] cbeer pushed 1 new commit to master: http://git.io/Hgt9yg
fcrepo-kitchen-sink/master 7a4c596 Chris Beer: add org.fcrepo.url to component scanning
* github-ff leaves
* kaarefc leaves10:11
<bljenkins>Project fcrepo-kitchen-sink build #342: SUCCESS in 2 min 43 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-kitchen-sink/342/10:12
<cbeer>looks like rhauch and hchiorean are working on our issues this week for the release
* ajs6f joins10:13
* github-ff joins10:17
[fcrepo4] eddies pushed 1 new commit to master: http://git.io/FH0Iag
fcrepo4/master 3fca3fa Edwin Shin: Adds a favicon image, which still needs to be wired into our views with something like...
* github-ff leaves
<pivotal-bot_>Chris Beer started "Create favicon.ico for fcrepo4" https://www.pivotaltracker.com/story/show/50144729
<eddies>hey i was just about to do that
<cbeer>eddies: wasn't sure if you saw the ticket or just got annoyed at the logging10:18
<cbeer>annoyance-driven development.
<eddies>i assumed the ticket was born out of annoyance =)
<cbeer>our call is in 40m, right?
i have no idea how to wire that favicon into our velocity template however10:19
<cbeer>isn't that pretty easy?
<eddies>probably if i knew anything about velocity =)
<cbeer>there's just something that looks like html sitting around somewhere
<eddies>i stuck a reference in a file called header.vm in fcrepo-http-api but evidentaly that's not used by any of the views i'm seeing10:20
<cbeer>i wonder why there isn't a common template
yeah, i think that's pre-escowles' work
(but we may still have some html sitting around)
<eddies>ah. as in, that should be nuked?
<ajs6f>If that was my work, it was the barest-bones stuff.
* travis-ci joins10:24
[travis-ci] futures/fcrepo4#632 (master - 3fca3fa : Edwin Shin): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/b4a418cf5b21...3fca3fa6c7d4
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/7736096
* travis-ci leaves
<eddies>how the hell can the build error on adding a PNG?!10:26
pardon my french
<ajs6f>That was one hell of a favicon.10:27
<fasseg>maven tagged your png as offensive -> permban
<eddies>ah: plugin: failed: Plugin org.jacoco:jacoco-maven-plugin: or one of its dependencies could not be resolved: Failed to collect dependencies for org.jacoco:jacoco-maven-plugin:jar:
dependency resolution failure. wonder what happened to that10:28
<fasseg>this goddamn thing gives me headaches with the fixity webapp also, since it tries to instrument the test classes two times when running a "mvn clean test verify"
<bljenkins>Project fcrepo4 build #739: FAILURE in 12 min: http://ci.projectblacklight.org/jenkins/job/fcrepo4/739/10:29
<fasseg><skip>true</skip> atm
<bljenkins>Edwin Shin: Adds a favicon image, which still needs to be wired into our views with something like
<eddies>i guess some maven mirrors are having issues? but odd since travis and our jenkins should be hitting different mirrors10:30
* leb joins10:34
<awoods>fasseg: you can also use the commandline flag: -DskipTests10:35
<fasseg>well but i want run the unit tests and the integration tests, thats why i do a "mvn clean test verify"10:36
but jacoco tries to instrument classes in the test and the verify phase and the second time fails with an AlreadyInstrumentedException10:37
so the <skip>true<skip> is at the jacoco plugin config atm not at the maven-test-plugin
* leb leaves10:40
<cbeer>eddies: has duraspace figured out the CRM stuff you were talking about ages ago?10:41
<awoods>not yet
cbeer: do you have a certain need?10:42
<cbeer>one of the things i noted during barmintor's talk is (before OR) we really need a way to collect features and user stories from institutions that are willing and able to contribute developers to their pet features
e.g. XACML
<cbeer>sequencer features
i heard mumblings about object validation here
<ajs6f>OWL Full support
afk bbi1010:44
* ajs6f leaves
<eddies>cbeer: not really. they were already using sugarcrm and for the stuff i was interested in, seems like the workflow that's been settled on is to use google spreadsheets to generate reports and fill in data manually...10:45
which is kinda gross, but…i don't have much say in it
<cbeer>for OR, maybe we can toss a google form on top of it?
* gregjansen leaves10:48
* ajs6f joins10:49
<awoods>cbeer: some mechanism for effectively collecting community use-cases would be productive for OR, and otherwise.10:50
<pivotal-bot_>Andrew Woods edited "Setup stand-alone client and repo for JMeter on futures1 and 2" https://www.pivotaltracker.com/story/show/5089705510:53
<cbeer>eddies: you tried to use the hangouts app, right?10:55
<eddies>cbeer: on my iphone, yes. but y'all seemed to be pretty unhappy w/ sound quality and i was on wifi, not 3G
<pivotal-bot_>Andrew Woods added comment: "http://docs.codehaus.org/display/SONAR/JMeter+Plugin" https://www.pivotaltracker.com/story/show/50913615
<cbeer>i guess i can always be called in if it doesn't work
<eddies>cbeer: actually, wait. i think i was using the google+ app. i couldn't figure out within the hangouts app how to join the event hangout10:57
<cbeer>i tried to get barmintor to do a live-demo of sdef/sdep10:58
i don't think it worked.
(getting him to do it. doing it would never work.)
<eddies>you mean fcrepo3?
he's doing fcrepo 101
<eddies>actually i did it at RIRI once
went ok
<cbeer>but you probably prepared for it
<eddies>that would be true10:59
<ajs6f>Eddie Shin: He prepares for things.11:00
<cbeer>ok, i can't figure it out11:01
<pivotal-bot_>Chris Beer edited "Have Maven Checkstyles" https://www.pivotaltracker.com/story/show/4983697911:08
* github-ff joins
[fcrepo4] eddies pushed 2 new commits to master: http://git.io/mBx84g
fcrepo4/master 57de615 Nigel Banks: Checkstyles https://wiki.duraspace.org/display/FF/Code+Style+Guide
fcrepo4/master 0d09923 Edwin Shin: Merge pull request #71 from nigelgbanks/49836979-maven-checkstyles...
* github-ff leaves
<pivotal-bot_>Chris Beer started "don't get heckled by fcrepo3 fanatics at NEFUG" https://www.pivotaltracker.com/story/show/5074815311:09
* gregjansen joins
<pivotal-bot_>Chris Beer edited "don't get heckled by fcrepo3 fanatics at NEFUG" https://www.pivotaltracker.com/story/show/50748153
Chris Beer accepted "don't get heckled by fcrepo3 fanatics at NEFUG" https://www.pivotaltracker.com/story/show/50748153
<awoods>gregjansen: https://plus.google.com/hangouts/_/event/ccnv0ba94h3fb2je3gscunqc49g
<pivotal-bot_>Chris Beer added comment: "https://docs.google.com/document/d/14gfdFioZhKsW2R5Wa5MFPsYKGjOzBPbVWgxcpEiyCi0/edit?usp=sharing" https://www.pivotaltracker.com/story/show/5098364511:10
<cbeer>three, three main utilities are..11:13
four, four main utilities
<ajs6f>"Bring the comfy chair!"
<cbeer>ajs6f: The COMFY chair?!11:15
<ajs6f>"And the soft pillows!"11:16
<pivotal-bot_>Nigel Banks finished "Have Maven Checkstyles" https://www.pivotaltracker.com/story/show/4983697911:17
Frank Asseg added "Deployment of fixity webapp on fcrepo.org" https://www.pivotaltracker.com/story/show/5099583111:21
Frank Asseg edited "Deployment of cleaned up fixity webapp on fcrepo.org" https://www.pivotaltracker.com/story/show/50995831
Chris Beer added "Update jmeter readme" https://www.pivotaltracker.com/story/show/5099680311:28
Chris Beer edited "Update jmeter README and remove unused files" https://www.pivotaltracker.com/story/show/5099680311:29
Chris Beer edited "Update jmeter README and remove unused files" https://www.pivotaltracker.com/story/show/50996803
<jcoyne>Justin Coyne added "Make it work just like fcrepo3"11:30
<pivotal-bot_>Chris Beer edited "Update jmeter README and remove unused files" https://www.pivotaltracker.com/story/show/50996803
crazy fools.11:32
you're going to be in hte templates for the favicon, no?
<pivotal-bot_>Edwin Shin edited "HTML response for /rest/ endpoint is missing node creation forms at /rest/objects" https://www.pivotaltracker.com/story/show/50982481
<cbeer>i think it's just a matter of adding the form to the mode:root template
hm. we should try to do drupal + cmis on fcrepo4 http://drupal.org/project/cmis11:36
<pivotal-bot_>Edwin Shin added "Profile report of jmeter test plan" https://www.pivotaltracker.com/story/show/5099792911:40
Edwin Shin edited "Profile report of jmeter test plan" https://www.pivotaltracker.com/story/show/50997929
<eddies>grr. the joint on my magsafe connector just frayed and split. i hope i have another power adapter somewhere or i'm only going to be running for another 2 hours11:41
cbeer: can't tell if serious about drupal + cmis11:42
<cbeer>eddies: i can't get the page to load
but i'm semi-serious.
<eddies>at the very least, we'd need to do what i did for webdav—which is to decorate the nodes coming in via cmis with our fedora mixins11:43
oh. that reminds me the other thing i did. i was testing randall's fixes for os x + modeshape's webdav. didn't seem to fix the problem, but i might have been using a stale local build so not confirmed either way yet. i want to finish testing that just in case there's anything last minute to address before MODE 3.3 goes out this week11:45
* travis-ci joins11:50
[travis-ci] futures/fcrepo4#633 (master - 0d09923 : Edwin Shin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/3fca3fa6c7d4...0d099233579a
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/7737963
* travis-ci leaves
<pivotal-bot_>Chris Beer added "oembed" https://www.pivotaltracker.com/story/show/5099962111:56
* gregjansen leaves12:02
* gregjansen joins
<nbanks>I seem to be getting HTTP ERROR 404 for every page I visit, nothing is printing to stdout though, any end points that I could check that are likely to work?12:04
* gregjansen leaves12:09
<nbanks>never mind12:13
* gregjansen joins12:21
* ff-github joins12:46
[fcrepo-fixity] fasseg pushed 3 new commits to cleanup: http://git.io/8tNmmg
fcrepo-fixity/cleanup 2a66bc5 fasseg: updated jersey to use jackson serializer, changed javascript client accordingly
fcrepo-fixity/cleanup be2df06 fasseg: fixed details javascript
fcrepo-fixity/cleanup 54fc14c fasseg: changed bg color to mitigate eye ache
* ff-github leaves
<bljenkins>Project fcrepo-fixity build #216: STILL FAILING in 1 min 54 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/216/12:47
* ff-github joins12:49
[fcrepo-fixity] fasseg pushed 1 new commit to master: http://git.io/MLp2pw
fcrepo-fixity/master 1015704 fasseg: Merge branch 'cleanup': this was a rewrite of the old fixity webapp with...
* ff-github leaves
<bljenkins>Project fcrepo-fixity build #217: NOW UNSTABLE in 3 min 36 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/217/12:53
* ff-github joins12:55
[fcrepo-fixity] fasseg pushed 1 new commit to master: http://git.io/zoKpHA
fcrepo-fixity/master f6612a8 fasseg: updated readme
* ff-github leaves
<pivotal-bot_>Chris Beer added "Fix sitemap URLs to use URIInfo" https://www.pivotaltracker.com/story/show/51005391
Chris Beer edited "Fix sitemap URLs to use URIInfo" https://www.pivotaltracker.com/story/show/51005391
Chris Beer edited "Fix sitemap URLs to use URIInfo" https://www.pivotaltracker.com/story/show/51005391
<cbeer>brown's api responses: https://repository.library.brown.edu/api/pub/items/bdr%3A200069/12:57
<bljenkins>Project fcrepo-fixity build #218: STILL UNSTABLE in 2 min 45 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/218/12:58
* ff-github joins12:59
[fcrepo-fixity] fasseg pushed 1 new commit to master: http://git.io/bPJSlw
fcrepo-fixity/master bf98a04 fasseg: added test.port config to test container
* ff-github leaves
<bljenkins>Project fcrepo-fixity build #219: ABORTED in 14 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/219/
* gregjansen leaves13:00
<bljenkins>Project fcrepo-fixity build #220: STILL UNSTABLE in 3 min 9 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/220/13:02
<cbeer>ajs6f: clearly, your protege is presenting here at NEFUG.13:03
<ajs6f>Who? What?
<cbeer>fcrepo + stanbol, karaf, camel, etc
<ajs6f>Oh, right on, brother!
* gregjansen joins
<cbeer>i'm not sure who
oh, if i looked at the url..13:04
from amherst
<ajs6f>Don't know that name, but I've heard from folks at Amherst asking about that kind of stuff.
* ff-github joins13:05
[fcrepo-fixity] fasseg pushed 1 new commit to master: http://git.io/o58RHA
fcrepo-fixity/master eaa67db fasseg: added check for test.port system property
* ff-github leaves
<cbeer>he lacks the suit, pipe and hat though.
<ajs6f>He probably lacks the anxieties and tensions, too.13:06
<eddies>no credibility whatsoever without the appropriate attire
<ajs6f>The fez?13:07
<bljenkins>Yippie, build fixed!13:08
Project fcrepo-fixity build #221: FIXED in 2 min 58 sec: http://ci.projectblacklight.org/jenkins/job/fcrepo-fixity/221/
<fasseg>im off for today.. see you guys tomorrow...13:09
* fasseg leaves
<cbeer>awoods: has amherst expressed any interest in fcrepo4?13:10
ajs6f: https://acdc.amherst.edu/wiki/13:11
<awoods>not that I know of... on a call
<ajs6f>Wow. They have actual documentation. I'm envious. :)
* gregjansen leaves13:17
<ajs6f>Oh, Coburn. It just hit me, that is the guy who got in touch with me. Glad to hear he's taking those ideas forward.13:20
NEFUG folks— is there a webcast of that, or some other way I could see what happened, or is more informal than that?13:31
<cbeer>no, nothing.13:32
<ajs6f>That = the whole NEFUG presentations.
<cbeer>beyond http://blogs.lib.uconn.edu/nefug/nefug2013/
<ajs6f>Oh, cool. DURABLE!
<cbeer>hm. looks like there's some notes up there
i /am/ surprised how many people are doing weird custom things with fcrepo
<ajs6f>Yeah, a little.
It's a framework.
<cbeer>and someone that likes SOAP is talking now
(i feel a little bad for uconn)13:33
<ajs6f>He/she has dev resources to donate to fcrepo4?
<jcoyne>cbeer: That's been my experience in this niche, everybody wants to be different.
<cbeer>jcoyne: yup.. "nothing works for us.. we're using islandora for now, until we build our own thing"13:34
and SOAP services.
ajs6f: i'm reading some of the notes from other session.. " discussion of Islandora issues; doesn’t work right out of the box and requires lots of work still, so maybe we should start from scratch with Fedora? "13:35
<ajs6f>I've mumbling incoherently for weeks now that there is a huge stratum of folks who aren't impressed by the scripted frameworks in play and end up starting from the repo. It's not just old-school sites like UVa or Tufts.13:36
<cbeer>yeah. i'm just coming to realize the scale of that community
<jcoyne>I'm going to start my own new framework in Go or Rust.
Maybe Elixer
<cbeer>jcoyne's Go framework isn't good enough for me. i'm going to write my own.
<ajs6f>I'm going to write a framework-framework for building Fedora frameworks quickly.
<cbeer>(SOAP? WADL?)
maybe we need an abstraction layer over that.13:38
<ajs6f>And under it.
<cbeer>barmintor was suggesting a SOAP-over-REST API for fcrepo4
<eddies>i step away for an hour and i find people are itching for SOAP?13:41
<ajs6f>More importantly, that there may be more to life that the scripted frameworks currently extant for Fedora.13:43
<eddies>actually, i wonder how much intersection there is in the SOAP and CMIS communities
<ajs6f>CMIS offers a SOAP exposure, I think.
<eddies>because maybe we can just give folks their SOAP via our CMIS support.
<ajs6f>But the model is agnostic.13:44
<eddies>yes, that's where i was going
<ajs6f>That would be some damn cheap SOAP.
I like cheap.
<eddies>and we can wash our hand of it
<ajs6f>Where are the Google Hangout sound effects when you need them?
<eddies>i think i need to find a photo of someone washing their hands with soap for my or presentation13:45
yes. i can totally see the slide
<ajs6f>I can totally hear the gorans.13:46
<eddies>goram soap
cbeer: how does someone go from "Islandora […] doesn’t work right out of the box and requires lots of work still, so maybe we should start from scratch with Fedora"?! at least look at Hydra and even scratching that off the list I still don't see how someone arrives at "SOAP or bust"13:49
* jcoyne leaves
* eddies grumbles in the corner disconsolately13:50
<pivotal-bot_>Chris Beer added "fix fixity report" https://www.pivotaltracker.com/story/show/5101166914:00
<ajs6f>I'm with you, for once, on the SOAP thing. It's just completely irrelevant.14:14
But I'm not sure that's actually what was said. I think cbeer might have been making a joke.14:15
cf: v
"- discussion of Islandora issues; doesn’t work right out of the box and requires lots of work still, so maybe we should start from scratch with Fedora?"
* ajs6f leaves
* ajs6f joins14:23
* kaarefc joins14:51
<cbeer>eddies: those were separate threads of discussion15:02
but not a joke15:03
<ajs6f>That's what I was trying to say.
* nbanks leaves15:11
<ajs6f>Should we seed this:15:12
<awoods>sequencers were floated as an idea15:15
<ajs6f>Yup. And I think stuff with indexing...15:18
* kaarefc leaves15:28
* ajs6f leaves15:33
* ajs6f joins15:36
* github-ff joins16:02
[fcrepo-auth-oauth] ajs6f created master (+1 new commit): http://git.io/pKolOA
fcrepo-auth-oauth/master 085c1d6 ajs6f: Initial commit
* github-ff leaves
* nbanks joins16:03
* nbanks leaves16:13
<eddies>DROID is incompatible with Java 7. That sort of throws a wrench in the DROID sequencer idea16:33
http://digital-preservation.github.io/droid/ sez: DROID requires Java 6 Standard Edition (SE). DROID was built and tested on Java 1.6 update 30. DROID will not function in a Java 7 environment.16:34
<ajs6f>cbeer is only working on two or three other open source projects already just to support this one. I'm sure he can do that too. :)16:42
* ajs6f leaves16:52
* github-ff joins16:53
[fcrepo-auth-oauth] ajs6f pushed 1 new commit to master: http://git.io/s7oHGg
fcrepo-auth-oauth/master 6c93394 ajs6f: Making authN filters
* github-ff leaves
<awoods>is there a good website on SCAPE?17:08
* nbanks joins17:19
* nbanks leaves17:23
* github-ff joins19:48
[ff-jmeter-madness] awoods pushed 1 new commit to master: http://git.io/q7F5dQ
ff-jmeter-madness/master 4eb361c Jason MacWilliams: Update jmeter configuration file to allow commandline setting of output directory.
* github-ff leaves
* jonathangee leaves20:29