Log of the #fcrepo channel on chat.freenode.net

Using timezone: Eastern Standard Time
* dwilcox joins08:35
* dhlamb joins08:43
* esm_ joins08:53
* acoburn joins08:56
<ruebot>awoods: just noticed something in the fcrepo ui -- the "Issue Tracker" link in the footer points to GitHub. Should it point to JIRA?09:08
* whikloj joins09:11
* osmandin joins09:12
* acoburn leaves09:14
<awoods>ruebot: JIRA would be better. thanks.09:17
* acoburn joins
* acoburn leaves09:26
<ruebot>awoods: shall i create a JIRA ticket for it then?09:29
* ruebot supposes he just created more work for himself
<awoods>ruebot: please ++
<ruebot>awoods: on it.
* acoburn joins
<awoods>ruebot: can you assign it to Bethany Seeger?09:30
<ruebot>awoods: oh, sure.
awoods: https://jira.duraspace.org/browse/FCREPO-173209:31
* ksclarke joins09:49
<nic-data>hi, i need some help getting Fedora to work on a MS Server 2012, anyone have any ideas how to get that to work? I posted in Fedora-Community last week but havnt got any further with our problems.10:03
<awoods>osmandin ^^10:14
acoburn: ping10:33
<acoburn>awoods: pong
<awoods>acoburn: good morning
<acoburn>awoods: it's a sunny day in vermont10:34
<awoods>acoburn: I am trying to work through the issues of deploying fcrepo-indexing-triplestore into karaf...
<acoburn>awoods: I am too
awoods: the issue is that jena-osgi doesn't like RDF/XML
awoods: my current tact is to use n-triples
<awoods>acoburn: would jena-core not make the feature happy?10:35
<acoburn>awoods: no
awoods: jena-core is part of jena-osgi
awoods: it's a deeper transitive dependency
awoods: like xercesImpl or xml-apis
<awoods>acoburn: is it a matter of finding the right library to add to the features.xml?
<acoburn>awoods: yes
<awoods>acoburn: it used to work, no?10:36
<acoburn>awoods: with clerezza, yes
awoods: that's why I chose clerezza — it actually works with OSGi
awoods: but with the new jena class introduced in 2.12.x, it caused the clerezza impl to fail10:37
awoods: since those bundles wrapped jena 2.11.x
awoods: so we needed to move fcrepo-camel to jena in order for them to pass the IT tests10:38
<awoods>acoburn: right
acoburn: It seems like you are on it. Is there anything in particular that I can do that would be helpful here?
<acoburn>awoods: I think I have a fix for it (to be verified once mvn install finishes)10:39
awoods: you can test it once the PR comes through
<awoods>acoburn: OK. Also, as you may know, dhlamb is going to be using the vagrant/karaf setup next Monday for the F4/HydraConnect workshop, and we wanted to send out the vagrant install directions today... or at the latest tomorrow.10:40
<acoburn>awoods: it will be ready this morning (hear that dhlamb?)
<dhlamb>acoburn: awoods: yup, i've been lurking10:41
<awoods>acoburn: that is great news, thanks!
<acoburn>awoods: dhlamb: yep, that fixed it10:43
awoods: I'll send a PR in just a sec
<awoods>acoburn: I'm ready10:44
acoburn: btw, did you happen to test the other features in camel-toolbox?10:45
* dwilcox leaves
<acoburn>awoods: most of them, but I'll do a more comprehensive test now10:46
awoods: really, though, we need to start using pax-exam for itesting
awoods: it's just that I wasn't able to figure it out when I last worked with it10:47
<awoods>acoburn++ pax-exam
<acoburn>awoods: I may have better luck now
<awoods>acoburn: if we have a plan, we can break up the investigation across more of the team.10:48
<acoburn>awoods: I think the plan is just "start using pax-exam for integration testing"10:49
<awoods>team ^^
<acoburn>awoods: and yes, it would be great to spread that across more of the team
awoods: it would be a great opportunity for whikloj to jump into the wonderful world of OSGi10:50
<awoods>acoburn: indeed, if only whikloj knew what an opportunity is at his doorstep.10:51
<whikloj>acoburn: if it is anything like the Wonderful world of Disney, count me out.
<osmandin>awoods: I'll take a look at the message from nic-data (In a meeting right now. Thanks.)
<whikloj>awoods: ....busy hiding
<acoburn>whikloj: it's nothing like the Wonderful World of Disney
<dhlamb>whikloj: fear is the mind killer
<awoods>nic-data: osmandin is our resident Windows liaison.
<whikloj>dhlamb: whale tuba
<f4jenkins>Project fcrepo4-T2 build #358: UNSTABLE in 4 min 33 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/358/10:52
<whikloj>awoods/acoburn: make your case, and it better not involve Mickey ;)10:53
<acoburn>whikloj: the case for OSGi: hot code swapping, runtime configuration, sane modularization, service lifecycle management10:54
<awoods>whikloj: https://ops4j1.jira.com/wiki/display/PAXEXAM3/Karaf+Container10:55
<whikloj>acoburn: I don't even understand maven
<acoburn>whikloj: it has nothing to do with maven
<dhlamb>acoburn: yes yes yes... tiny little workers all doing your bidding... that's OSGi
<awoods>whikloj: in other words, it won't help you better understand maven.
<dhlamb>acoburn: and if you don't lke one, you can change it without shutting down the whole stack10:56
<whikloj>awoods/acoburn: ok I'll have a look at the page. Is this a ticket or still in discussion?
<acoburn>whikloj: it's just an IRC discussion10:57
<dhlamb>whikloj: aside from some dense documentation, maven gets the job done. just another xml format you gotta learn
<awoods>acoburn: it sounds like whikloj would be interested in working a ticket
<whikloj>awoods: is there a WebAC call on Thursday too?10:58
<acoburn>whikloj: since there's no ticket yet, you can define it
whikloj: it's on Wed
<awoods>acoburn: what do you think a reasonable, incremental first step would be for pax-exam?
<whikloj>acoburn: thanks10:59
<awoods>whikloj: yes, as acoburn says, the webac meeting is Wed @3pm ET
<acoburn>awoods: for the nonce, forget about camel-toolbox and just focus on fcrepo-camel
<awoods>whikloj: https://ops4j1.jira.com/wiki/display/PAXEXAM3/Getting+Started+with+OSGi+Tests
<whikloj>awoods/acoburn: stepping away for a coffee, will do some reading later.11:00
<acoburn>awoods: all manual test passed11:06
manual tests--
<awoods>acoburn: passing++
acoburn: did you test all of the features?
acoburn: or just triplestore-indexing11:07
<acoburn>awoods: indexing-triplestore, indexing-solr, fixity, reindexing, audit-triplestore
awoods: that's all of them
<awoods>acoburn: great... I have a meeting @11:30, but am otherwise ready to review/test a PR11:08
* dwilcox joins11:17
<acoburn>awoods: the -toolbox is pegged to version 4.3.0 for fcrepo-audit. Should I leave that in place for this PR?11:20
awoods: I'd like to move away from -SNAPSHOT dependencies, wherever possible
<awoods>acoburn: what are you suggesting here?11:21
<acoburn>awoods: the current -toolbox code depends on fcrepo-audit/4.3.0 (not the -SNAPSHOT)
<awoods>acoburn: correct
<acoburn>awoods: I could either change that be 4.3.1-SNAPSHOT or leave it as is
<awoods>acoburn: and you would prefer to leave it at 4.3.0?11:22
<acoburn>awoods: if we begin going down the route of decoupling these components (and their releases), then I would expect dependencies would not include -SNAPSHOT
<awoods>acoburn: it seems like camel-toolbox should keep up to date with fcrepo-camel, no?
<acoburn>awoods: correct
awoods: that's pegged to -SNAPSHOT11:23
<awoods>acoburn: -SNAPSHOT allows up to keep projects up to date with each other.
<acoburn>awoods: right, but only if the release cycle keeps in lock step
<awoods>acoburn: for example, had fcrepo-camel-toolbox been depending on fcrepo-camel-SNAPSHOT, we would have known that there was a compilation problem.11:24
<acoburn>awoods: yes, I'm actually not suggesting removing that11:25
<awoods>acoburn: using ${project.version} requires lock-step. using some.version-SNAPSHOT does not.11:26
acoburn: what are you suggesting here?
<acoburn>awoods: I'm suggesting keeping fcrepo-audit pegged at 4.3.0
awoods: as it currently is
<awoods>acoburn: fcrepo-audit dependency defined in fcrepo-toolbox?11:28
<acoburn>awoods: yes
<awoods>acoburn: ok... going onto a call.11:29
acoburn: at what point would you be inclined to bump up that dependency version?11:30
<acoburn>awoods: when a new version is *released*
awoods: the idea is that the releases are decoupled
<awoods>acoburn: yes... but that is not the same thing as avoiding snapshot dependencies.11:32
<acoburn>awoods: I updated the PR: https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/pull/52
awoods: I'll need to think about this some more. I may find that I will contradict myself later11:35
dhlamb: I updated the fcrepo-camel-toolbox PR if you'd like to give it a try: https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/pull/5211:44
dhlamb: I tested it in karaf 4.0.1
* ksclarke leaves12:20
* ksclarke joins12:33
* ajs6f joins12:39
<dhlamb>acoburn: testing now. if it works, all that's left is figuring out the sed commands for any configuration that may be required12:46
acoburn: dumb question: what's the best file extension for a karaf script?12:52
<acoburn>dhlamb: not sure12:53
<dhlamb>acoburn: right now i'm using .script b/c that's what i saw in karaf documentation
<acoburn>dhlamb: that's what I'd use
<dhlamb>acoburn: k, rolling with it. they stick out like sore thumbs, tho!12:54
<awoods>acoburn: https://jira.duraspace.org/browse/FCREPO-1730 and PR-52 look good. Shall I squash/merge?13:13
<acoburn>awoods: yes, that would be great
dhlamb: how did your test go?13:14
* github-ff joins13:16
[fcrepo-camel-toolbox] awoods pushed 1 new commit to master: http://git.io/vZ5dD
fcrepo-camel-toolbox/master 56c1d73 Aaron Coburn: Replace clerezza code with jena...
* github-ff leaves
<dhlamb>acoburn: i'll let you know in a minute13:17
acoburn: vagrant is up. compiling now so i can point to my local maven and see
* github-ff joins
[fcrepo-camel-toolbox] awoods closed pull request #54: Update fcrepo dependencies to project.version (master...fcrepo-1731) http://git.io/vZ9z7
* github-ff leaves
* github-ff joins13:18
[fcrepo-camel-toolbox] awoods closed pull request #52: replace clerezza code with jena (master...fcrepo-1730) http://git.io/vZSnm
* github-ff leaves
* avmich joins
<awoods>dhlamb: PR-52 is in13:19
<avmich>I'm looking into the latest Fedora code. Is there any documentation regarding code layout? Where are the essential parts?
<awoods>dhlamb: do you think you will have your vagrant PRs ready for today?
<dhlamb>awoods: yes, might be in the PM, tho13:20
awoods: doing the juggle
<awoods>dhlamb: I would like to test your PR and get the message out either tonight or first-thing tomorrow.
dhlamb: we are cutting it close13:21
<dhlamb>awoods: but if this works, then all that's left is some seds to get the config going. then another branch and PR to include image magick for the conference
awoods: understood13:23
<awoods>avmich: are you looking for something in particular? interested in developing a feature or fixing a bug?13:25
* avmich leaves13:26
* esm_ leaves
* travis-ci joins13:28
fcrepo4-exts/fcrepo-camel-toolbox#149 (master - 56c1d73 : Aaron Coburn): The build passed.
Change view : https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/compare/44eeac17fb82...56c1d73dc750
Build details : https://travis-ci.org/fcrepo4-exts/fcrepo-camel-toolbox/builds/80476889
* travis-ci leaves
* avmich joins13:32
<dhlamb>awoods: ok, works on the vagrant build. on to configuration.13:34
<terrellt>Trying to spin up solr/fedora in one tomcat, I think log4j is exploding13:46
I get https://gist.github.com/terrellt/186cb7c0b05f1655ffb613:47
Ah, maybe permissions, but that's odd13:48
Needed to adjust my upstart script for tomcat to set the working directory to catalina_home.13:58
* dwilcox leaves14:06
* dwilcox joins14:12
<dhlamb>awoods: doing a clean destroy/up cycle on the config stuff. i'll keep you posted14:16
<awoods>go dhlamb, go!14:44
* github-ff joins14:49
[fcrepo4-vagrant] daniel-dgi opened pull request #21: Resolves FCREPO-1686 (master...karaf) http://git.io/vZdDX
* github-ff leaves
<dhlamb>acoburn: awoods: that's just for the OSGi deployment. i'll still have to make a branch for hydra connect
awoods: best way to go about that, considering it doesn't have/warrant a ticket?14:51
<awoods>dhlamb: I will create a branch for HC3 and you can create a PR against that branch14:52
* github-ff joins14:55
[fcrepo4-vagrant] awoods created hydraconnect3 from master (+0 new commits): http://git.io/vZd9L
* github-ff leaves
<awoods>dhlamb: here is your branch: https://github.com/fcrepo4-labs/fcrepo4-vagrant/tree/hydraconnect3
<dhlamb>awoods: ok, one last destroy/up cycle because i'm paranoid, and then i'll issue the pull to the HC3 branch15:00
<avmich>will wait :)15:01
<dhlamb>awoods: assuming the other PR is ok. someone vagrant up'ing it on a non-mac is probably a pretty good idea
<awoods>avmich: see http://irclogs.fcrepo.org/2015-09-15.html @13:2515:02
<avmich>thanks, awoods. Got disconnected. I'd like to get an overall understanding of the code layout, so I can better hunt problems15:04
I'm now trying to setup a Fedora cluster on two virtual machines15:05
<awoods>avmich: I assume you are starting with the core code base: https://github.com/fcrepo4/fcrepo4 ?15:06
<awoods>avmich: the primary modules are:
<avmich>I've checked https://wiki.duraspace.org/display/FEDORA4x/How+to+setup+a+Fedora+cluster , the last part15:07
didn't work
avmich: are you working from HEAD or a specific release?
<avmich>HEAD . Bad idea?15:08
<awoods>avmich: either way
<avmich>I don't think right now, at my current level of understanding, it's that important15:09
<awoods>avmich: I am spinning up a cluster right now, myself, using 4.3.0 on AWS
<awoods>avmich: amazon
<avmich>can I use VirtualBox?
<awoods>avmich: would expect so
<avmich>yeah, me too. But looks like it's not enough to mention IP addresses of cluster machines in maven startup line15:11
* dwilcox leaves15:12
<awoods>avmich: what error are you seeing?
<avmich>both instances work fine. When I upload a test file into one, I don't see it in the another15:14
I'm sure setting up a cluster is more than that though15:15
still confused about - which set of config files Fedora uses? The Git repo has several sets
<awoods>avmich: I need to focus on something else for a moment
* github-ff joins15:17
[fcrepo4-vagrant] daniel-dgi opened pull request #22: Imagemagick for HC3 (hydraconnect3...imagemagick) http://git.io/vZdAd
* github-ff leaves
* dwilcox joins15:18
<dhlamb>awoods: ok, PR is up. let me know what you guys find, and i'll pivot and take care of both PR's if need be
<awoods>dhlamb: thanks... reviewing it now. I noticed that audit events are not ending up in the triplestore.15:25
<dhlamb>awoods: oops, ok. i was checking solr and fuseki manually to make sure changes were there, but dropped the ball on that. didn't see any errors though, so probably a config issue15:26
awoods: will investigate
<awoods>dhlamb: what is "wrapper"? here: https://github.com/fcrepo4-labs/fcrepo4-vagrant/pull/21/files#diff-aa72fb46afa9a0e47c0cda586d6a3ea1R115:28
<dhlamb>awoods: that's karaf's 'service wrapper'15:29
awoods: lets you run it as a linux service. e.g. $ sudo service karaf-service start
<awoods>dhlamb: cool
<dhlamb>awoods: seems like maybe this event.baseUri config entry is an issue? event.baseUri=http://example.com/event15:32
awoods: not sure what that means in this context
acoburn: escowles: you around? ^^
<awoods>dhlamb: that should be ok
<acoburn>dhlamb: in a meeting
<escowles>dhlamb: yep15:33
<awoods>dhlamb: that is the prefix that shows up in the event messages
<escowles>dhlamb: I think that's the baseURI that you'd want to use in your triplestore -- could point to your repo if you're storing events in the repo
<awoods>dhlamb: hold on... you are good
dhlamb: I am seeing the audit events in the triplestore now... there must have been a delay15:34
<dhlamb>awoods: ?
awoods: ok, i'll take it :D
awoods: let me try on my install too, just to confirm
<awoods>dhlamb: that would be good...
dhlamb: I see my first audit event... but not the second yet... must be a delay.15:35
dhlamb: The "LIMIT" clause in the fuseki interface was throwing me off.15:38
dhlamb: by default it LIMITs to 25 results
<dhlamb>awoods: sample query for audit events?
awoods: i'm trying to query for premis:hasEventDateTime and i'm not getting anything :(15:39
<awoods>dhlamb: I am just using the "give me everything query"
SELECT ?subject ?predicate ?object
?subject ?predicate ?object
<dhlamb>awoods: good old SPO
<awoods>dhlamb: the audit events are the ones with "subject" being prefixed with "http://example.com"15:40
<dhlamb>awoods: alright, i see them now, but it is indeed trying to redirect me to 'example.com/event/blah-blah...'
awoods: and that's ok?15:41
awoods: wouldn't localhost:8080/fcrepo/rest/audit make sense for this?
<awoods>dhlamb: technically, yes.15:42
dhlamb: but it makes a good talking point as is.15:43
dhlamb: it makes it clear where the configuration item should be swapped in for the attendees
dhlamb: PR-21 is looking really good. Let me test the other features...15:44
dhlamb: fcrepo-fixity is not working because port 9080 is not open on the vagrant box15:47
* avmich leaves
<awoods>dhlamb: https://github.com/fcrepo4-exts/fcrepo-camel-toolbox/tree/master/fcrepo-fixity
<dhlamb>awoods: ah, ok. that's easy enough to fix15:48
<awoods>dhlamb: a fix to that will also enable the reindexer to work15:49
dhlamb: also running on port 9080
dhlamb: are you going to push another commit to your karaf branch?15:51
<dhlamb>awoods: yup
<dhlamb>awoods: momentarily
<osmandin>nic-data: do you still have an issue running fcrepon in Win 2008 R2, etc?15:53
<dhlamb>awoods: ok, ports are opened up on both PRs15:56
* ajs6f leaves
<dhlamb>awoods: but my features for fixity and re-indexing weren't installed. i'm doing a rebuild to make super sure15:57
* github-ff joins
[migration-utils] mikedurbin tagged 4.3.1-SNAPSHOT at master: http://git.io/vZFte
* github-ff leaves
* dhlamb needs to figure out docker...16:01
<awoods>dhlamb: is port 9000 needed?16:04
dhlamb: I thought both fixity and reindexing ran on 9080
<dhlamb>awoods: oh... cripes.... i need to increase the font on my monitor
* dhlamb facepalms
awoods: good catch16:05
awoods: hang on, build is almost finished. i'll double check and then strip out that 9000
<awoods>dhlamb: although a "vagrant provision" did not open up the port...
<dhlamb>awoods: i did 'vagrant reload' and it seemed to do it16:06
<awoods>dhlamb: I would be surprised if a full "vagrant destroy" is needed.
* awoods reloading...
<dhlamb>awoods: crazy, my install craps out after the audit-triplestore
* travis-ci joins
fcrepo4-exts/migration-utils#90 (4.3.1-SNAPSHOT - 59470e4 : nruest): The build passed.
Change view : https://github.com/fcrepo4-exts/migration-utils/compare/4.3.1-SNAPSHOT
Build details : https://travis-ci.org/fcrepo4-exts/migration-utils/builds/80505513
* travis-ci leaves
<dhlamb>awoods: yet a vagrant provision indeed does the trick...16:10
awoods: that's super odd... don't know why it just stops after audit-triplestore on my environment
* dhlamb facepalms again...
<awoods>dhlamb: you got it?16:11
<dhlamb>awoods: yeah, seems fine after running vagrant provision
awoods: services are up and running
awoods: testing more thoroughly now16:12
curl -X POST "localhost:9080/fcrepo/rest" -H "Content-Type: application/json" \16:14
-d '["activemq:queue:fixity"]'
awoods: that seem like an appropriate url?16:15
awoods: sry, first time using most of this
<awoods>curl -XPOST localhost:9080/reindexing/a -H"Content-Type: application/json" -d '["activemq:queue:fixity"]'
<acoburn>dhlamb: do a GET on localhost:9080/reindexing/16:16
dhlamb: that will give you a little help display
<dhlamb>acoburn: thx16:17
awoods: yes, re-indexing and fixity seem to be working now
<awoods>dhlamb: same here
dhlamb: nice job16:18
<dhlamb>acoburn: you continue to blow my mind. with the re-indexing, we can set up our listeners for creating drupal nodes and bootstrap an islandora site from any fedora 4 that follows PCDM!!!!
<acoburn>awoods++ dhlamb++
dhlamb: that's the idea — you can tie it into *any* endpoint
<dhlamb>acoburn: awoods: escowles: thx for the help guys.16:19
* dwilcox leaves
<acoburn>dhlamb: and you can start at any point in the repository, either from the root or from any arbitrary node
<dhlamb>acoburn: man, more and more, i see it. just keep decoupling till you have the smallest camel utility possible and it will get lots of re-use16:20
<awoods>dhlamb: I am going to push your commits... then send out a message to the attendees.
dhlamb: ...after testing your imagemajick branch
<acoburn>dhlamb: middleware++16:21
alright, i'm out of the office for the night, but will be online late. shoot me an email if you need anything from me16:22
* github-ff joins
[fcrepo4-vagrant] awoods pushed 1 new commit to master: http://git.io/vZFZL
fcrepo4-vagrant/master 82d1da5 Daniel Lamb: Karaf installation...
* github-ff leaves
* dhlamb leaves16:23
* github-ff joins
[fcrepo4-vagrant] awoods closed pull request #21: Resolves FCREPO-1686 (master...karaf) http://git.io/vZdDX
* github-ff leaves
* github-ff joins16:42
[fcrepo4-vagrant] awoods pushed 1 new commit to hydraconnect3: http://git.io/vZF8P
fcrepo4-vagrant/hydraconnect3 fc19d18 Daniel Lamb: Imagemagick for HC3 workshop
* github-ff leaves
* github-ff joins16:43
[fcrepo4-vagrant] awoods closed pull request #22: Imagemagick for HC3 (hydraconnect3...imagemagick) http://git.io/vZdAd
* github-ff leaves
* dwilcox joins16:44
* escowles leaves16:45
* avmich joins16:47
* dwilcox leaves16:56
* whikloj leaves16:57
* github-ff joins17:00
[fcrepo4-vagrant] awoods tagged fcrepo4-vagrant-4.3.1-SNAPSHOT at hydraconnect3: http://git.io/vZFuL
* github-ff leaves
* acoburn leaves17:01
* acoburn joins17:42
* acoburn leaves
* ksclarke leaves18:17
* the_mgt_ joins19:28
* the_mgt leaves19:32
* avmich leaves19:35
* ksclarke joins20:04
* sbmarks leaves20:21
* sbmarks joins20:22
* dhlamb joins
* ksclarke leaves21:00
* dhlamb leaves23:27