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

Using timezone: Eastern Standard Time
* awead joins05:55
* awead leaves06:37
* awead joins06:38
* dwilcox joins07:47
* awead_ joins08:16
* awead leaves08:17
* dhlamb joins08:24
* dwilcox leaves08:56
* dwilcox joins08:59
* jgpawletko joins09:02
* acoburn joins09:04
* umgrosscol joins09:29
* osmandin joins09:37
* acoburn leaves09:56
* acoburn joins10:00
* github-ff joins10:34
[fcrepo4-vagrant] awoods pushed 1 new commit to master: http://git.io/vcSzi
fcrepo4-vagrant/master 9d284d7 mesemus: Update README.md...
* github-ff leaves
* github-ff joins10:35
[fcrepo4-vagrant] awoods closed pull request #26: README.md - workaround instructions if reindexing service is not started (master...master) http://git.io/vcWrL
* github-ff leaves
* whikloj joins10:39
<awoods>whikloj: ping10:42
<whikloj>awoods:
<awoods>whikloj: good morning10:43
<whikloj>awoods: morning, hope it's nice there. Raining here.
<awoods>whikloj: nice here... sunny and chilly 40F
* ksclarke joins10:44
<awoods>whikloj: As you know, we will be sending out the latest fcrepo4-vagrant and webac script today...
whikloj: for webac stakeholders
whikloj: thanks for your work this weekend
whikloj: As you likely saw, I added some comments to your PR
<whikloj>awoods: np, I can make the code review changes this morning
<awoods>whikloj: I wanted to get this out soonish... let me know if you would like me to take any of those comments.10:45
<f4jenkins>Project fcrepo4-T2 build #378: UNSTABLE in 5 min 4 sec: http://jenkins.fcrepo.org/job/fcrepo4-T2/378/10:54
* sdidomenico joins11:17
<whikloj>awoods: I'll leave you the publicImage fix if you don't mind, my java knowledge for objects in 3 dimensions is lacking.11:18
<sdidomenico>Hi all, quick question about Fedora v3: I'm seeing quite a large amount of temporary files in fedora/tomcat/temp/mulgara_<user>
the files are named things like tuples3337988283230522439_ml11:19
Can I just delete these files, or will only a restart of Fedora clear them? The reason I ask is because the volume is close to filling up.
<awoods>sdidomenico: There is not much Fedora3 activity on this channel... you will probably get a better response from the fedora-tech@googlegroups.com mailing list.11:20
<sdidomenico>Ok thanks awoods11:21
<awoods>whikloj: I am trying to map your five scenarios (01.acl - 05.acl) to the use cases: https://wiki.duraspace.org/display/FF/Design+-+WebAccessControl+Authorization+Delegate#Design-WebAccessControlAuthorizationDelegate-Phase111:32
whikloj: If you already have that mapping, that would help.
<whikloj>awoods: ok, no I don't I can look at them as well. I was working on the ASCII art thing11:33
<awoods>whikloj: just checking.
whikloj: have you tested your ACLs? I am not sure how to authenticate the agents defined in the authorizations since the vagrant instance only defines three users: fedoraAdmin, testuser, adminuser.11:37
<whikloj>awoods: no, I was going to try direct curls and see if they work.11:38
<awoods>whikloj: the users still need to authenticate
whikloj: even with curl11:39
<barmintor>sdidomenico: I'm only not replying to your list posting b/c "I don't know" isn't a useful response.
<whikloj>awoods: ok, I can replace all the users with testuser in the acls.
<awoods>whikloj: that makes sense... and having a consistent naming convention for the resources would be nice.11:40
<sdidomenico>barmintor: Thanks, well it's at least more helpful at the moment than no response at all :) Thanks!
<awoods>whikloj: looking at what is on fcrepo4-vagrant, we can pass in group information with the header of "some-header"11:41
<whikloj>awoods: via a curl call you mean?
<awoods>whikloj: yes: 'curl -utestuser:password1 -H"some-header: Registrar" localhost:...'11:42
<whikloj>awoods: Really only the first scenario uses a specific target user, the rest use a group (ie. Editors, Admins) so I will update the one acl
<awoods>whikloj: and the groups can be passed in with the header, which is the recommended approach for the first phase.11:43
<whikloj>awoods: exactly
* umgrosscol leaves11:48
<barmintor>sdidomenico: FWIW I'm looking at Mulgara release notes and not seeing anything through the 2.x line that mentions temp files11:50
sdidomenico: but the FCR3 buffering upstream is nor, IIRC, done in tmp files
* osmandin leaves11:52
<sdidomenico>Great, thanks barmintor. Our current guess is that these files are used as temporary space for searching (not indexing) -- in which case we could delete them12:05
barmintor: What we're going to try now is turning off some of our ingestion scripts to give Fedora some idle time, and hopefully it will do some of its own housecleaning.
<barmintor>sdidomenico: I'll cross my fingers for you12:06
<sdidomenico>barmintor: if that doesn't work, we'll be restarting Fedora, which should really do it
barmintor: haha thanks
* jrgriffiniii joins12:46
* jrgriffiniii leaves12:57
* jrgriffiniii joins
* umgrosscol joins13:01
* ajs6f joins13:16
barmintor: Were you complaining to me because you put lots of stuff in APT last week?13:17
<barmintor>ajs6f: still am depositing13:18
<ajs6f>barmintor: Rock it like a hurricane. A hurricane of care for the scholarly record. Not a real hurricane.
<barmintor>ajs6f: prolly ~3.5T when we're done13:19
we're at 4500 out of 5.2k objects13:20
<ajs6f>barmintor: And every bit is special in its own special way.
* acoburn leaves
<barmintor>meh
* acoburn joins
<ajs6f>barmintor: Andew Diamond (APT) just got up and said that there are now X bags in APT, almost all of which came in last week from you and us. That's why I asked.13:21
<barmintor>yes, that is true: it is durbin's fault our deposit is going so slowly
<ajs6f>barmintor: wha?
<barmintor>ajs6f: we were queued behind the UVa uploads that Durbin was responsible for AFAIK13:22
<ajs6f>barmintor: Oh, right. Forgot about the joint queuing.
barmintor: I wonder if the direct access via I2 could enable parallel action.13:23
* ajs6f leaves13:25
* acoburn1 joins14:34
* acoburn1 leaves
* acoburn1 joins
* acoburn1 leaves
* jrgriffiniii leaves14:36
* acoburn leaves14:37
* jrgriffiniii joins14:38
* sdidomenico leaves14:59
* dwilcox leaves15:12
* jrgriffiniii leaves15:38
* jrgriffiniii joins15:40
* acoburn joins15:41
* ajs6f joins15:47
* acoburn leaves15:55
* acoburn joins15:56
* acoburn leaves16:01
* acoburn joins
<awoods>whikloj: any news?16:05
<whikloj>awoods: no none, you mean on the PR?
<awoods>whikloj: yes
<whikloj>awoods: Did you miss my request about 3 dimensional things in java (trying to tie URI, acl URI and rdf:type into a static something)16:06
awoods: which I figured you could do with your eyes closed, underwater, while asleep
* dhlamb leaves16:07
<awoods>awoods: There was a request?
whikloj: There was a request?
whikloj: can you point me to the request?16:08
<whikloj>awoods: sorry maybe I dreamt that, or started to type it and failed to finish.
awoods: It would have been in IRC, and I can't so it was my mistake. My apologizes
^^ I can't "see it" so..
awoods: Can either a) take over the PR, or b) give me some easy way to get three Strings per List here https://github.com/whikloj/fcrepo-sample-dataset/blob/fcrepo-1736/src/main/java/org/fcrepo/repository/WebACProtectedResources.java#L5016:10
<awoods>whikloj: Go ahead and checkin your latest, and I will see what I can do with the PR tonight/tomorrow.16:11
<whikloj>awoods: This is my latest, I hope it goes far enough. https://github.com/whikloj/fcrepo-sample-dataset/tree/fcrepo-173616:12
<awoods>whikloj: thanks! I will keep you posted.
* ajs6f leaves17:04
<awoods>acoburn: ping17:11
<acoburn>awoods:
<awoods>acoburn: I am creating an fcrepo camel route against a webac-enable repo, and wondering how to get the authZ working...17:12
acoburn: https://github.com/fcrepo4-exts/fcrepo-camel#setting-basic-authentication
* jrgriffiniii leaves
<awoods>acoburn: I'm wondering where to set those properties
<acoburn>awoods: there are two ways to do this17:13
awoods: on a per-endpoint basis or a component-wide basis
<awoods>acoburn: component-wide seems clean
acoburn: in this scenario
<acoburn>awoods: if the auth values might change for each fcrepo: call you do the first, otherwise, the second is usually better
awoods: in that case, in the blueprint.xml file you create a bean with id="fcrepo"17:14
awoods: on that page ^^^, this is described in the "configuring the fcrepo component" section
<awoods>acoburn: I was missing the bean part... let me give it a shot.17:15
* acoburn leaves
* acoburn1 joins17:16
awoods: here's an example: Unable to locate Spring NamespaceHandler for XML schema namespace [http://java.sun.com/xml/ns/javaee]
awoods: oops darned copy/paste
awoods: http://git.io/vcQm717:17
awoods: I'm thinking of making a fcrepo-camel-blueprint maven archetype, so that it's easier to start out with creating new camel routes17:19
<awoods>acoburn1: works, thanks17:20
<acoburn1>awoods: that's great
<awoods>acoburn1: your archetype would be for creating more toolbox-style features?17:21
<acoburn1>awoods: exactly
<awoods>acoburn1: that could be nice
<acoburn1>awoods: it's much better than cp -R
<awoods>acoburn1: and cooler17:22
<acoburn1>awoods: well, if it's cooler, you know it needs to be a priority17:23
<awoods>acoburn1: ;)
afk
* umgrosscol leaves17:24
* acoburn1 leaves
* dshalvi joins17:41
* ajs6f joins17:49
* ajs6f leaves17:50
* whikloj leaves17:59
* dshalvi leaves18:10
* barmintor leaves18:45
* dwilcox joins19:25
* the_mgt_ joins19:45
* the_mgt leaves19:48
* dwilcox leaves19:49
* awead_away leaves20:15
* ksclarke leaves20:51
* ksclarke joins21:01
* ksclarke leaves21:02
* ksclarke joins21:14
* ksclarke leaves21:18
* jgpawletko leaves21:31
* dhlamb joins21:46
<awoods>dhlamb: ping21:59
<dhlamb>awoods: pong
<awoods>dhlamb: do you have a moment for a camel/thumbnail question?22:00
<dhlamb>awoods: absolutely
<awoods>dhlamb: https://github.com/daniel-dgi/fcrepo-camel-workshop/blob/master/06-ThumbnailGenerator.xml
dhlamb: I am getting 0 size png files on the upload
dhlamb: but if a put a <log> command after line:38, I see jpeg binary.22:01
dhlamb: thoughts on debugging?22:02
<dhlamb>awoods: run convert from the command line on the image and see what pops. if it's a blank file, it's an issue with the flags to convert
<awoods>dhlamb: running locally works fine22:03
<dhlamb>awoods: also, beware the stream. if you log it, the file is gone once the stream is read
awoods: hrm... ok
<awoods>dhlamb: sure... I only logged the stream to inspect the issue
dhlamb: do you have your sample jpegs somewhere?
dhlamb: I wonder if mine are too big
<dhlamb>awoods: i just used some memes. i can zip some up if you wanna try on others.22:04
awoods: how big is too big?
<awoods>dhlamb: mine are in the 2M range22:05
<dhlamb>awoods: i can't imagine that's the problem22:06
<awoods>dhlamb: hmm
<dhlamb>awoods: hang on, let me try it out to see if i can recreate
awoods: i've got the box running from when i tested out the url encoding stuff. should be recent enough?22:07
<awoods>dhlamb: yes, basically the same box
<dhlamb>awoods: hang on, got grace period on the deploy :(22:11
awoods: ok, was missing camel-exec22:19
<awoods>dhlamb: I have camel-exec22:20
karaf@root()> feature:list|grep camel-exec
camel-exec | 2.15.1 | x | Started | camel-2.15.1
<dhlamb>awoods: oh yes, that was just for me. wasn't on the h3c branch22:21
awoods: testing now
awoods: ok, worked fine for me22:22
<awoods>dhlamb: that is good and bad22:23
<dhlamb>awoods: indeed
<awoods>dhlamb: can you send me some jpegs?
<dhlamb>awoods: will do
awoods: emailed to your duraspace account22:26
* awoods waiting...22:27
dhlamb: got it
dhlamb: I am going to wipe my vagrant and start fresh... seeing the same issue.22:29
<dhlamb>awoods: good idea22:30
awoods: ping me if you still are having problems
<awoods>dhlamb: will do... thanks22:33
dhlamb: same issue on a clean vagrant... hmm22:58
<dhlamb>awoods: so if the convert command works fine in bash, and you can log the file... doesn't leave much22:59
awoods: and after the convert, got an image when you log?23:00
<awoods>dhlamb: and the error stack trace shows the body of the generated png...23:01
dhlamb:yes
dhlamb: but null and 0 size gets to the fcrepo PUT exchange
<dhlamb>awoods: hrm... have you tried PUTs with fcrepo-camel outside of this context?23:03
<awoods>dhlamb: no23:04
dhlamb: that seems like a reasonable test
* github-ff joins23:11
[fcrepo-camel-toolbox] acoburn opened pull request #61: Replace CamelBlueprintTestSupport with plain old CamelTestSupport (master...fcrepo-1670-with-spring) http://git.io/vc70Z
* github-ff leaves
* github-ff joins23:12
[fcrepo-camel-toolbox] acoburn closed pull request #51: Update camel toolbox to use fcrepo-transform from fcrepo4-exts (master...fcrepo-1670) http://git.io/vZ2Xl
* github-ff leaves
<dhlamb>awoods: getting 403 on PUTs. maybe my environment. is authentication enabled by default now?23:18
<awoods>dhlamb: yes, webac is enabled23:20
dhlamb: oddly, now 06-ThumbnailGenerator.xml do not deploy
<dhlamb>awoods: grace period?23:21
<awoods>dhlamb: grace period?
<dhlamb>awoods: log?
<awoods>dhlamb: what are you saying?
<dhlamb>awoods: Status of 06-ThumbnailGenerator.xml when you run bundle:list23:22
<awoods>dhlamb: 164 | GracePeriod | 80 | 0.0.0 | 06-ThumbnailGenerator.xml
<dhlamb>awoods: does it say grace period? stopped?
<awoods>dhlamb: what is the GracePeriod?23:23
<dhlamb>awoods: it's failed a runtime dependency. it's attempting for a window to resolve before stopping entirely
awoods: hang on, let me check what to grep for in the logs23:24
* dhlamb searches bash history
<awoods>dhlamb: it was camel-exec23:25
<dhlamb>awoods: ok, good23:26
awoods: don't forget apt-get install imagemagick23:27
awoods: let me know
<awoods>dhlamb: it is there, thanks
dhlamb: I'm unable to PUT a text string to fcrepo... same error:23:34
org.fcrepo.camel.FcrepoOperationFailedException: HTTP operation failed invoking http://localhost:8080/fcrepo/rest/collection/test3/try/thumbnail with statusCode: -1 and message: null
<dhlamb>:(23:35
* dhlamb leaves23:51
* awoods leaves00:08
* ksclarke joins00:22
* ksclarke leaves01:05