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

Using timezone: Eastern Standard Time
* fasseg joins04:47
* github-ff joins07:37
[fcrepo4] escowles pushed 1 new commit to projection-tests: http://git.io/4v6aKg
fcrepo4/projection-tests 5cdf71a Esmé Cowles: Adding debugging output to filesystem tests that are failing in Jenkins
* github-ff leaves
* travis-ci joins08:00
[travis-ci] futures/fcrepo4#1695 (projection-tests - 5cdf71a : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/7811dc5c40b5...5cdf71a09182
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22686237
* travis-ci leaves
* github-ff joins08:24
[fcrepo4] escowles pushed 1 new commit to projection-tests: http://git.io/_hASmw
fcrepo4/projection-tests 89258de Esmé Cowles: Debugging temp directory
* github-ff leaves
<pivotal-bot>Esme Cowles added comment: "All of the tests are failing for the same reason: uploading a file to the filesystem projection is failing. ..." https://www.pivotaltracker.com/story/show/6868919808:35
* travis-ci joins08:39
[travis-ci] futures/fcrepo4#1696 (projection-tests - 89258de : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/5cdf71a09182...89258de1ae83
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22688919
* travis-ci leaves
<pivotal-bot>Andrew Woods added comment: "BTW, please make your updates and new commits in the same branch so that they will automatically be applied..." https://www.pivotaltracker.com/story/show/6492915608:43
* mikeAtUVa joins09:02
* ksclarke joins09:16
* ksclarke leaves09:17
* gregjansen joins
* ksclarke joins09:20
* gregjansen leaves09:48
* scossu joins09:53
<ksclarke>awoods: the readyshare line isn't being used for anything else right now is it? I'd like to test connecting to via the dial-out method in a google hangout (my other computer to phone method stopped working recently)10:26
<awoods>ksclarke: no, the readytalk line is not being used right now... shall I start it up?10:27
<ksclarke>yes, thanks
<awoods>ksclarke: hit it10:28
* awoods lonely10:29
* gregjansen joins10:36
<pivotal-bot>Andrew Woods added "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/6926697210:42
Andrew Woods estimated "Ensure benchtool multi-threaded connection pool" as 1 point https://www.pivotaltracker.com/story/show/69266972
Andrew Woods started "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/69266972
Andrew Woods added comment: "https://github.com/futures/benchtool/pull/14" https://www.pivotaltracker.com/story/show/6926697210:43
Andrew Woods edited "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/69266972
Andrew Woods finished "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/69266972
Andrew Woods delivered "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/69266972
Andrew Woods accepted "Ensure benchtool multi-threaded connection pool" https://www.pivotaltracker.com/story/show/69266972
* github-ff joins10:46
[fcrepo4] escowles pushed 1 new commit to projection-tests: http://git.io/OXbuNw
fcrepo4/projection-tests 58865d9 Esmé Cowles: Disabling versioning test that enabled auto-versioning
* github-ff leaves
<awoods>escowles: the excitement: http://ci.fcrepo.org/jenkins/view/FF/job/fcrepo4-experimental/10:47
<escowles>awoods: i think the issue might be FedoraVersionsIT.testRepositoryWideAutoVersioning() which updates the CND to enable auto-versioning10:48
<awoods>escowles: that would make sense
<escowles>if that's running before the filesystem tests, then that would explain why versioning is enabled
<awoods>escowles: if that is the cause, then we need to update the cnd after the versioning to disable the config.10:49
<escowles>awoods: yes, i've just disabled the test for now, but i can update it to restore the original CND
<awoods>escowles: that is what the "projection-testt" branch is for, your tweaking tests.10:50
* gregjansen leaves10:56
* martinjd joins10:57
* Longshou joins11:01
* travis-ci joins11:15
[travis-ci] futures/fcrepo4#1697 (projection-tests - 58865d9 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/89258de1ae83...58865d9ec56b
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22699119
* travis-ci leaves
<ksclarke>what's the current state of ingest from f3 to f4? is there a tool for that already?11:47
<awoods>mikeAtUVa ^^11:50
<mikeAtUVa>ksclarke: there's not yet a tool that I think would be useful...11:51
<ksclarke>ok, thanks
<mikeAtUVa>ksclarke: There's a lot of questions that remain... do people want to migrate in a lossless way, or do they want to take advantage of new features and best practices?
<ksclarke>I remember hearing (I think) about file projection over the older fedora file system, but was wondering more about the new features/best practices route11:52
<escowles>ksclarke: https://wiki.duraspace.org/display/FF/Fedora+3+to+4+Upgrade11:53
<ksclarke>thanks escowles
<escowles>awoods: the jenkins build passed when i disabled the versioning CND change, so i'll update that and create a clean PR for doing that12:06
<awoods>escowles: Are you talking about re-enabling the versioning test (with the update to back-out the CND updates that that test introduces after the test's completion)?12:07
<escowles>awoods: yes, i'll update the test to undo the CND changes so it won't conflict with the filesystem federation test12:08
<awoods>escowles: great12:09
escowles/ksclarke/longshou: plans for the day?
<escowles>when i'm done with that, i can move on to some of the other small tickets in the backlog (or i could try the version deletion ticket)
<awoods>mikeAtUVa: does it make sense to hand-off/work-with escowles and the version deletion?12:11
<ksclarke>so still not seeing any slow down in ingest performance across the backends (with larger runs); should I document https://wiki.duraspace.org/display/FF/Single-Node+Test+Results or on another page? wondering what is needed to wrap up that ticket.12:12
<mikeAtUVa>awoods: sure... looking at the pull request from the "reversion" ticket should illuminate every file that would need to be touched.
<awoods>escowles: where does index-rebuild stand? https://www.pivotaltracker.com/story/show/65168574
<pivotal-bot>feature: Enable index rebuild (unstarted) / owner: Esme Cowles
<awoods>ksclarke: Maybe create a new page under: https://wiki.duraspace.org/display/FF/Performance+Testing12:13
<escowles>awoods: it needs this ticket done so the IT will pass without a sidecar fcrepo4: https://www.pivotaltracker.com/story/show/69192280
<pivotal-bot>feature: Configure indexer webapp to run fcrepo4 for integration tests (unscheduled) / owner:
<mikeAtUVa>awoods: I'm working through the locking ticket... there's quite a bit to consider before I get to the implementation.
<awoods>ksclarke: that is focused on "flat hierarchies"
<mikeAtUVa>awoods: also, I will be out tomorrow and won't make the standup meeting (or make any progress)12:14
<awoods>escowles: so it if you could give this a review, that might be helpful: https://www.pivotaltracker.com/story/show/60883346
<pivotal-bot>feature: Version Management (finished) / owner: Mike Durbin
<ksclarke>then I'll look at ACL/writer ticket
<escowles>awoods: definitely, i'll take a look at that PR12:15
<awoods>mikeAtUVa: we will miss you tomorrow
As for myself, I am going to update the 4.0 Feature Documentation (adding "admin console" and "linked data"), and push some more on the clustering-for-scalabilty tests (given ksclarke's recent benchtool updates). I would also like to have a better idea of what to expect out of the filesystem-connector and txns (as well as other rest-api operations over that connector)12:18
<pivotal-bot>Andrew Woods edited "Enable Version Deletion" https://www.pivotaltracker.com/story/show/6917756012:19
* github-ff joins12:34
[fcrepo4] lsitu opened pull request #280: Adding a "${jms.host}" variable to make jms more easily configurable as (master...feature/jms_configuration) http://git.io/qQbbxA
* github-ff leaves
<pivotal-bot>Longshou Situ added comment: "Sure. https://github.com/futures/fcrepo4/pull/280" https://www.pivotaltracker.com/story/show/6885335412:35
* github-ff joins
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/keRhiQ
fcrepo4/master 23eb685 lsitu: Adding a "${jms.host}" variable to make jms more easily configurable as...
fcrepo4/master 94cb07d Andrew Woods: Merge pull request #280 from lsitu/feature/jms_configuration...
* github-ff leaves
* Longshou leaves12:41
* scossu leaves12:58
* github-ff joins13:01
[fcrepo4] escowles pushed 1 new commit to projection-tests: http://git.io/9dSxhQ
fcrepo4/projection-tests f306d90 Esmé Cowles: Re-enabling auto-versioning test with update to undo changes after the test is done
* github-ff leaves
* fcrepo-bot joins13:13
* martinjd leaves13:15
* github-ff joins13:18
[fcrepo4] escowles pushed 1 new commit to projection-tests: http://git.io/Kd2pNA
fcrepo4/projection-tests 0df0b84 Esmé Cowles: Removing unneeded import and logging
* github-ff leaves
* github-ff joins13:20
[fcrepo4] escowles opened pull request #281: Versioning tests now undo changes to fedora:resource CND (master...projection-tests) http://git.io/pV6M8g
* github-ff leaves
<pivotal-bot>Esme Cowles added comment: "I've updated the FedoraVersions IT to undo the CND changes it makes, and re-enabled the filesystem federatio..." https://www.pivotaltracker.com/story/show/6868919813:21
Esme Cowles finished "Fix Projection Integration Tests" https://www.pivotaltracker.com/story/show/68689198
* scossu joins13:28
* gregjansen joins13:31
* travis-ci joins13:35
[travis-ci] futures/fcrepo4#1700 (projection-tests - f306d90 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/58865d9ec56b...f306d908fe4d
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22710140
* travis-ci leaves
* longshou joins
* scossu leaves13:37
* travis-ci joins13:46
[travis-ci] futures/fcrepo4#1701 (projection-tests - 0df0b84 : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/f306d908fe4d...0df0b849f070
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22711199
* travis-ci leaves
* fcrepo-bot leaves14:20
<pivotal-bot>Esme Cowles added comment: "This PR worked for me on my local machine. I had a few very minor comments, but otherwise this PR looks goo..." https://www.pivotaltracker.com/story/show/6088334614:31
* github-ff joins14:32
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/Qp9jzA
fcrepo4/master 789823d Esmé Cowles: Re-enabling filesystem federation copy tests...
* github-ff leaves
<pivotal-bot>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo4/commit/789823db687608b187d848374bee3786eaff7268" https://www.pivotaltracker.com/story/show/68689198
Andrew Woods delivered "Fix Projection Integration Tests" https://www.pivotaltracker.com/story/show/68689198
* github-ff joins14:33
[fcrepo4] awoods closed pull request #281: Versioning tests now undo changes to fedora:resource CND (master...projection-tests) http://git.io/pV6M8g
* github-ff leaves
* scossu joins14:45
* travis-ci joins15:05
[travis-ci] futures/fcrepo4#1703 (master - 789823d : Esmé Cowles): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/94cb07d332c2...789823db6876
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22716628
* travis-ci leaves
<pivotal-bot>Esme Cowles started "Enable Version Deletion" https://www.pivotaltracker.com/story/show/6917756015:06
<mikeAtUVa>Esme: as a little background... the first iteration of versioning support didn't have reasonable identifiers for versions created without an explicit label provided. When we improved the version listing page to keep people out of the JCR system space we "assigned" identifiers to them that *happen* to be the UUID of the frozen node. They may not always be that UUID and we don't want users to assume they are. This is a start in t15:12
he (now daunting) task of hiding the leakage of JCR concepts into Fedora 4.
escowles: also, thanks for the helpful review comments on the PR:)
<escowles>mikeAtUVa: that makes sense -- if we decide we need to use something other than the node UUID as the version identifier, we can switch to that later15:15
* martinjd joins15:18
<awoods>ksclarke: Do you have AWS access through UCLA? or is that a personal instance?15:30
<ksclarke>yes, through UCLA now... very nice
(though I also have personal instances for personal projects)15:32
<awoods>ksclarke, that is nice. escowles, does UCSD have an AWS relationship?15:33
<escowles>awoods: not that i know of -- i don't remember anyone at UCSD using AWS15:34
<awoods>escowles: It seems like a significant step for state university systems to support the use of AWS. I assume if UCLA has it, an agreement exists at the state level.15:35
<ksclarke>it's very new for us; it's only been "turned on" since Mar 12
and from what I hear there was a bit of legal wrangling to make it acceptable15:36
<escowles>awoods: that would make sense -- it's entirely possible that i just haven't heard about it (though the UC campuses often do things on their own without talking to the others)
e.g., UCSD is doing Hydra, UCLA is doing Islandora, and UCOP has Merritt
* ksclarke nods15:37
<pivotal-bot>Longshou Situ added comment: "I see the "jmsBroker" dependency (https://github.com/futures/fcrepo4/blob/master/fcrepo-webapp/src/main/re..." https://www.pivotaltracker.com/story/show/6885335415:52
* scossu leaves
<pivotal-bot>Andrew Woods added "Unversioned Node Exception" https://www.pivotaltracker.com/story/show/6929655815:59
<awoods>mikeAtUVa: say the word when you are done with: https://www.pivotaltracker.com/story/show/6088334616:00
<pivotal-bot>feature: Version Management (finished) / owner: Mike Durbin
<awoods>...it looks like you might be.
<mikeAtUVa>awoods: yeah, I pushed a commit to address all the coments.16:01
<awoods>mikeAtUVa: looks good to me. escowles?16:02
<escowles>yep, it looks good to me too
<pivotal-bot>Andrew Woods edited "Configure indexer webapp to run fcrepo4 for integration tests" https://www.pivotaltracker.com/story/show/6919228016:06
Andrew Woods edited "Configure indexer webapp to run fcrepo4 for integration tests" https://www.pivotaltracker.com/story/show/69192280
Andrew Woods edited "Configure indexer webapp to run fcrepo4 for integration tests" https://www.pivotaltracker.com/story/show/69192280
Andrew Woods accepted "Fix Projection Integration Tests" https://www.pivotaltracker.com/story/show/6868919816:12
<awoods>longshou: What issue are you having with jms-broker configuration on non-localhosts, exactly?16:16
* scossu joins16:22
<longshou>I didn't see the records ingested showing up in fuseki so I was thinking about the JMS indexer may not be able to receive the events from that remote fcrepo instance.
<awoods>longshou: could it be a firewall issue? Do you have confidence that the jms connection is not being made?16:24
longshou: does it make sense to update your /etc/hosts to define two hostnames for localhost to enable you to test the setup on a single machine?16:25
<pivotal-bot>Andrew Woods edited "A link to the endpoint fcr:versions should not be presented in the UI for unversionable nodes since it will always ..." https://www.pivotaltracker.com/story/show/6902691416:28
* scossu leaves16:30
<longshou>I'll discuss it with Esme regarding the update for /etc/hosts later. Esme just mentioned that the index type triple need to be added to enable fuseki indexing. I'll try it to see whether it works for both VMs or not.16:31
<awoods>longshou: and let me know if it becomes too frustrating.16:34
* ksclarke1 joins16:35
<longshou>Not at all. It's interesting for me to know how the system works. Thanks.
<awoods>longshou: or how it doesn't work ;)16:37
* ksclarke1 leaves
* github-ff joins16:39
[fcrepo4] awoods pushed 1 new commit to master: http://git.io/c4nBCQ
fcrepo4/master 7df0138 Michael Durbin: Add support for version reversion...
* github-ff leaves
<pivotal-bot>Andrew Woods added comment: "Resolved with: https://github.com/futures/fcrepo4/commit/7df01382197478dd31c07fe70cb8c8e9ebae577b" https://www.pivotaltracker.com/story/show/6088334616:40
* github-ff joins16:41
[fcrepo4] awoods closed pull request #279: Added support for version reversion. (master...revert-version) http://git.io/F0uTSw
* github-ff leaves
<pivotal-bot>Andrew Woods delivered "Version Management" https://www.pivotaltracker.com/story/show/6088334616:42
* scossu joins16:45
* github-ff joins16:51
[fcrepo4] awoods force-pushed master from 83ff331 to 7df0138: http://git.io/0BHdXg
* github-ff leaves
* mikeAtUVa leaves16:56
* travis-ci joins
[travis-ci] futures/fcrepo4#1707 (master - 83ff331 : Andrew Woods): The build has errored.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/7df013821974...83ff3311ebd4
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22724809
* travis-ci leaves
* travis-ci joins17:12
[travis-ci] futures/fcrepo4#1706 (master - 7df0138 : Michael Durbin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/789823db6876...7df013821974
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22724746
* travis-ci leaves
<pivotal-bot>Andrew Woods started "Configure indexer webapp to run fcrepo4 for integration tests" https://www.pivotaltracker.com/story/show/6919228017:15
Andrew Woods accepted "Version Management" https://www.pivotaltracker.com/story/show/6088334617:16
* ksclarke1 joins17:24
* ksclarke1 leaves17:25
* travis-ci joins17:26
[travis-ci] futures/fcrepo4#1708 (master - 7df0138 : Michael Durbin): The build passed.
[travis-ci] Change view : https://github.com/futures/fcrepo4/compare/83ff3311ebd4...7df013821974
[travis-ci] Build details : http://travis-ci.org/futures/fcrepo4/builds/22725424
* travis-ci leaves
<pivotal-bot>Longshou Situ added comment: "To support the messages produced by two or more Fedora installations consumed by a single fcrepo-jms-index..." https://www.pivotaltracker.com/story/show/6885335417:33
* fasseg leaves17:46
* longshou leaves17:47
<pivotal-bot>Andrew Woods added comment: "Option #1 (2 fedoras posting to one queue) should be possible with: ""17:50
https://github.com/futures/fcrepo4/pull..." https://www.pivotaltracker.com/story/show/68853354
<awoods>ksclarke: the y-axis on your graphs is milliseconds?17:52
<ksclarke>awoods: yes17:56
* ksclarke goes to figure out how to tell gnuplot to output that17:57
<awoods>ksclarke: Good luck... or you can just note it.
<pivotal-bot>Longshou Situ added comment: "No. I think the "jmsBroker" dependency (https://github.com/futures/fcrepo4/blob/master/fcrepo-webapp/src/m..." https://www.pivotaltracker.com/story/show/6885335417:58
<awoods>longshou: Are you saying that the jms.host configuration via a SystemProperty passed into fcrepo4 does not work?18:10
* martinjd leaves18:53
<escowles>awoods: i think the issue is that the ticket says to use a single indexer: https://www.pivotaltracker.com/story/show/6885335419:14
<pivotal-bot>feature: Multiple Fedoras Feeding Single Triplestore (started) / owner: Longshou Situ
* scossu leaves
<escowles>but the use case doesn't say anything about the indexer: https://wiki.duraspace.org/display/FF/Use+case+%5Bstub%5D%3A+External+triple+store+spanning+several+repositories19:15
and i thought you said it would be OK to use multiple indexers if that's what we needed to do to get it working:
<awoods>escowles: yes, two or more indexers are fine
<escowles>hmmm.... looks like longshou's not here19:16
awoods: i just emailed longshou about it, so i think that was the main sticking point for getting that working19:18
<awoods>escowles: thanks for the help.
* longshou joins19:21
* longshou leaves20:53
* scossu joins22:44
* scossu joins23:18

Generated by Sualtam