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

Using timezone: Eastern Standard Time
* dwilcox joins00:32
* ksclarke leaves00:52
* dwilcox leaves01:24
* dwilcox joins01:51
* dwilcox leaves01:59
* dwilcox joins04:18
* awead_ joins07:09
* awead leaves07:12
* dwilcox leaves07:16
* acoburn joins08:46
awoods: what do you think about cutting a release of fcrepo-transform?08:47
* osmandin joins08:50
* ksclarke joins08:52
* chadmills joins09:10
<awoods>acoburn: We could. Is there a reason to put one out so soon after it was just released within fcrepo-webapp in 4.4.0?09:16
<acoburn>awoods: so that all of the ancillary projects don't have to depend on a SNAPSHOT version, that's the only reason
<awoods>acoburn: Do other projects besides fcrepo-webapp-plus depend on fcrepo-transform?09:18
* umgrosscol joins
<acoburn>awoods: fcrepo-camel, fcrepo-camel-toolbox
<awoods>acoburn: Is there a reason why those projects can not depend on the 4.4.0 version of fcrepo-transform?09:19
<acoburn>awoods: wasn't that what fcrepo-1670 was all about?09:20
awoods: getting these projects to *not* use the fcrepo-transform released in core?
awoods: there's no reason that such a release would imply the release of any other artifacts09:21
<awoods>acoburn: That is an interesting perspective. My understanding of fcrepo-1670 was a recognition that some pieces currently in core do not need to be there.09:22
acoburn: all of that said, it is not clear to me that there is a rush on getting an fcrepo-transform release out, but if you want to do it, that would be just fine.09:23
<acoburn>awoods: no, there's definitely no rush09:24
<awoods>acoburn: I will be on travel for the rest of the week starting tomorrow.
<acoburn>awoods: question about the modeshape upgrade09:41
awoods: I've merged the branch locally with master and I keep getting OOM errors when building it09:42
<awoods>acoburn: that is curious
<acoburn>awoods: it's always the -http-api integration test with versions09:43
<awoods>acoburn: what are your JAVA_OPTS or MAVEN_OPTS?
<acoburn>awoods: I've tried various things up to 4096mx and it makes not difference09:44
s/not/no/
awoods: I'm trying the modeshape-4.4.0 branch itself again09:45
<awoods>acoburn: I have not seen the OOM issue, but would be happy to try to replicate if you have a process for me.
<acoburn>awoods: that would be great. The process is this:09:46
awoods: checkout the modeshape-4.4.0 branch
awoods: rebase with master
awoods: squash commits
awoods: checkout master09:47
awoods: merge the modeshape-4.4.0 branch
awoods: mvn clean install
* ksclarke leaves
<awoods>acoburn: I assume squashing commits is not necessary?
<acoburn>awoods: presumably not
awoods: this is the error:09:49
WARN 09:49:16.128 (CommonAnnotationBeanPostProcessor) Invocation of destroy method failed on bean with name 'containerWrapper': java.lang.OutOfMemoryError: Metaspace
awoods: it's during container shutdown after the http-api integration tests09:50
awoods: I'm going to start over with a clean repo
<awoods>acoburn: Maybe it is as simple as updating our container management approach
<acoburn>awoods: is that a simple thing?09:52
<awoods>acoburn: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-kernel-modeshape/src/main/java/org/fcrepo/kernel/modeshape/spring/ModeShapeRepositoryFactoryBean.java09:53
acoburn: on a call
<acoburn>awoods: ok, I'm familiar with that class09:54
<awoods>acoburn:09:59
500 git checkout master
501 git pull
502 git checkout modeshape-4.4.0
503 git rebase master
504 mvn clean install
SUCCESS
<acoburn>awoods: and should I worry about this:10:02
Retrieving transaction manager Transaction: unknown
* ksclarke joins10:04
<acoburn>awoods: just tried again twice, and both succeeded10:12
* whikloj joins10:18
<acoburn>awoods: but, when I squash the commits and merge with master, I get the OOM error10:21
<ruebot>awoods: can you make me an owner of fcrepo4-exts so i can do the transfer?10:33
<awoods>ruebot: done10:36
acoburn: If you rebase on master, I am not sure what the merge is supposed to accomplish.10:38
<ruebot>awoods10:41
awoods: Transferred! https://github.com/fcrepo4-exts/fcrepo4-vagrant-base-box
<awoods>ruebot: great
ruebot: thanks10:42
<ruebot>awoods: updated ticket too. i think we're good to go on that one :-)
awoods: as for hashicorp account, is there a general fedora email address we should setup for that? i'm just using mine now.10:43
awoods: if that email address is setup with gravatar, even better
<awoods>ruebot: good question. What would an appropriate email be?10:50
<ruebot>awoods: for islandora, we community@islandora.ca -- do you have something like that, that you and dwilcox share?10:51
<awoods>ruebot: info@fedora-commons.org10:53
ruebot: if you feel like https://jira.duraspace.org/browse/FCREPO-1771 is ready for review... please change the ticket state.10:59
<ruebot>awoods: updated email address. i think you should get an email notification11:00
awoods: gotta present again right now. i'll pass you the creds in a private message if you're cool with that when i'm done
<awoods>ruebot: got the email...11:01
but I do not have the username/password
ruebot++
<acoburn>awoods: back11:12
awoods: I'm clearly not doing the merge properly
<awoods>acoburn: is the merge necessary?11:13
<acoburn>awoods: that's what I mean
awoods: should I just push to master from the modeshape-4.4.0 branch?11:14
awoods: after sqaushing the commits?
<awoods>acoburn: you also need to rebase on master
<acoburn>awoods: of course
<awoods>acoburn: of course
acoburn: no merge required11:15
acoburn: on another note, in order to move forward with: https://jira.duraspace.org/browse/FCREPO-1776 , would you mind creating a new ticket to address the GONE transaction / ClientErrorException issue?
<acoburn>awoods: sure11:16
<awoods>acoburn: perfect
<acoburn>awoods: I've generally been following this workflow for git: https://wiki.duraspace.org/display/FCREPO/Git+Guidelines+and+Best+Practices11:17
awoods: which involves merging the branch into master
<awoods>acoburn: that is a legacy wiki page... I should give it a look.11:19
acoburn: That is not the flow I use.11:20
<acoburn>awoods: seems that the preferred flow here would be "git push origin modeshape-4.4.0:master"11:21
<awoods>acoburn: agreed11:22
<acoburn>awoods: ok, I'm starting over with this flow:11:23
git clone https://github.com/fcrepo4/fcrepo4.git
git checkout modeshape-4.4.0
git rebase master
git rebase -i HEAD~4 # to squash the commits11:24
mvn clean install # sanity check
git push origin modeshape-4.4.0:master
<awoods>acoburn: great, as a side note...11:25
acoburn: my checkout command would have been: git checkout --track origin/modeshape-4.4.011:26
<acoburn>awoods: doesn't that accomplish the same thing?
awoods:11:27
git checkout modeshape-4.4.0
Branch modeshape-4.4.0 set up to track remote branch modeshape-4.4.0 from origin.
Switched to a new branch 'modeshape-4.4.0'
<awoods>acoburn: good question. I think the difference might be that "--track" create an upstream for the local branch.
acoburn: hmm11:28
acoburn: sounds similar
acoburn: anyways, you should be good to go
<acoburn>awoods: ok, thanks11:29
awoods: one question though
awoods: is this INFO line an issue? "Retrieving transaction manager Transaction: unknown"
awoods: does that mean that the tx manager is all in-memory?
<awoods>acoburn: in production it is an issue... that has been resolved by adding the necessary dependency11:30
<acoburn>awoods: arrgggggggg
awoods: another OOM
<awoods>acoburn: let me find the ticket...
acoburn: see: https://jira.duraspace.org/browse/FCREPO-129311:31
<acoburn>awoods: I think I am just jinxed with this merge. perhaps ruebot or whikloj might be able to look at it?11:52
https://github.com/fcrepo4/fcrepo4/pull/91011:53
<whikloj>acoburn: what do you need?
<acoburn>whikloj: I have tested and reviewed the PR above and am happy with it11:54
whikloj: however, when I squash the commits and run mvn clean install, I get OOM errors
<whikloj>acoburn: I'll give it a try
<acoburn>whikloj++11:55
whikloj: it's the modeshape-4.4.0 branch
<awoods>All: are you finding jenkins to be particularly slow to load? http://jenkins.fcrepo.org/12:20
<whikloj>awoods: nope12:21
<awoods>whikloj: hmm, a reload takes about a minute for me12:22
<whikloj>awoods: of the front page or a particular project?
<awoods>whikloj: just the link I mentioned12:23
<whikloj>awoods: nope, it is loading nice and quick for me
<awoods>whikloj: that is encouraging
whikloj: thanks
<whikloj>awoods: for me it is :)
awoods: np12:24
acoburn: ok, I pulled the modeshape branch, it built. Rebased on master (possibly not needed). Squashed the 4 commits to one and was able to build it.12:35
<acoburn>whikloj: that's great12:36
<whikloj>acoburn: so I have a modeshape branch that is nice. but I don't think I can push it anywhere12:37
<acoburn>whikloj: can't you push it to master?
awoods: does whikloj have write access to the fcrepo4 repository?
<whikloj>acoburn/awoods: I do12:38
<awoods>acoburn: I would think so...
whikloj: then why do you not think that you can push the branch?
<whikloj>awoods/acoburn: Consider me cautious, but to confirm. I am now pushing this merge to master?12:39
<acoburn>whikloj: that's my understanding: git push origin modeshape-4.4.0:master12:40
whikloj: or "upstream" if that's how you've set up your local repo
<awoods>whikloj: yes, this is just a normal codereview/push situation.
* chadmills leaves
<whikloj>awoods/acoburn: ok
* github-ff joins12:41
[fcrepo4] whikloj pushed 1 new commit to master: http://git.io/vWIgG
fcrepo4/master e0acd0c Andrew Woods: Update ModeShape -> 4.4.0, and ISPN -> 7.2.3...
* github-ff leaves
<acoburn>whikloj++
<whikloj>acoburn: and I close the PR with the commit, correct?12:42
<acoburn>whikloj: yes, please
whikloj: and update the ticket in jira: https://jira.duraspace.org/browse/FCREPO-158112:43
* github-ff joins
[fcrepo4] whikloj closed pull request #910: Modeshape 4.4.0 (master...modeshape-4.4.0) http://git.io/vnJuS
* github-ff leaves
<whikloj>acoburn: will do
* github-ff joins
[fcrepo4] awoods deleted modeshape-4.4.0 at c64f203: http://git.io/vWIg9
* github-ff leaves
* github-ff joins12:50
[fcrepo4-vagrant] awoods pushed 1 new commit to master: http://git.io/vWIV6
fcrepo4-vagrant/master 5e66c37 nruest: Use the fcrepo4-vagrant-base-box...
* github-ff leaves
* github-ff joins12:51
[fcrepo4-vagrant] awoods closed pull request #27: Use the fcrepo4-vagrant-base-box. Address FCREPO-1771. (master...FCREPO-1771) http://git.io/vCdGu
* github-ff leaves
<f4jenkins>Project fcrepo4 build #3114: UNSTABLE in 10 min: http://jenkins.fcrepo.org/job/fcrepo4/3114/12:52
jwhiklo: Update ModeShape -> 4.4.0, and ISPN -> 7.2.3
* travis-ci joins13:06
fcrepo4/fcrepo4#4094 (master - e0acd0c : Andrew Woods): The build has errored.
Change view : https://github.com/fcrepo4/fcrepo4/compare/989be975cce4...e0acd0c69229
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/86437727
* travis-ci leaves
<acoburn>awoods: seems like I wasn't the only one getting OOM errors :-(13:08
<awoods>acoburn: is jenkins resolving quickly for you?13:10
<acoburn>awoods: it's snappy
* chadmills joins
<awoods>I need to step away13:11
* awead_ joins13:19
<whikloj>acoburn: is the jenkins build matching you experience?13:21
* awead leaves
<acoburn>whikloj: travis-ci is, I'm not sure about jenkins
<whikloj>acoburn: perhaps the difference in the Readme recommended -Xmx settings and travis-ci's settings is an issue?13:25
<acoburn>whikloj: on my local machine I changed the memory settings to 4GB and it had no effect13:26
<whikloj>acoburn: that is an issue, but I would say 512m is still too low for travis13:28
<acoburn>whikloj: I agree13:30
whikloj: for the jenkins failures, those seem to be from an invalid infinispan configuration13:34
whikloj: I wonder if something isn't getting cleared before the build starts
<whikloj>acoburn: I am not well versed in jenkins, so I went to the console output13:37
<acoburn>whikloj: that's what I was looking at, too.
<whikloj>acoburn: yes, but then I just smile and nod until I feel like things will get better :)13:39
acoburn: also, Grizzly and I have agreed to give each other a wide berth13:40
acoburn: so I should get back to pax exam
* acoburn leaves14:17
* github-ff joins14:21
[fcrepo4] awoods pushed 2 new commits to master: http://git.io/vWLTQ
fcrepo4/master cf386a2 Aaron Coburn: remove jul-to-slf4j dependency; import * in osgi config
fcrepo4/master 19639e1 Andrew Woods: Merge pull request #918 from acoburn/fcrepo-1776...
* github-ff leaves
* travis-ci joins14:42
fcrepo4/fcrepo4#4096 (master - 19639e1 : Andrew Woods): The build passed.
Change view : https://github.com/fcrepo4/fcrepo4/compare/e0acd0c69229...19639e1998f3
Build details : https://travis-ci.org/fcrepo4/fcrepo4/builds/86458263
* travis-ci leaves
* github-ff joins15:16
[migration-utils] awoods created awoods-patch-1 (+1 new commit): http://git.io/vWLBb
migration-utils/awoods-patch-1 c86d401 Andrew Woods: Migrated creation date?...
* github-ff leaves
* travis-ci joins15:24
fcrepo4-exts/migration-utils#102 (awoods-patch-1 - c86d401 : Andrew Woods): The build failed.
Change view : https://github.com/fcrepo4-exts/migration-utils/commit/c86d401b475d
Build details : https://travis-ci.org/fcrepo4-exts/migration-utils/builds/86469308
* travis-ci leaves
* acoburn joins
* github-ff joins15:28
[migration-utils] awoods deleted awoods-patch-1 at c86d401: http://git.io/vWLg9
* github-ff leaves
* f4jenkins leaves15:33
* f4jenkins joins15:36
* dwilcox joins15:43
* github-ff joins15:51
[fcrepo4-client] awoods opened pull request #33: Update create-version response code (master...fcrepo-1755) http://git.io/vWLKO
* github-ff leaves
* awead joins15:55
* awead_away leaves15:57
* osmandin leaves16:36
* whikloj leaves16:59
* dwilcox leaves17:09
* dwilcox joins17:15
* chadmills leaves17:31
* acoburn leaves17:54
* umgrosscol leaves18:13
* ksclarke leaves18:20
* dwilcox leaves19:11
* dwilcox joins19:24
* the_mgt_ joins19:28
* the_mgt leaves19:31
* dwilcox leaves20:24
* ksclarke joins20:29
* dwilcox joins21:08
* dwilcox leaves21:25
* ksclarke leaves22:17
* awoods leaves22:21
* ksclarke joins22:32
* dwilcox joins23:12
* ksclarke leaves00:00
* dwilcox leaves00:11
* dwilcox joins00:13
* ksclarke joins00:20

Generated by Sualtam