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

Using timezone: Eastern Standard Time
* awead joins08:04
* dhlamb joins08:33
* dhlamb leaves08:34
* dhlamb joins08:37
* peichman joins08:58
* awoods joins09:24
* github-ff joins09:51
[fcrepo-webapp-plus] awoods pushed 2 new commits to master: http://git.io/vWoMo
fcrepo-webapp-plus/master 9d7b166 nruest: Add maintainer section. Address FCREPO-1745.
fcrepo-webapp-plus/master 72deea2 Andrew Woods: Merge pull request #26 from ruebot/FCREPO-1745...
* github-ff leaves
* ajs6f joins09:52
* github-ff joins
[fcrepo-karaf] awoods pushed 2 new commits to master: http://git.io/vWoMH
fcrepo-karaf/master 02708b7 Aaron Coburn: add maintainer section
fcrepo-karaf/master 12d0deb Andrew Woods: Merge pull request #13 from acoburn/add_maintainers...
* github-ff leaves
* travis-ci joins09:55
* travis-ci leaves
* acoburn joins09:56
* travis-ci joins09:57
* travis-ci leaves
* peichman1 joins09:58
* peichman leaves
<peichman1>@acoburn: Is the WebAC sprint meeting at 11 today?09:59
<acoburn>peichman1: yes it is10:00
<peichman1>@acoburn: same hangout as last time?10:01
<acoburn>peichman1: yes
<peichman1>@acoburn: cool, see you guys then
<acoburn>peichman1: excellent, see you then10:02
* peichman1 leaves10:03
* peichman joins
* umgrosscol joins10:12
* dwilcox joins10:15
* whikloj joins10:37
* github-ff joins10:41
[fcrepo4] nianma opened pull request #920: change strong ETags to weak ETags (master...fcrepo-1754) http://git.io/vWKvH
* github-ff leaves
* dredmon joins10:42
Do folks still use this channel or is it better to use the mailing list for questions?10:44
<awoods>dredmon: This channel is a good starting point if you want something quick. The mailing list obviously provides access to a much larger audience.10:45
<dredmon>Fair enough - looking for something that I hope will be relatively quick....10:46
I've been combing through the documentation looking for an answer to this, but couldn't seem to find it: Does anyone know of a way to rebuild the infinispan cache?10:47
<awoods>dredmon: good question. The only way I know of is to iterate over your repository content.10:48
* peichman leaves
* peichman joins10:50
* dwilcox leaves
<dredmon>How do you do that? By reingesting?
<awoods>dredmon: no, by performing HTTP GET requests on your resources.10:51
* github-ff joins10:54
[fcrepo4] nianma opened pull request #921: add prefix check before update properties (master...fcrepo-1766) http://git.io/vWKLO
* github-ff leaves
<awoods>dredmon: what are you actually trying to accomplish?
<dredmon>well, let me back up....
The cache got corrupted due to that disk filling up and now I'm getting errors when I try to restart fedora10:57
The binaries are stored in another location and appear to be intact
I've fixed the space on my hypervisor, but can't get Fedora to restart, the issue appears to be due to issues with the infinispan cache, I can DM more information from the logs if desired11:00
<awoods>on a call11:03
* chadmills joins11:05
* github-ff joins11:18
[fcrepo-webapp-plus] yinlinchen opened pull request #27: fcrepo-webapp-plus: default build - webac and audit (master...fcrepo-1773) http://git.io/vWKnK
* github-ff leaves
* Nianli joins11:24
* awead leaves11:32
* chadmills leaves11:39
* awead joins11:44
<awoods>dredmon: you should join a call we are having today at 3pm ET on backup strategies, on google-hangout: https://plus.google.com/hangouts/_/event/c1glu6soq43r1rr6ou17qtobug811:51
<dredmon>Okay, I'll try to participate - I have a few calls this afternoon and if they don't run over, I may be able to participate
* jrgriffiniii leaves11:55
<dredmon>awoods: Are you still on a call?11:57
* dwilcox joins12:00
* jrgriffiniii joins12:02
<awoods>dredmon: no, but planning for my next call12:05
<dredmon>Okay
<awoods>dredmon: There are others on this channel... no need to wait for me.12:12
<dredmon>yes, if anyone else knows - I would love other input!12:13
When I attempt to start fedora, I get Corruption: checksum mismatch for leveldb12:14
<awoods>whikloj: can you please resolve the review of this ticket: https://jira.duraspace.org/browse/FCREPO-175512:15
dredmon: can you please post a gist of your error output?
<dredmon>from catalina.out?
<awoods>dredmon: probably12:16
dredmon: have you seen this: https://wiki.duraspace.org/display/FEDORA4x/How+to+inspect+LevelDB
<dredmon>No, I haven't seen that yet - I'll look through it. Thanks!12:17
<whikloj>awoods: I'll try, this is the client one that has no documentation. So I'll have to read the code to learn how to use it first.
<awoods>whikloj: or you can get in touch with mikeAtUVa or escowles.12:18
whikloj: the PR is trivial: https://github.com/fcrepo4-labs/fcrepo4-client/pull/33/files12:19
* escowles joins13:00
<awoods>https://wiki.duraspace.org/display/FF/2015-10-26+Performance+-+Scale+Meeting13:03
<ajwagner>Is the call underway? I used the web dialer and just have hold music.13:07
<ruebot>it is underway
<ajwagner>Hrm.13:08
Okay, I can't reach the number from my land line. I'll try with Skype like last time I guess.
<awoods>ajwagner: (712) 775-703513:11
Participant Code: 479307#
<ajwagner>@awoods when I dial that from my land line I get "We're sorry the number you have dialed can not be reached from your line. 416 19" Anyway, I'll try to get it sorted out fro the next one.13:13
<awoods>ajwagner: https://wiki.duraspace.org/display/FF/Conference+Call+Information13:14
* dwilcox leaves13:20
* dwilcox joins13:23
<dhlamb>yes! that's definitely the time to do it! do you have checksums from the initial deposit, though?13:27
<escowles>dhlamb: or you could load data first and do a second pass to verify checksums later... if you wanted to get stuff in as fast as possible13:28
<dhlamb>escowles: indeed
* awead_ joins13:29
* awead leaves
* dwilcox leaves13:31
<dhlamb>it's not stupid. things get corrupted all the time13:33
just general entropy
don't feel bad for wanting fixity :D
* jrgriffiniii leaves13:39
<ruebot>dhlamb++13:44
<dhlamb>representative testing data set is a great start13:47
* jrgriffiniii joins13:49
* jgpawletko joins13:56
* dwilcox joins
<ruebot>sharing++13:57
...and now i have to disappear to another meeting13:59
later y'all
* peichman leaves14:03
* dhlamb leaves14:10
* jrgriffiniii leaves14:17
* dhlamb joins14:22
awoods: escowles: creating internal ticket at dgi to loop in other developers / managers w/r/t creating a reasonable first testing plan14:24
* jrgriffiniii joins14:25
<escowles>dhlamb++
<awoods>dhlamb: thanks14:26
* dwilcox leaves14:48
* dwilcox joins14:54
* dwilcox leaves14:58
* mohamedar joins15:03
* github-ff joins15:11
[fcrepo4] acoburn opened pull request #922: update max metaspace size (master...fcrepo-1796) http://git.io/vW65X
* github-ff leaves
* jrgriffiniii leaves15:12
* jrgriffiniii joins15:36
* jgpawletko leaves15:39
* dwilcox joins15:55
* dredmon leaves15:56
<awoods>All: is anyone running Redhat who is available for a quick test?15:57
<escowles>awoods: btw, my clean build of master fcrepo4 succeeded16:00
<awoods>escowles: redhat?
<escowles>awoods: no, macosx
<ajs6f>awoods: Not on my workstation, but if no one else volunteers, I can go to a server that is RH. I mean, I won't do it pleasantly or cheerfully, I'll do it grumbling and with a sneer, but I will do it. It will take some time to download build tools, too. Basically, you really don't want to rely on my help, because I am a thin reed to lean on.
<awoods>escowles: thanks
ajs6f: could you pop your head into mikeAtUVa's office to ask for his help, on my behalf. I believe he runs redhat.16:01
<ajs6f>awoods: no, I'm at home with a thrown-out back.
<awoods>ajs6f: it is nice of you to keep the back company
<ajs6f>awoodsL Oh, I took it out and slipped it into a tumbler of bleach and aspirin to unwind.16:02
<awoods>ajs6f: the issue can be seen in the Jenkins log: http://jenkins.fcrepo.org/job/fcrepo4/org.fcrepo$fcrepo-http-api/3120/consoleFull
searching for "passivation"
<ajs6f>awoods: WTF? Is that a problem in our XML?16:03
awoods: Or do you suspect a problem in the XML proc libraries?
<awoods>ajs6f: it relates to the recent upgrade of Modeshape/ISPN... new ispn schema
ajs6f: but for some reason, it looks like an old schema parsing library is trying to read the new schema16:04
<ajs6f>awoods: Oh. But it's behaving differently on Mac and RH?
<awoods>ajs6f: I am trying to find out
ajs6f: I believe acoburn said he sees the same issue
<ajs6f>awoods; Right. Could be differetn JVMs, too, eh?
<awoods>ajs6f: but acoburn is likely swamped
ajs6f: yes, the issue could be coming from any number of differences16:05
ajs6f: at this point, I am looking for someone else who sees the same issue
ajs6f: redhat is the first attempt
<ajs6f>awoods: Why not just ignore it and hope that no user ever sees it?
<acoburn>awoods: ajs6f: it's easy for me to test this on centos/ubuntu
<ajs6f>awoods: If you want, I can try and use that RH instance I mentioned.
<awoods>acoburn: would you mind giving master an mvn clean install spin?16:06
<acoburn>awoods: ajs6f: I don't see *this* issue on macosx — only the MetaSpace issue
awoods: for centos? or ubuntu? or both?
<awoods>acoburn: jenkins is redhat
<acoburn>awoods: ok, centos is closest
<ajs6f>centos < rh. A problem that would show up with rh is likely to shop up with centos.16:07
<awoods>acoburn: fyi, I wiped out .m2 and the work directory on jenkins... still seeing the issue
* peichman joins16:08
<escowles>awoods: just ran the build on ubuntu 14.04.3 and got the passivation attribute error16:10
* dhlamb leaves
<ajs6f>escowles: You must be using ACTIVEFedora. See what I did there?
* bseeger joins16:11
* escowles groans
<peichman>what is the best way to propagate changes from a locally built fcrepo4 into the fcrepo-webapp-plus?
context: I am making some changes to the ServletContainerAuthenticationProvider and I want to build the webapp with webac and then deploy16:12
<escowles>peichman: you should be able to stop the locall-built webapp, and run the fcrepo-webapp-plus version with the same data directory
<peichman>@escowles: is fcrepo-webapp-plus supposed pick up the changes from the local fcrepo4? I tried running mvn install on fcrepo4, then building the fcrepo-weabapp-plus, but AFAICT, the fcrepo-webapp-plus is still using the libraries from stock 4.4.1, without my local changes16:18
(I am still a bit mystified by the build process)
* jgpawletko joins16:19
<escowles>peichman: assuming both projects have the same version (currently 4.4.1-SNAPSHOT), then fcrepo-webapp-plus should pick up your locally-built fcrepo4 changes16:20
<acoburn>peichman: when you build fcrepo, do you run mvn install? that's the only way to have your local changes make it into webapp-plus16:22
<peichman>@acoburn: I did, but I think I may have deployed it incorrectly into the vagrant16:23
<escowles>peichman: the vagrant build scripts will pull a pre-built war file from github. but you can take your locally-built fcrepo-webapp-plus version and deploy it in the vagrant after it's setup16:24
* ajs6f leaves16:25
<peichman>@escowles: I did that, but I don't see my changes (in particular, there is a info level log message I added to the ServletContainerAuthenitcationProvider::authenticate method that is not showing up)
<escowles>peichman: did you look at the tomcat logs (in /var/log/tomcat7) to make sure your war file deployed successfully?16:26
<peichman>@escowles: everything looks fine in the catalina log, including other info messages
let me see if I have this process straight:16:27
1. cd fcrepo4 && mvn install16:28
2. cd fcrepo-webapp-plus && mvn install -Paudit,webac
<acoburn>awoods: build success on centos
<peichman>3. cp fcrepo-webapp-plus/target/*.war to the vagrant
4. replace the exisitng fcrepo.war on the vagrant with the one I just built and copied16:29
* Nianli leaves
<peichman>4. restart Tomcat
<escowles>peichman: you shouldn't need to restart tomcat -- it should notice the updated file and redeploy it automatically
<peichman>okay
<escowles>and restarting tomcat might prevent it from doing that and so it might still be using the old version of fcrepo416:30
<peichman>gonna retry steps 1-4 explicitly, to make sure I didn't miss anything
<escowles>and steps 1-4 are correct
* osmandin leaves
* jrgriffiniii leaves16:31
* jrgriffiniii joins16:39
<awoods>escowles/acoburn: thanks for running the builds16:40
escowles: what is your java -version?
escowles: and your "mvn -version"16:41
<peichman>okay, that worked; I realized in my prior attempts I had fat-fingered the path to the webapps directory, so it wasn't really replacing the old fcrepo.war
<escowles>peichman: excellent -- glad you got it figured out
<awoods>peichman: not replacing fcrepo.war makes it harder to get an updated version.
<escowles>awoods: java version "1.8.0_60" (1.8.0_60-b27)
awoods: mvn version "3.0.5"16:42
<acoburn>awoods: just ran this on ubuntu (for good measure) and encountered no errors16:45
awoods: java version "1.8.0_66"
<awoods>escowles/acoburn: hmm, pondering the commonalities...
escowles/acoburn: Jenkins has:
mvn -version16:46
Apache Maven 3.0.5 (r01de14724cdef164cd33c7c8c2fe155faf9602da; 2013-02-19 08:51:28-0500)
Maven home: /opt/maven3/apache-maven-3.0.5
Java version: 1.8.0_25, vendor: Oracle Corporation
Java home: /usr/lib/jvm/jdk1.8.0_25/jre
Default locale: en_US, platform encoding: UTF-8
OS name: "linux", version: "3.19.0-30-generic", arch: "amd64", family: "unix"
...failing
<acoburn>awoods: the centos machine uses java version1.8.0_65
awoods: and it uses maven 3.3.316:47
<awoods>acoburn: locally I run:
Maven home: /opt/maven3/apache-maven-3.0.5
Java version: 1.8.0_25, vendor: Oracle Corporation
Java home: /usr/lib/jvm/jdk1.8.0_25/jre
<acoburn>awoods: it seems from the error it's using a cached version of the ISPN config
<awoods>acoburn: indeed... but where is it coming from?16:48
<acoburn>awoods: nevermind, I'm thinking of something else
* peichman leaves
<awoods>acoburn: rather, a cached version of the libraries used to parse the new version of the ISPN config
<acoburn>awoods: are you setting anything special in JAVA_OPTS?16:50
<awoods>acoburn: I am trying a build on the jenkins machine now, but with my own user...16:51
acoburn: it looks like it may have gotten past the fcrepo-http-api issue
<acoburn>awoods: that seems sensible
<awoods>acoburn: which would indicate it has something to do with the jenkins config
<acoburn>awoods: on my builds, the machines were completely new (fresh installation of maven, java, etc)16:52
* jgpawletko leaves16:55
<acoburn>awoods: if you're changing the jenkins configs, you can remove the MaxPermSize and UseSplitVerifier params16:59
<awoods>acoburn: yes... although I will minimize the changes during this debugging first.
<acoburn>awoods: that makes sense; it's just smth I noticed when looking through the logs17:00
<awoods>acoburn: thanks
* umgrosscol leaves17:03
* escowles leaves
* bseeger1 joins17:04
* bseeger leaves
<whikloj>acoburn/awoods: Is there a way to dynamically add to the resource headers from an optional extension like webac?17:05
<acoburn>whikloj: you mean in order to produce a Link header in the response?17:06
<whikloj>acoburn: yes
<acoburn>whikloj: I think this will involve a change in fcrepo-core (probably the fcrepo-http-api module)17:07
<whikloj>acoburn: I like your getEffectiveAcl as it does all the work...but I need to get the response HTTP level.
<acoburn>whikloj: awoods can correct me if I'm wrong
<awoods>whikloj: are you working the link-header issue?
<whikloj>awoods: yes
<awoods>whikloj: I thought that was a "nice to have"17:08
<whikloj>awoods: It is but I essential to the Solr authentication issue
^^ but is essential
<awoods>whikloj: right
<whikloj>Can I optionally include code from an extension? Or check if a module is installed?17:09
<acoburn>whikloj: this seems like a case for dependency injection, no?17:10
<whikloj>acoburn: sure....I'll go look up how to do that now :)
acoburn: do you think stripping this out to a utility class is good, bad or ugly? https://github.com/fcrepo4/fcrepo-module-auth-webac/blob/master/src/main/java/org/fcrepo/auth/webac/WebACRolesProvider.java#L266
<acoburn>whikloj: why? do you need access to it from another module?17:11
<awoods>whikloj: probably not very cleanly. You may need to create a JAX-RS MessageBodyWriter.
whikloj: a MessageBodyWriter would be injected into to request/response flow by Jersey17:12
<whikloj>acoburn: It does traversal to get the Acl, which is what I need to add to the Link Header
<acoburn>whikloj: I'd follow awoods' suggestion about writing to the JAX-RS response17:13
<whikloj>awoods: MessageBodyWriter, so I would add this to WebAC and if enabled it could inject itself into the flow? That seems nice.
<awoods>whikloj: indeed
* acoburn leaves17:19
* bseeger1 leaves
* f4jenkins leaves17:27
* f4jenkins joins17:28
* dwilcox leaves17:36
<awoods>whikloj: here is one example: https://github.com/fcrepo4/fcrepo4/blob/master/fcrepo-http-commons/src/main/java/org/fcrepo/http/commons/responses/RdfStreamProvider.java#L6217:38
* jrgriffiniii leaves17:39
<whikloj>awoods: So this would replace an existing MessageBodyWriter, correct?17:42
<awoods>whikloj: I don't think so... I think it would just be chained.17:43
<whikloj>awoods: ok, I'll give it a try and see what blows up.
<awoods>whikloj: but you may need to do some testing
<whikloj>awoods++
<awoods>whikloj++
* mohamedar leaves17:47
* dwilcox joins17:48
* whikloj leaves17:59
* jrgriffiniii joins18:01
* dwilcox_ joins
* dwilcox leaves
* chadmills joins18:02
* chadmills leaves18:37
* dwilcox_ leaves18:47
* jrgriffiniii leaves18:53
* jrgriffiniii joins19:06
* jgpawletko joins19:10
* dwilcox joins19:12
<jrgriffiniii>For any currently attending the DLF, are there any fcrepo-related dining or events planned for this evening?19:13
* dwilcox leaves19:48
* jrgriffiniii leaves19:53
* dwilcox joins20:21
* the_mgt_ joins20:25
* dwilcox leaves20:26
* mohamedar joins20:28
* the_mgt leaves
* mohamedar leaves20:33
* jrgriffiniii joins21:32
* dwilcox joins21:58
* jgpawletko leaves22:00
* dwilcox leaves22:02
* github-ff joins22:57
[fcrepo4-client] whikloj pushed 2 new commits to master: http://git.io/vWPjI
fcrepo4-client/master 7555313 Andrew Woods: Update create-version response code...
fcrepo4-client/master 3928cec Jared Whiklo: Merge pull request #33 from awoods/fcrepo-1755...
* github-ff leaves
<f4jenkins>Yippee, build fixed!23:01
Project migration-utils build #255: FIXED in 1 min 55 sec: http://jenkins.fcrepo.org/job/migration-utils/255/
* travis-ci joins23:02
fcrepo4-labs/fcrepo4-client#90 (master - 3928cec : Jared Whiklo): The build passed.
Change view : https://github.com/fcrepo4-labs/fcrepo4-client/compare/a3ef377a74ea...3928cec7e461
Build details : https://travis-ci.org/fcrepo4-labs/fcrepo4-client/builds/87590945
* travis-ci leaves
* github-ff joins23:05
[fcrepo4] awoods opened pull request #923: Remove unused ISPN configuration file and clustering properties (master...fcrepo-1739) http://git.io/vWXeV
* github-ff leaves

Generated by Sualtam