IQSS logo

IRC log for #dataverse, 2018-05-02

Connect via chat.dataverse.org to discuss Dataverse (dataverse.org, an open source web application for sharing, citing, analyzing, and preserving research data) with users and developers.

| Channels | #dataverse index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary

All times shown according to UTC.

Time S Nick Message
00:04 dataverse-user joined #dataverse
06:46 jri joined #dataverse
08:24 icarito[m] joined #dataverse
13:48 pameyer joined #dataverse
13:59 pameyer joined #dataverse
14:12 pameyer1 joined #dataverse
14:32 donsizemoer joined #dataverse
14:32 donsizemore joined #dataverse
15:25 pameyer joined #dataverse
15:28 pameyer joined #dataverse
15:29 pameyer joined #dataverse
15:30 pameyer joined #dataverse
15:31 pameyer joined #dataverse
15:31 pameyer joined #dataverse
15:32 pameyer joined #dataverse
15:37 pameyer joined #dataverse
16:39 jri joined #dataverse
17:06 pameyer joined #dataverse
17:45 pameyer joined #dataverse
18:03 pdurbin donsizemore: hi! Can I pick your brain about Data Explorer installation?
18:13 donsizemore @pdurbin you might get more if you picked my nose... but how may i help? ;)
18:34 donsizemore @pdurbin also, I promise the LimitNOFILE line at least makes it from systemd into /proc:
18:34 donsizemore Max open files            32768                32768                files
18:38 pdurbin heh, I believe you. off to QA! I'm wondering if these instructions are sufficient because they don't mention the "prep" file stuff: https://github.com/scholarsportal/Dataverse-Data-Explorer#installation
18:39 pdurbin donsizemore: you already had TwoRavens installed so you did all of setup to generate "prep" files: http://guides.dataverse.org/en/4.8.6/installation/r-rapache-tworavens.html#d-version-conflict-check-preprocess-r
18:39 pameyer joined #dataverse
18:39 pdurbin but on a brand new installation of Dataverse, this hasn't been done yet
18:40 pdurbin so loading up Data Explorer as an external tool isn't enough, is it?
18:40 pdurbin you still have to do the setup for prep files, I assume
18:40 pdurbin you have to put the R code into place, etc
18:49 donsizemore @pdurbin we've called DDI Explorer on Nevermore from test Dataversia that didn't have TwoRavens installed (on the Dataverse side) though you're correct there's a pre-existing TwoRavens installation on Nevermore. but doesn't the prep stuff all happen on the Dataverse side?
18:50 donsizemore @pdurbin I thought Dataverse required R for ingest(?)
18:57 donsizemore @pdurbin if you'd like once I get done here I can stand up a test to see what I can call from a clean dataverse
18:58 pameyer @donsizemore: you can definitely run Dataverse without R
18:58 pameyer not sure if that results in unexpected behavior when ingesting some filetypes though
19:01 donsizemore lemme grab a hot chocolate and i'll setup up a DDI explorer without the tworavens installation (and without R?)
19:33 pdurbin donsizemore: that would be great! Yeah, no R either please. No worries if this is too time consuming. I just don't want to oversell how easy it is to set up Data Explorer.
19:34 pdurbin And I'm thinking that probably the Data Explorer installation instructions should mention all the setup that's required.
19:34 donsizemore @pdurbin well, there's a schema.xml problem (I think) in Dataverse-Ansible which I'll track down next
19:35 donsizemore @pdurbin but I did a simple git-pull to a throwaway webserver to see if DDI Explorer is self-contained
19:35 pdurbin Hmm, ok. Did you see we bumped the Solr version again? Still 7.x.
19:39 donsizemore @pdurbin yes sir, i bumped it in my solr7 branch
19:40 donsizemore @pdurbin and DDI Explorer seems to work fine on a plain-old webserver (though I just checked and saw that R is installed) from a plain-old Dataverse (which also has R installed)
19:41 donsizemore @pdurbin but no apache modules, no hours of compiling libraries, just vanilla EPEL-provided R
19:42 pameyer joined #dataverse
19:43 pameyer @donsizemore: your mention of apache modules reminded me of an experiment I did with using nginx with dataverse
19:44 pameyer that works (although I'm pretty sure it would lose shib, so probably not a realistic option)
19:46 pameyer and if you're curious about http2 for dataverse, my nginx experiment didn't show a significant performance improvement
20:28 pameyer joined #dataverse
20:28 pdurbin donsizemore: just vanilla R. Ok. Thanks! Interesting.
21:15 jri joined #dataverse
21:58 donsizemore @pdurbin I thought Dataverse needed R to do some of the ingest functions but maybe it's all java-based. the DDI Explorer I texted with was straight from a 'git clone' and worked out of the box
21:58 donsizemore ^^tested
22:00 pdurbin donsizemore: I'm sure R is needed on some level to created an Rdata version of a tabular file. But maybe it just needs to communicate with Rserve on a separate box? I'm not sure. The prep file is created by running an R script I think. One sec.
22:01 pdurbin Yeah, the preprocess.R file mentioned at http://guides.dataverse.org/en/4.8.6/installation/r-rapache-tworavens.html#d-version-conflict-check-preprocess-r
22:05 pdurbin Anyway, thanks for poking around at this donsizemore

| Channels | #dataverse index | Today | | Search | Google Search | Plain-Text | plain, newest first | summary

Connect via chat.dataverse.org to discuss Dataverse (dataverse.org, an open source web application for sharing, citing, analyzing, and preserving research data) with users and developers.