Time
S
Nick
Message
05:34 poikilotherm joined #dataverse
06:48 juancorr joined #dataverse
07:04 poikilotherm joined #dataverse
07:05 jri joined #dataverse
07:52 icarito[m] joined #dataverse
07:53 candy` joined #dataverse
08:52 poikilotherm joined #dataverse
10:14 pdurbin joined #dataverse
10:20 pdurbin
Good morning, all.
10:29 poikilotherm
Good morning from Göttingen :-)
10:31 pdurbin
Hackathon? :)
10:32 poikilotherm
Dataverse DE meeting :-)
10:33 pdurbin
cool
10:36 poikilotherm
We are going to have a little Skype meeting with Slava later
10:36 poikilotherm
Wanna join
10:36 poikilotherm
?
10:39 pdurbin
Sure! When?
10:42 kiru joined #dataverse
10:58 poikilotherm
10am PST
10:58 poikilotherm
Ooops
10:58 poikilotherm
EDT
11:00 pdurbin
I think that will work for me. Do you want to post something to https://groups.google.com/forum/#!forum/dataverse-dev ?
11:22 poikilotherm
https://twitter.com/poi_ki_lo_therm/status/1135506939465654272
11:30 pdurbin
Heh. Nice shirt. Say hi to Jonas and Péter for me! I don't *think* I recognize anyone else. Looks fun!
11:32 pkiraly
Hi Philip!
11:32 poikilotherm
pkiraly could you send pdurbin the details how to join later?
11:35 pdurbin
pkiraly: hi!
11:36 pkiraly
@pdurbin: my Skype ID is kirunews. If you send yours, or ping me I'll call you at 16:00 CET (10:00 am in Cambridge, MA if I am correct)
11:36 pdurbin
pkiraly: ok, on Skype I'm "greptilian". :)
11:44 stefankasberger joined #dataverse
12:01 pdurbin
stefankasberger: hi! https://github.com/IQSS/dataverse/issues/5235 is in our sprint and I was thinking about picking it up and using pyDataverse to implement it. Or at least give pyDataverse a try. :)
12:06 stefankasberger
@pdurinb: i was thinking of the same for our internal test setup. to automatically populate dataverse with data for tests and so on.
12:07 stefankasberger
but you have to wait for release 0.2.0, which then has the needed functionality to use metadata with pyDataverse. will hopefully come until the end of the week. if not, i would say, during or after the community conference.
12:10 pdurbin
stefankasberger: ok, do you plan to create an issue at https://github.com/AUSSDA/pyDataverse/issues and assign it to a milestone?
12:20 jonas80 joined #dataverse
12:21 doigl joined #dataverse
12:22 eRATG joined #dataverse
12:22 aseeland joined #dataverse
12:22 poikilotherm
Greetings pdurbin :-)
12:22 poikilotherm
We are talking about setting up communication channels here in Göttingen
12:23 poikilotherm
Are there any plans for that GDCC chat thing?
12:23 poikilotherm
Do you know what happens "over there"?
12:24 lsteilen joined #dataverse
12:24 poikilotherm
jonas80 mentioned sth. that they might come up with a Slack instance?
12:24 jonas80
like using Slack, Mattermost or something similar?
12:25 rigelk
hi there! pdurbin do you know if anything came out of discussion about activitypub after the last community call? Or if anything ensued?
12:40 donsizemore joined #dataverse
12:41 donsizemore
ooh ooh @stefankasberger "to automatically populate dataverse with data for tests and so on"
12:44 pdurbin_m joined #dataverse
12:45 pdurbin_m
rigelk: yes, please see and give feedback on https://github.com/IQSS/dataverse/issues/5883
12:47 pdurbin_m
poikilotherm Jonas80 et al. I am strongly anti-Slack for open source. Have you read my recent manifesto on "Searchable Linkable Open Public Indexed (SLOPI) Communication"? :)
12:47 pdurbin_m
Please see and give feedback on https://github.com/good-labs/slopi-communication
12:48 pdurbin_m
pronounced "sloppy" :)
12:49 jonas80
Is there any tool that we could use right now? E.g. mattermost instead of slack?
12:51 donsizemore
@jonas80 i like this!
12:52 donsizemore
@jonas80 though their licensing isn't very FLOSS
12:52 pdurbin_m
jonas80: yes, someone could install RocketChat and configure it for SLOPI communication like Inkscape did: https://chat.inkscape.org
12:54 pdurbin_m
Does Mattermost support SLOPI communication?
12:54 jonas80
funny, i was just about to mention here that RocketChat came up in the discussion - and the question if it can be set up to enable slopi-style communication
12:54 jonas80
current state of the disussion here is that it is not
12:55 stefankasberger
@pdrubin: am not planning to go into project management activities with milesontes and issues for this release. i think, after 0.2.0 i will go more open with this, but right now i just want to release the stuff i need as fast as possible to do the migrations (and make people at AUSSDA happy). :)
12:57 stefankasberger
@donsizemore: was it a "danger" ooh ooh or a "awesome" ooh ooh?
12:57 donsizemore
@stefankasberger a wonderful ooh ooh. i wedged a kludgy ansible framework in place for this but would love a python module
12:58 pdurbin_m
stefankasberger: that's fine
12:59 pdurbin_m
donsizemore: I'd like to set up a Jenkins job for pyDataverse
12:59 pdurbin_m
jonas80: did you see my Inkscape link above? It's possible. :)
13:03 jonas80
sorry, that statement was about mattermost
13:04 lsteilen left #dataverse
13:06 pdurbin_m
jonas80: oh. I hope it's possible. Should I reach out to the Mattermost devs?
13:14 donsizemore
@pdurbin_m rocketchat licensing and features seem much more FLOSS than mattermost
13:15 pdurbin_m
Hmm. ok.
13:22 donsizemore
@stefankasberger are you cool with the jenkins job for pydataverse Phil wants? a push notification to https://jenkins.dataverse.org should do the trick
13:23 donsizemore
@pdurbin do we want a rocketchat or mattermost server, for funsies? can always tear them down
13:34 jonas80
we also talked about finding the resources to contribute a rocketchat server to community effort
13:35 jonas80
(like UNC's jenkins server)
13:35 pdurbin
donsizemore: as long as it's SLOPI, sure! :)
13:36 donsizemore
@jonas80 now that you mention it, UNC's network security would probably crawl all over me for running an open chat server. they almost had a coronary over the Tor nodes
13:36 pdurbin
heh
13:37 pdurbin
Well, open in what way?
13:37 pdurbin
I'm just talking about the logs and data being open.
13:38 donsizemore
in general, they frown when people not affiliated with UNC are able to access and use UNC network resources.
13:38 pdurbin
Sure, that makes sense.
13:40 pdurbin
What about some SLOPI chat rooms under https://gitter.im/GlobalDataverseCommunityConsortium ?
13:42 donsizemore
:thumbs up:
13:44 jonas80
this 404s for me
13:45 jonas80
i need to sign up first, don't i
13:48 poikilotherm
Do you have a Github account? :-D
13:49 pdurbin
If you look at https://gitter.im/jupyter/home you can see many chat rooms. The idea would be to do something similar for Dataverse.
13:49 pdurbin
1111 people in https://gitter.im/jupyter/jupyter it says. :)
13:53 stefankasberger
@donsizemore: sure. what do i need to do for it?
13:56 pdurbin
Welcome to all the new people here! aseeland doigl eRATG :)
14:01 stefankasberger
@donsizemore @pdurbin : have setup the webhook in the pyDataverse GitHub repo. should now work from my side.
14:09 stefankasberger
@pdurbin: if you invite me, i would join the call with dataverse DE too. skype: cheeseman1983
14:16 jonas80
@stefankasberger: did we meet at EDDI18?
14:17 jonas8026 joined #dataverse
14:17 jonas81 joined #dataverse
14:24 doigl
Thanks, aseeland and I are the ones from Stuttgart in the Dataverse-DE call
14:25 stefankasberger
Jonas Recker? Then yes. :)
14:42 stefankasberger
oh yes, now i see you better on the cam. :)
14:54 donsizemore
@pdurbin how often would you say demo.dataverse.org gets wiped out?
14:55 donsizemore
@pdurbin i see dataverses from 2015...
15:01 donsizemore
@stefankasberger: py36: commands succeeded coverage: commands succeeded coveralls: commands succeeded docs: commands succeeded flake8: commands succeeded packaging: commands succeeded dist: commands succeeded congratulations :)
15:03 poikilotherm joined #dataverse
15:08 jonas82 joined #dataverse
15:18 jonas82
What is the best way we can participate in the community meeting without being there physically?
15:25 jonas82
@stefankasberger - jonas kahle, we discussed about workflow software, meaning business process management in this instance
15:25 jonas82
(too late...)
15:29 pdurbin
jonas82: that's a great question
15:39 pdurbin
jonas82: I just emailed the committee
15:40 pdurbin
pkiraly: can we link those notes from your meeting in here? This channel is logged and public.
16:08 pkiraly
pdurbin: please don't. One of our participants does not like to share it. Maybe we will create a "public" version later.
16:21 pdurbin
Ok. If a public version becomes available, please let me know. I guess that's for doigl jonas82 and poikilotherm :)
16:32 pdurbin
rigelk: thanks for you comments at https://github.com/IQSS/dataverse/issues/5883#issuecomment-498306349 !
16:34 jri joined #dataverse
16:44 andrewSC joined #dataverse
17:10 poikilotherm joined #dataverse
17:15 dan-drexel joined #dataverse
17:15 dan-drexel
Can someone help me with an issue Im having installing Dataverse?
17:15 dan-drexel
I cant get Solr to start and run
17:16 dan-drexel
Loaded: loaded (/usr/lib/systemd/system/solr.service; disabled; vendor preset: disabled) Active: failed (Result: start-limit) since Mon 2019-06-03 13:11:47 EDT; 2min 21s ago Process: 5891 ExecStart=
17:16 dan-drexel
Job for solr.service failed because the control process exited with error code. See "systemctl status solr.service" and "journalctl -xe" for details.
17:23 poikilotherm joined #dataverse
17:24 poikilotherm
pdurbin I just created https://gitter.im/IQSS/dataverse
17:24 poikilotherm
Let's go try this :-D
17:33 pdurbin
poikilotherm: huh? How did you have permission to create that?
17:33 pdurbin
dan-drexel: Solr trouble? Are you on CentOS?
17:34 poikilotherm
I think I got permission to do so by being an admin on IQSS/dataverse-kubernetes
17:35 donsizemore joined #dataverse
17:35 poikilotherm
All admin from IQSS org have admin rights on that channel by default
17:37 pdurbin
But is IQSS the right namespace? I guess I was thinking that https://gitter.im/GlobalDataverseCommunityConsortium might scale better. In the future when there is more than one room. What do you think?
17:38 poikilotherm
Hmm - I thought that we should reuse existing channels
17:38 poikilotherm
And it seemed you are open to let IRC die in favor of Gitter/SLOPI-capable channels
17:39 poikilotherm
If you think this is inappropriate, please go ahead and delete the room
17:39 poikilotherm
No spoiled milk yet
17:40 pdurbin
I'm thinking. And I do still like IRC and freenode. They're old friends. That said, I care more about the data, the messages, than the platform. Same with Google Groups.
17:41 pdurbin
I like that you're a do-er. :)
17:42 poikilotherm
:-D
17:42 pdurbin
But its troubling than a non-owner of IQSS can create a room like that.
17:42 pdurbin
poikilotherm: I'm going to PM you something.
17:46 pdurbin
donsizemore: Hi! Did you see that dan-drexel is having Solr trouble?
17:49 donsizemore
@pdurbin I did not. @dan-drexel what happens when you run "/usr/local/solr/bin/solr start" (or equivalent) as the solr user?
18:03 pdurbin
poikilotherm: I just create a spreadsheet to help think more about the namespace for Gitter: https://docs.google.com/spreadsheets/d/129ANYQotm2IX-uJRAh-EqR41nwjJjbgsZh3eMUnZ_kE/edit?usp=sharing
18:03 pdurbin
Everyone here is welcome to request edit access.
18:06 * poikilotherm
goes looking...
18:07 pdurbin
poikilotherm: maybe I should organize it by description of the room. We could probably use a single tab that way.
18:07 pdurbin
My point is that I think we'll (eventually) want more than one room. And I'm not sure that "IQSS" is the best namespace for this.
18:07 poikilotherm
Meh the table does not load on slow ICE train WLAN :-(
18:08 pdurbin
For IQSS Slack we prepend "#dv-" to a lot of channnels, #dv-general for example.
18:10 poikilotherm
When changing the org name in GItter, wouldn't the same apply for Dataverse on Github?
18:10 poikilotherm
There are a lot of repos popping up these days
18:11 poikilotherm
One could think of moving IQSS/dataverse-kubernetes to dataverse/kubernetes, too
18:11 poikilotherm
That might support the nature of being a community thing
18:11 poikilotherm
Or Docker images on Dockerhub under "dataverse", not "iqss"
18:12 pdurbin
Well, that's why the consortium exists, in part, I believe. To gather us together. To be larger than IQSS.
18:13 pdurbin
But my point is that I think we might want multiple channels on Gitter, eventually.
18:13 poikilotherm
Yeah. But writing "GlobalDataverseCommunityConsortium" in links, image tags and more is quite unpleasant
18:13 pdurbin
Sure, but with Gitter you can create artibrary org names that aren't tied to a GitHub org.
18:13 poikilotherm
Totally agreed.
18:14 poikilotherm
My point is that Dataverse should be transfered as a name consistently. It even is a trademark
18:14 poikilotherm
Currently, this is mostly done with "IQSS"
18:14 poikilotherm
It is consisten
18:14 poikilotherm
IMHO one should stick with being consistent
18:15 pdurbin
On Twitter we are @dataverseorg.
18:15 pdurbin
I think we might even be on Facebook but I'm not sure.
18:17 pdurbin
I'd like to get iq.dataverse.org out of the Java packages and switch it to dataverse.org
18:17 pdurbin
er, that was totally wrong
18:18 pdurbin
I meant to say this:
18:18 pdurbin
Instead of "package edu.harvard.iq.dataverse.api;" in the Java I'd rather see "package org.dataverse.api;"
18:18 pdurbin
or something
18:19 poikilotherm
org.dataverse... ;-)
18:19 poikilotherm
dataverseorg is also totally fine for image names, github repos and the like
18:19 poikilotherm
(also I like it shorter)
18:19 pdurbin
when we started Dataverse 4 we didn't own https://dataverse.org
18:19 poikilotherm
+1 :-D
18:19 poikilotherm
Together with a checkstyle enforcement please :-D
18:20 poikilotherm
I think you had to pick dataverseorg on twitter as dataverse was already taken: https://twitter.com/dataverse
18:21 poikilotherm
2008 vs 2012
18:21 pdurbin
Yeah, some guy in Germany has it. Can you please talk to him? I've emailed him several times.
18:21 poikilotherm
LOL
18:21 pdurbin
He wasn't in Berlin. I looked.
18:22 poikilotherm
He is in Bonn according to Twitter :-D
18:22 pdurbin
Ok. Please find him for me.
18:25 poikilotherm
Oh he is a professor in Darmstadt
18:25 poikilotherm
https://fbi.h-da.de/personen/michael-vonrueden/
18:26 pdurbin
him, that email is not the one I've been using. thanks!
18:27 pdurbin
hmm*
18:27 poikilotherm
But I might be wrong... The spelling is a bit different
18:28 pdurbin
Maybe he knows the right guy though. :)
18:30 poikilotherm
At least the topics on http://www.michael-vonrueden.de/ match with his description on the website of UA darmstadt
18:30 pdurbin
I'm emailing him.
18:33 poikilotherm
pdurbin: is it correct that I don't see any entries in your spreadsheet except the two explaining lines?
18:33 pdurbin
there are 3 tabs
18:33 pdurbin
email sent
18:33 poikilotherm
Ah!!!
18:33 poikilotherm
Ok, I see
18:34 pdurbin
But what I was trying to say is that perhaps we could have one tab organized by description.
18:34 poikilotherm
Can't those be merged in one tab and different columns?
18:34 pdurbin
yes, fixing
18:35 pdurbin
fixed
18:39 donsizemore
@pdurbin akio and i are talking payara5 (or other) for future dataversia. just curious if the future platform will/could be a BoF discussion at the community meeting?
18:40 pdurbin
donsizemore: sure! Did he see Steve M's comments on the Payara issue? Do you think Akio could start a pull request?
18:40 donsizemore
we have #2628 up on the screen =)
18:41 poikilotherm
pdurbin: I just requested access to the doc, would you be so kind to grant permission? :-)
18:41 donsizemore
@pdurbin part of that would involve re-architecting intro microservices, which he's doing now WRT ingest and TRSA
18:41 donsizemore
^into
18:42 pdurbin
poikilotherm: done!
18:42 poikilotherm
Gnaaah... It's a pity I can't come to #dcm2019 :-(
18:42 pdurbin
donsizemore: Steve M had some specific suggestions for the search service bean or something
18:43 poikilotherm
pdurbin: thx!
18:43 pdurbin
I betcha poikilotherm would appreciate it if Akio makes a pull request to get Dataverse running better on Payara 5. The search service stuff or whatever it was. :)
18:46 poikilotherm
Sure thing! I was going to ask you guys who wants to get the things done, so we can rock and roll...
18:46 poikilotherm
Happy to help with this
18:46 pdurbin
donsizemore: if Akio gets the branch up thre I can try to hack on it.
18:47 pdurbin
poikilotherm: or you can create a branch :)
18:47 donsizemore
we're looking at https://i.stack.imgur.com/uuPbl.jpg
18:47 donsizemore
akio notes dataverse currently needs jax-ws, jms, javamail, jaxb though ws is questionable
18:47 poikilotherm
Pretty look at MicroProfile :-D
18:47 poikilotherm
+please
18:49 pdurbin
donsizemore: we can have our cake and eat it too with Payara, Wildfly, and Liberty. They support Java EE and Microprofile. You can mix and match APIs. Why not both! That's the idea.
18:50 donsizemore
@pdurbin this discussion started by akio asking about the outcome of your platform discussion during tech hours last week =)
18:50 poikilotherm
donsizemore: before heading for heavy refactoring to micro service, it would be already a great enhancement to setup resources etc from code, refactor timers etc.
18:50 poikilotherm
Get things into using Hazelcast
18:51 poikilotherm
See #5292
18:51 poikilotherm
Need to change my train now, will be back online in about 30-40 minutes
18:51 poikilotherm
Oh, not true! Got another 10 minutes :-D
18:51 pdurbin
So it's not like we need to slim Dataverse down to fit into Microprofile. Dataverse can continue to be fat. Dataverse can continue to use all the Java EE Full APIs. Does that make sense, donsizemore ?
18:52 pdurbin
donsizemore: did you show Akio my notes after that tech hours meeting? I can shoot you the link.
18:52 poikilotherm
(Don't get me wrong - I am all into cuttting and slicing... But this is a HUGE thing, and there might be other low hanging fruits that would make that step easier)
18:52 donsizemore
@pdurbin i first dug up @poikilotherm's post to dataverse-dev, then we started traipsing through issues
18:52 donsizemore
@pdurbin a link would be great
18:53 poikilotherm
-might
18:53 pdurbin
donsizemore: here are my notes after last week's tech hours discussion: https://github.com/IQSS/dataverse/issues/2628#issuecomment-496709057
18:53 poikilotherm
+are
18:54 donsizemore
@pdurbin we were on that issue =) but hadn't scrolled up enough from the bottom
18:54 pdurbin
lots of chatter! the way I like it :)
18:55 poikilotherm
pdurbin: I added some stuff to your doc :-)
18:56 poikilotherm
Alright guys, arriving at Frankfurt now. See you later
18:56 pdurbin
poikilotherm: I like the "dataverse" namespace best, I think
18:56 donsizemore
@pdurbin oh. and seeing you TRSA question in that issue: it currently runs payara5
18:56 pdurbin
donsizemore: cool, that's why I was saying we should coordinate with you
18:59 donsizemore
@pdurbin check out question 3 at https://arjan-tijms.omnifaces.org/2018/05/java-ee-survey-2018-results.html
18:59 donsizemore
@pdurbin akio points out that wildfly is backed by red hat, the new caretaker for open source java
19:00 pdurbin
Question 3 - Which application servers have you recently used?
19:00 pdurbin
So why is Akio using Payara instead? :)
19:00 donsizemore
yis
19:01 donsizemore
path of least resistance
19:01 pdurbin
Do we have to pick just one?
19:01 pdurbin
Can we get one of Payara, Wildfly, or Liberty working and then support the other two later?
19:02 donsizemore
i just asked if he wanted me to test deploying TRSA in wildfly or liberty
19:02 pdurbin
nice!
19:02 donsizemore
he points out that resource configuration may be totally different across platforms
19:03 jri joined #dataverse
19:07 poikilotherm joined #dataverse
19:08 poikilotherm
donsizemore: config is somewhat different. using code annotations might take care of some of this
19:08 poikilotherm
e.g. for JDBC connections. IIRC for JMS, too
19:09 poikilotherm
What I like about Payara is Hazelcast out of the box
19:09 pdurbin
donsizemore: I *think* the Microprofile Config API might help. poikilotherm probably knows better.
19:09 poikilotherm
pdurbin you like to run multi instance dataverse, right? then hazelcast will help a lot
19:10 poikilotherm
COnfig api is usable for resources in annotations, yes
19:10 pdurbin
I do have my eye on hazelcast.
19:10 pdurbin
poikilotherm: donsizemore and Akio have a (different) app deployed to Payara and are wondering how hard it would be to get it deployed to Wildfly and Liberty.
19:10 pdurbin
donsizemore: what's the app called again, please?
19:11 donsizemore
https://github.com/OdumInstitute/trsa-web
19:11 pdurbin
thanks! poikilotherm ^^
19:11 donsizemore
i can take on test deployments if you like
19:11 donsizemore
it's a pretty small web app but does (temporarily) use h2 for its DB
19:11 pdurbin
Honestly, for Dataverse, I think we should for for Payara for now. And worry about Wildfly and Liberty later.
19:12 pdurbin
go* for
19:13 dan-drexel
@pdurbin and @donesizemore I get this Port 8983 is already being used by another process (pid: 5621) Please choose a different port using the -p option.
19:13 poikilotherm
donsizemore why are you on EE7?
19:13 dan-drexel
and also I am on CentOS 7
19:15 pdurbin
dan-drexel: you can use lsof to figure out what's listening on that port. Sounds like you know the pid already. Is it another installation of Solr?
19:15 poikilotherm
donsizemore you don't seem to use things like glassfish-resources.xml, so in theory you should be good to go
19:16 poikilotherm
Deploying resources manually however will be different for Wildfly/LIberty
19:17 dan-drexel
@pdurbin It's showing java as using port 5621
19:17 pdurbin
Let's focus for a moment on helping dan-drexel with his Solr problem. We can debate this or that app server in a bit. :)
19:17 poikilotherm
Sure :-)
19:17 donsizemore
@pdurbin oh, i'm not debating — akio was asking
19:17 dan-drexel
Thanks!!! I appreciate the help!
19:18 pdurbin
dan-drexel: ok, and Solr is Java. So maybe you already have Solr running?
19:19 dan-drexel
@pdurbin: when I do try 'service solr status' I get this:
19:19 dan-drexel
Loaded: loaded (/usr/lib/systemd/system/solr.service; disabled; vendor preset: disabled) Active: failed (Result: start-limit) since Mon 2019-06-03 15:10:36 EDT; 7min ago Process: 15448 ExecStart
19:19 dan-drexel
So it shows it as failed for active
19:20 poikilotherm
dan-drexel could you please post the output of `ps -f 5621 | cat`
19:20 poikilotherm
Then we will know what exactly is going on (java is used by a few things here...)
19:20 dan-drexel
UID PID PPID C STIME TTY STAT TIME CMD solr 5621 1 0 13:11 ? Sl 0:17 java -server -Xms512m -Xmx512m -XX:NewRatio=3 -XX:SurvivorRatio=4 -XX:TargetSurvivorRatio=90 -XX:MaxTenuringThreshold=8 -XX:+UseConcMarkSweepGC -XX:ConcGCThread s=4 -XX:ParallelGCThrea
19:21 poikilotherm
Did you use the installer?
19:21 poikilotherm
(And thx for posting)
19:21 dan-drexel
not yet, just working through the pre-reqs
19:22 poikilotherm
Ok, this definitly looks like you have a Solr process running. Is this a machine dedicated to Dataverse or is sth. else going on here?
19:22 dan-drexel
Also if it helps this is a fresh CentOS install as a virtualbox vm
19:22 poikilotherm
Thx!
19:22 dan-drexel
Nope, just for Dataverse
19:22 poikilotherm
This might be a process which systemd lost track of
19:22 poikilotherm
Could you please kill the process?
19:23 poikilotherm
And then try to restart Solr via systemctl?
19:23 dan-drexel
I'm actualy fairly new to Linux - how do i kill the process?
19:23 poikilotherm
kill 5621
19:23 dan-drexel
Ahh, that worked!!! It's now active and running
19:24 dan-drexel
What do we think happened here?
19:24 poikilotherm
Nice
19:24 dan-drexel
Thanks for the help!!!
19:24 poikilotherm
To be sure, can you please restart the VM and let us know if there is still a Solr process running after that?
19:25 dan-drexel
Sure, I will do that now
19:25 poikilotherm
Thx :-)
19:25 poikilotherm
While we are waiting for the restart: did you install Solr as described in the install guides?
19:25 poikilotherm
Or from RPM/...
19:26 dan-drexel
Yep, following the guide exactly as it says
19:26 poikilotherm
Ok, then I suppose that was the pitfall
19:26 poikilotherm
The guide lets you start Solr to create the core
19:27 donsizemore
correct
19:27 poikilotherm
And afterwards, you install the init script
19:27 poikilotherm
And simply call start with systemd
19:27 poikilotherm
That will obviously crash
19:27 poikilotherm
So this is a docs bug :-D
19:27 dan-drexel
I see. Yep, just restarted and still running
19:27 pdurbin
Can someone please create a GitHub issue to fix the guides? Want to, dan-drexel?
19:27 poikilotherm
Hey, pdurbin, maybe you can ask him your favorite question :-D
19:28 poikilotherm
*dingding*
19:28 poikilotherm
:-D
19:28 pdurbin
way ahead of you
19:28 * poikilotherm
needs to calm down giggling
19:28 dan-drexel
Hahaha what is this question?
19:29 pdurbin
dan-drexel: I'm asking (like I usually do) if you would be so kind as to create an issue at https://github.com/IQSS/dataverse/issues about the Solr doc bug. :)
19:29 poikilotherm
dan-drexel, can you please "restart" in the guide where it says "systemctl start solr.service"?
19:29 poikilotherm
Should be all good now
19:29 poikilotherm
(Except you might find another docs bug :-D)
19:30 poikilotherm
(With restart I mean goto and continue from there)
19:30 pdurbin
poikilotherm: that reminds me. donsizemore rewrote some of the prereq page. I forget if you were still hacking on it or not.
19:30 dan-drexel
Ohh I see! That makes sense now. No problem, I'll log the bug
19:30 pdurbin
dan-drexel: thank you!!
19:30 poikilotherm
pdurbin donsizemore: feel free to pipe in your changes.
19:30 dan-drexel
Thank you guys for your help!
19:30 poikilotherm
I feel my current work is far from complete
19:30 poikilotherm
And it will need changing once on Payara 5 anyway :-D
19:30 pdurbin
poikilotherm: ok, Merce liked your diagram, for what it's worth. :)
19:31 poikilotherm
Thx for the flowers :-)
19:31 pdurbin
poikilotherm: if we can get back to Gitter for a moment, what do you think of "I am a member of an organization on Github and it appears I have full rights to delete public and private rooms" at https://gitlab.com/gitlab-org/gitter/webapp/issues/882 ?
19:32 poikilotherm
Doe!
19:33 poikilotherm
Jesus Christ, is Gitter a good choice?
19:33 pdurbin
I'm starting to wonder.
19:33 pdurbin
173 people in https://github.com/IQSS
19:34 poikilotherm
BTW they still haven't fixed the issue with the "online indicator"
19:34 donsizemore
information wants to be free!
19:34 pdurbin
heh
19:34 poikilotherm
dan-drexel: you are most welcome. If you find any other doc bugs, please report :-D
19:35 pdurbin
Let's not be too mean to Gitter. They provide a valuable service to the open source community.
19:35 poikilotherm
It is always superb to have new people on the block finding our slips
19:35 donsizemore
@pdurbin in my shell history looks like i ignored the solr start call and called it with systemd instead.
19:35 dan-drexel
One more question...
19:35 donsizemore
@pdurbin want a PR to just say that instead?
19:36 dan-drexel
When I do service solor restart I get this: /sbin/service: line 3: /etc/init.d/functions: Not a directory
19:36 pdurbin
donsizemore: sure!
19:36 donsizemore
@dan-drexel a lot of the docs still reference rhel/centos6... if you're on 7 you'll want to do most things with systemd
19:37 dan-drexel
so then "systemctl restart solr.service"?
19:37 donsizemore
yes sir
19:37 poikilotherm
One might argue that novice linux users might not understand the current docs on this... Maybe add a hint about distros using old init.d
19:38 dan-drexel
Excellent! Yep, for me I would not recognize that.
19:38 poikilotherm
Or simply remove that. IMHO the days of CentOS 6 and other oldies are long gone, at least for new installations
19:38 pdurbin
The last CentOS 6 server that I care about is phoenix.dataverse.org
19:39 donsizemore
@poikilotherm agreed, but some folks at IQSS are still on 6 and wanted to keep them
19:39 poikilotherm
Ok, then add a deprecated notice with "will be removed on next major release" :-D
19:39 pdurbin
dan-drexel: thank for creating the issue! The next question I always ask is if I can coach you through a pull request. :)
19:39 poikilotherm
(When sticking to 5.0 on Payara 5, this is not far away... hihihi)
19:42 dan-drexel
@pdurbin no problem! what is a pull request?
19:42 dan-drexel
@pdurbin do you by chance work at Harvard?
19:43 pdurbin
A pull request is a suggestion for a change to a GitHub repo. Yes, I work at Harvard.
19:44 dan-drexel
I see, should that have been a pull request rather then a issue report?
19:45 dan-drexel
My co-worker Steve Melly used to work at Harvard, do you know him?
19:45 dan-drexel
he was a GIS person
19:46 pdurbin
Hmm. Nope but from searching my email I see his name in the 2012-2015 timeframe.
19:47 pdurbin
And an issue first is perfect.
19:49 poikilotherm
Man this is Dataverse on speed. Going at 300km/h from Frankfurt to Cologne...
19:49 pdurbin
Dataverse doesn't go that fast.
19:49 poikilotherm
(186mph)
19:50 poikilotherm
Yeah, sometime you get the impression it takes as long as from Frankfurt to Cologne till it deploys on startup
19:50 pdurbin
yeah
19:52 poikilotherm
Oi pdurbin do you see an option for running a RocketChat server on IQSS AWS bill?
19:52 pdurbin
poikilotherm: maybe if someone volunteers to be a sysadmin for it
19:53 poikilotherm
https://rocket.chat/2018/11/28/rocket-chat-launches-on-aws-marketplace/
19:55 poikilotherm
And obviously https://rocket.chat/docs/installation/paas-deployments/aws/
19:55 poikilotherm
It might be worth a try...
19:57 pdurbin
Again, for me it comes down to the care and feeding of the service. Who would fix it when it goes down? That sort of thing.
19:58 pdurbin
dan-drexel: I requested code review from donsizemore and poikilotherm on your pull request.
19:58 poikilotherm
Right. There might be room for those kind of resources. While discussing at Göttingen, it seemed much more complicated to provide computing resources for us
19:58 * poikilotherm
heading over to the PR
19:59 poikilotherm
I will try to reach Jonas and talk to him about htis
19:59 poikilotherm
donsizemore would you welcome a change from IRC to sth. like RocketChat? Not sure what you wrote earlier this day
20:03 poikilotherm
Posted a review :-)
20:06 pdurbin
thanks!
20:07 poikilotherm
BTW pdurbin: today I learned some IT personell is rejecting using Dataverse in Docker/Kubernetes just for in the docs saying "this is highly experimental". This definitly needs to change :-D
20:08 pdurbin
poikilotherm: uh oh. Maybe we should link to the installation guide?
20:09 pdurbin
To suggest a more traditional installation approach, I mean.
20:09 pdurbin
Or are you saying that using Docker is a hard requirement for them?
20:09 poikilotherm
Maybe when refactoring that part of the install guide we should say sth about "community driven" and maybe add a doc of users in dataverse-kubernetes, so people can get an impression
20:10 candy`_ joined #dataverse
20:10 pdurbin
Sure, that all sounds fine.
20:10 poikilotherm
No no no. The are _rejecting_ using Docker or even Kubernetes
20:10 poikilotherm
They believe in docs, which is good. So the docs should change ;-)
20:11 poikilotherm
Get more attraction to the project... ;-)
20:11 pdurbin
If you make a pull request I'm happy to review it. :)
20:11 poikilotherm
Yeah yeah. Let donsizemore push his changes first
20:12 pdurbin
Oh, I thought you were talking about the dataverse-kubernetes README.
20:12 poikilotherm
I am talking about the install guide :-D
20:13 poikilotherm
<poikilotherm> Maybe when refactoring that part of the install guide we should say sth about "community driven" and maybe add a doc of users in dataverse-kubernetes, so people can get an impression
20:13 poikilotherm
Gotta switch trains again. No WLAN in there :-(
20:13 poikilotherm
Have a good night / evening everyone
20:13 pdurbin
Are you talking about http://guides.dataverse.org/en/4.14/developers/containers.html#future-production-use-on-minishift-openshift-kubernetes ?
21:06 donsizemore joined #dataverse
21:14 jri joined #dataverse
