Time
S
Nick
Message
07:12 juancorr joined #dataverse
07:43 MrK joined #dataverse
08:25 stefankasberger joined #dataverse
08:26 stefankasberger3 joined #dataverse
08:31 stefankasberger joined #dataverse
09:07 stefankasberger joined #dataverse
09:25 stefankasberger joined #dataverse
09:45 stefankasberger4 joined #dataverse
09:45 stefankasberger4 left #dataverse
09:46 stefankasberger joined #dataverse
10:17 pdurbin joined #dataverse
10:23 poikilotherm joined #dataverse
11:31 pdurbin
Uh oh. "Breaking API changes in 4.12?" https://github.com/IQSS/dataverse/issues/5724
11:54 donsizemore joined #dataverse
12:32 pdurbin
need more continuous integration
12:32 pdurbin
I just renamed that issue to "Breaking API changes in 4.12: RSpace (and other?) integrations broken" and posted some screenshots.
12:34 poikilotherm
:-D
12:34 poikilotherm
Yeah!!!
12:34 poikilotherm
More CI :-)
12:35 pdurbin
Did everyone see the new Jenkins installation donsizemore set up? https://jenkins.dataverse.org
12:35 poikilotherm
Fancy
12:36 poikilotherm
I might be interested for things using k8s
12:36 poikilotherm
Docker or similar available?
12:36 donsizemore
@poikilotherm absolutely
12:38 donsizemore
@poikilotherm i'm kicking the tires in dataverse-kubernetes/microk8s this morning =)
12:39 poikilotherm
YEAH!
12:39 poikilotherm
That sound marvellous
12:39 poikilotherm
+s
12:40 poikilotherm
When you have problems with microk8s, you could take a look at K3S too
12:40 poikilotherm
https://k3s.io/
12:40 poikilotherm
Might be even easier and smaller - its just a simple curl
12:40 poikilotherm
Made for CI :-)
12:44 pdurbin
Would anyone like to create an issue with a title something like "As a member of the Dataverse community, I'm interested in learning more about continuous integration solutions that are available to me"? We should put something in the dev guide about this new Jenkins service. And write more about Travis CI as well.
12:44 stefankasberger joined #dataverse
12:46 donsizemore
@pdurbin you should know I _hate_ creating issues =p
12:47 poikilotherm
ROFL :-D
12:47 pdurbin
:)
12:49 pdurbin
Did we ever decide where to create the repo? I'm happy to create one under IQSS GitHub called jenkins.dataverse.org or dataverse-jenkins or whatever. Maybe dataverse-jenkins would be better in case orgs in the community want to run their own Jenkins.
12:50 donsizemore
you're the boss
12:51 pdurbin
no I'm not :)
12:51 pdurbin
donsizemore: would you rather have the repo under https://github.com/GlobalDataverseCommunityConsortium ?
12:51 poikilotherm
pdurbin I was thinking this morning about renaming to dataverse-k8s... I love short names :-D
12:52 donsizemore
@pdurbin you're asking me the hard questions.
12:52 pdurbin
poikilotherm: too unfriendly, in my opinion, but it's your repo
12:53 pdurbin
donsizemore: I'll just stick it under IQSS for now if that's ok. You'll be able to contribute your ansible configs or whatever? The goal would be that I'd be able to stand up my own installation of Jenkins. Or anyone who runs a fork of Dataverse.
12:54 poikilotherm
Er
12:54 poikilotherm
Just a suggestions
12:54 poikilotherm
-s
12:54 poikilotherm
When you target not just Jenkins, but also others like GitLab CI, Travis, etc, maybe it makes sense to have a more general name?
12:55 pdurbin
hmm
12:55 poikilotherm
People often get opinionated when it comes to choosing CI servers ;-)
12:55 pdurbin
but is anyone actually using GitLab CI?
12:55 * poikilotherm
raises his finger
12:55 pdurbin
anyone in the Dataverse community, I mean
12:55 * poikilotherm
raises his finger again
12:56 pdurbin
I can make you a dataverse-gitlab-ci repo if you want. Do you have something you'd like to share with the community?
12:56 poikilotherm
And I know that a lot of others in the german science community are adapting to using Gitlab
12:56 poikilotherm
Nope, not yet
12:56 poikilotherm
But that k8s stuff might be interesting for this, too
12:57 poikilotherm
And not just for Jenkins
12:57 pdurbin
sure
12:57 pdurbin
we went with Jenkins because donsizemore already had it set up for pull requests Akio is working on (and who knows what else) :)
12:57 poikilotherm
Just my 2 cents. Feel free to act as you prefer :-)
12:57 poikilotherm
That's perfectly fine
12:58 pdurbin
And IQSS uses Jenkins too, of course.
12:58 poikilotherm
I was just thinking about what you said above: when people want to run their own stuff...
12:58 pdurbin
But I think a dedicated Jenkins instance for the whole Dataverse community will be quite valuable.
12:58 pdurbin
Sure, when you run a fork you have extra needs.
12:58 poikilotherm
Totally!
12:59 pdurbin
Jim has a nice diagram of their CI setup.
12:59 poikilotherm
I had the impression you guys are starting to collect scripts etc for CI in that repo
12:59 poikilotherm
And those could be run in other enviroments too
12:59 poikilotherm
Like Travis, GitLab CI, etc
12:59 pdurbin
Well, the repo doesn't exist yet. :)
12:59 poikilotherm
Aye
13:00 poikilotherm
That's why I suggested naming it not "*jenkins*" but keep it more general
13:00 pdurbin
But I think you both see what I'm getting at. If I run a fork of Dataverse, I want to set up some CI to make sure everything is still working.
13:00 poikilotherm
Changing this later on is always tricky ;-)
13:00 poikilotherm
Yeah
13:00 pdurbin
meh, I'm not worried about renaming it later
13:00 poikilotherm
Alright :-D
13:00 poikilotherm
Then fire at will ;-)
13:01 pdurbin
again, I'm happy to create a repo for the gitlab ci users, even if there's only one right now in the dataverse community :)
13:01 poikilotherm
Oh by the way: did you see https://pcottle.github.io/learnGitBranching/?demo
13:01 poikilotherm
Would this be a good adendum to the dev guide?
13:02 pdurbin
I've played with that pcottle thing a bit. Try the long link at http://wiki.greptilian.com/git/branching/ . Sure, add it to the dev guide.
13:03 poikilotherm
:-)
13:04 poikilotherm
And another BTW: Stuttgart and Göttingen are trying to get a Dataverse DE community meeting together
13:04 pdurbin
orly
13:07 poikilotherm
Yeah, I'll see what my supervisors tell me about this.
13:07 poikilotherm
Wanna come to Göttingen? :-P
13:08 pdurbin
Sure! Wait... Stuttgart...
13:09 pdurbin
residence: stuttgart according to https://en.wikipedia.org/wiki/Frank_Karlitschek
13:10 pdurbin
The reason I bring this up is that I chatted with Frank at LibrePlanet a couple weekends back. And then he posted this: https://groups.google.com/d/msg/dataverse-community/A5l4P9xKwYI/Xi23owkBBQAJ
13:13 donsizemore
i think it should be in stuttgart for the hymntune
13:13 poikilotherm
LOL
13:14 poikilotherm
Well I could suggest that and tell them you might come when we change the location ;-)
13:14 poikilotherm
We could have another mini-hackathon :-D
13:14 pdurbin
more than 20 minutes this time, please
13:14 poikilotherm
Aye!
13:17 donsizemore
@poikilotherm how long does it typically take for your bootstrap pod to become ready? i'm at 3m of hardly any system activity... i suspect a stalled glassfish download?
13:17 poikilotherm
It feels like taking ages
13:18 poikilotherm
The deployment takes minutes
13:18 poikilotherm
On a reasonable machine it should deploy at about 3 minutes
13:18 poikilotherm
or 2
13:18 poikilotherm
The bootstrap job simply waits for the deployment to finish
13:18 poikilotherm
Have a look at the logs of the dataverse pod
13:19 poikilotherm
glassfish is downloaded at image build time
13:19 poikilotherm
it really is "just" the deployment
13:19 pdurbin
donsizemore: I just made you an admin of https://github.com/IQSS/dataverse-jenkins . Can you please push for Ansible config or whatever there?
13:20 poikilotherm
As pdurbin stated in his issue/dev mail, it is increasingly getting slower
13:20 pdurbin
s/for/your/
13:20 donsizemore
@poikilotherm deploy failed. it can't connect to the DB container.
13:21 poikilotherm
Any chance we can have a remote session?
13:21 donsizemore
FATAL: role "root" does not exist
13:21 poikilotherm
I did not yet try this with mikrok8s
13:21 poikilotherm
Used minikube and a full-grown k8s install only so fa
13:21 poikilotherm
There might be some pitfalls when using micro-k8s
13:22 donsizemore
i blame microk8s's use of microk8s.* naming of binaries
13:22 donsizemore
i had minikube but trashed it when ubuntu advertised microk8s
13:22 poikilotherm
*G*
13:22 poikilotherm
Yeah, lots of opinions out there :-D
13:23 donsizemore
i like distribution-packaged-anything when possible
13:23 donsizemore
if you want i can pop my experience into the issue for posterity
13:24 poikilotherm
Sure.
13:24 poikilotherm
Let me check this
13:24 donsizemore
everything looks fine except for that postgres error
13:24 poikilotherm
Will install snap on my Fedora machine
13:24 poikilotherm
Did that postgres container deploy properly?
13:24 poikilotherm
And is that service online and registred?
13:25 donsizemore
it thinks it's happy:
13:25 donsizemore
dls
13:26 donsizemore
forgot not to paste snippets, sorry.
13:26 poikilotherm
Errrr
13:26 poikilotherm
Nope.
13:26 poikilotherm
It is totally unhappy
13:26 donsizemore
and postgres thinks it's happy "LOG: database system is ready to accept connections" but then "FATAL: role "root" does not exist"
13:27 donsizemore
i'll nuke all this and try k3s maybe
13:27 poikilotherm
There is no solr and not postgres pod
13:27 poikilotherm
You might need to create PVs
13:28 poikilotherm
Not sure on that for micro-k8s
13:28 donsizemore
they're jumbled in there, and say they're running
13:28 poikilotherm
Hmm don't see them in the log above?
13:28 donsizemore
lemme nuke it all and try in something that isn't microk8s
13:28 poikilotherm
:-D
13:28 poikilotherm
Okidoki
13:29 donsizemore
i can see them via 'get pods' and 'get svc' — i think it all comes down to the DB role error
13:30 pdurbin
donsizemore: can I assign https://github.com/IQSS/dataverse-jenkins/issues/1 to you? :)
13:35 poikilotherm
Wow, k3s was definitly very quick to setup :-D
13:38 pdurbin
so nice that xarthisius is back :)
13:39 pdurbin
Does anyone want to play around with launching Jupyter notebooks or RStudio from Dataverse using Whole Tale? Try it from the "Explore" button at https://dev2.dataverse.org/file.xhtml?fileId=29
13:39 pdurbin
andrewSC bjonnh bricas_ candy` dzho icarito[m] juancorr MrK pmauduit poikilotherm sivoais yoh ^^
13:56 jonas47 joined #dataverse
14:01 donsizemore
@pdurbin you're welcome to assign it to me, but you're giving me too much credit. it's just a jenkins rpm on centos7
14:02 donsizemore
@pdurbin for an ansible role i would defer to jeff geerling https://github.com/geerlingguy/ansible-role-jenkins
14:07 pdurbin
Should we have a Vagrantfile?
14:09 donsizemore
we can make one
14:10 pdurbin
cool
14:11 pdurbin
or whatever, a script to spin it up on ec2
14:11 poikilotherm
pdurbin: https://github.com/IQSS/dataverse-kubernetes/releases/tag/v4.12
14:12 poikilotherm
This is a pre-release and I hope it proves usefull
14:13 pdurbin
poikilotherm: thanks! Sadly, Craig won't be using it. :( Please see https://github.com/IQSS/dataverse/issues/4152
14:14 poikilotherm
Ok
14:14 poikilotherm
But it seems like donsizemore is interested :-D
14:16 pdurbin
Me too. Want to help us add it to https://jenkins.dataverse.org ?
14:16 pdurbin
How would it make sense to add it to https://github.com/IQSS/dataverse-jenkins ?
14:16 pdurbin
Let's create the job from the command line.
14:17 pdurbin
just like Pete does: https://github.com/sbgrid/data-capture-module/tree/0.6/jenkins
14:18 poikilotherm
I am trying to get it running in K3S
14:19 poikilotherm
Nearly there
14:19 poikilotherm
Then this should be easy :-)
14:20 poikilotherm
Hmm probably it should be "Almost there"
14:28 donsizemore
@poikilotherm it's beating the snot out of minikube on the same box. i take that as a good sign =)
14:29 donsizemore
@poikilotherm poopycakes: CrashLoopBackOff
14:29 poikilotherm
Logs plz
14:29 donsizemore
can't find secrets
14:30 poikilotherm
Did you create them?
14:30 donsizemore
yes
14:30 poikilotherm
O.o
14:31 poikilotherm
secretPlease double check with kubectl get
14:31 poikilotherm
seret
14:31 donsizemore
promsies i did.
14:31 poikilotherm
secret
14:31 donsizemore
i have 4 opaques and a default token
14:31 poikilotherm
Ok sounds good
14:32 donsizemore
this is in a fresh minikube on linux_x86_64
14:32 poikilotherm
Did you just use the commands I placed in README?
14:32 donsizemore
yes. unless you just updated the README and i hadn't reloaded the page
14:32 poikilotherm
Nope, that part did not change for a few weeks now... ;-)
14:32 donsizemore
no, i had current instructions
14:32 poikilotherm
Could you give me some more detailed logs?
14:33 donsizemore
sure thing. github issue, pastebin, e-mail?
14:33 poikilotherm
As you prefer
14:33 poikilotherm
Pastebin sounds like quick n easy
14:34 poikilotherm
No real secrets in there, so this should be alright
14:40 donsizemore
@poikilotherm i'm reclaiming my title, "Danger-prone Daphne of Dataverse" https://pastebin.com/BhU7BUBy
14:42 poikilotherm
Ok, secrets are alright
14:42 poikilotherm
Its just saying it doesn't find those for doi and rserve
14:43 poikilotherm
Hmm doi should be found, need to check on that. Might be a bug in my descriptor
14:43 donsizemore
i'm always ready to blame kubernetes
14:43 poikilotherm
Those Exceptions are normal :-(
14:43 poikilotherm
This is due to old stuff inside the WAR and scanners
14:44 poikilotherm
There has to be some other output from glassfish
14:44 poikilotherm
Otherwise start domain would not have failed
14:44 poikilotherm
Could you paste the full log? I'll have a look at it later or tomorrow
14:44 donsizemore
oh oh. further up i get errors about the doi and rserve password envs
14:45 poikilotherm
That's ok!
14:45 poikilotherm
Those are on me... Didn't mount them in the descriptor and I had no chance to test this yet
14:45 poikilotherm
Will add that
14:46 donsizemore
https://pastebin.com/ZZApZKtp
14:49 poikilotherm
Hmm this could be due to memory restrictions
14:49 poikilotherm
Because there are no errors I don't see on a successfull deployment
14:49 poikilotherm
Wil re-check tomorrow
14:49 poikilotherm
Cu guys, gotta pick up kids
14:49 donsizemore
this is a hand-me-down PC with 8 cores and 16GB of RAM ... could be memory.
14:49 donsizemore
thank you, have a great evening!
15:04 donsizemore
@pdurbin may i pick your institutional memory?
15:08 pdurbin
donsizemore: hit me
15:08 pdurbin
jonas47: welcome!
15:35 donsizemore
@pdurbin n/m. i'm assigning something the "unable to reproduce" looney tunes GIF and hiding from mandy
15:35 pdurbin
man, I should dig that gif up
15:35 jonas47
pdurbin: hey! thanks :D
15:36 jonas47
so i to get to "Candy Trade" in WholeTale, but it just shows something like 5% on the progress bar since hours
15:37 jonas47
so i ...tried... to
15:37 pdurbin
jonas47: bah. Oh, hey, real quick. Stuttgart or Göttingen? Please see http://irclog.iq.harvard.edu/dataverse/2019-04-08#i_90251
15:38 donsizemore
stuttgart!
15:38 jonas47
Göttingen!
15:38 donsizemore
@jonas47 mandy and thu-mai shared your experience. p.s. hi =)
15:39 jonas47
(i'm totally biased on this ---- my home town is like 20min from Göttingen, both my parents and both my siblings studied there, i lived there twice for about half a year each)
15:39 pdurbin
ah nice
15:39 pdurbin
xarthisius: any ideas on the 5% progress thing?
15:39 pdurbin
jonas47: what if you click on Browse and then back?
15:39 jonas47
stuttgart on the other hand is part of the evil conservative south of germany ;)
15:41 jonas47
browse & back leads me to the place where i can choose the environment - in both cases (jupyter and rstudio) i get to the 5% loading screen
15:44 pdurbin
Ok. You don't see "Candy Trade" under "Launched Tales" under Browse?
15:44 jonas47
"No tales have been launched"
15:44 pdurbin
:(
15:47 jonas47
so i have one major feedback on wholetale in general: it is an external tool and it is therefore not suitable for sensitive data - which feedback goes in which issue (dataverse-ansible vs. dataverse)?
15:48 jonas47
FYI: i used ORCID to sign up on WT
15:50 pdurbin
jonas47: I think you should open issue at https://github.com/IQSS/dataverse/issues called something like "feedback on Whole Tale external tool"
15:55 jonas47
ok! https://github.com/IQSS/dataverse/issues/5728
15:57 donsizemore
@jonas47 what if you could open wholetale (or other external tools) in a secure enclave?
16:01 xarthisius
there's a bug in the progress bar handling on .stage. Sorry about that, I said that's a stable environment, but it is a place where test things and sometimes things are broken
16:01 jonas47
that would need some sort of formal agreement with the external provider (https://gdpr-info.eu/art-28-gdpr/)
16:01 jonas47
Processing by a processor shall be governed by a contract or other legal act under Union or Member State law, that is binding on the processor with regard to the controller and that sets out the subject-matter and duration of the processing, the nature and purpose of the processing, the type of personal data and categories of data subjects and the obligations and rights of the controller.
16:01 jonas47
i don't think i could promote this feature if it has such requirements
16:02 jonas47
i'd rather have them use a local jupyterhub instance
16:02 jonas47
(local = on the local network)
16:02 xarthisius
jonas47: point of WT is that it's not a centralized service. You can run it locally, firewalled for your user. Just like JHub
16:03 xarthisius
we run common instance at .wholetale.org to show what it does
16:03 pdurbin
https://wholetale.readthedocs.io/en/stable/admin_guide/install.html
16:04 jonas47
okay, that was a misunderstanding then! thanks for clearing it up
16:04 xarthisius
we'll be adding "run on GKE" hopefully soon^{TM}
16:05 xarthisius
so that you're not bound to OpenStack
16:06 jonas47
so i must make one thing clear: i don't have any institutional backing when it comes to advanced features.... sadly, for most of my colleagues, the repository is just a place where one can put data files and have metadata with it
16:10 xarthisius
I'm trying to understand how are we incompatible with GDPR, is the fact that you need to login?
16:12 xarthisius
because we don't really keep the data in any way (sensitive or not)
16:12 xarthisius
well, it's being cached while user is operating on it
16:27 jonas47
i am not very knowledgeable on the GDPR, but in any case, i would not transfer any sensitive data (e.g. where school children were surveyed and are still identifiable) to an external service
16:30 pdurbin
jonas47: I think what you really want is a Trusted Remote Storage Agent (TRSA). donsizemore knows about this. You said "secure enclave". :)
16:31 jonas47
for hosting it ourself: do you see Slava's approach (Dataverse Store, 4tikhonov opened this Issue 12 days ago, https://github.com/IQSS/dataverse/issues/5688) as a place where wholetale could be included?
16:32 xarthisius
yeah, I don't know how to mitigate that. What we're doing is equivalent to a user wgetting a file somewhere. We unfortunately don't have an option to control what users are doing
16:32 pdurbin
jonas47: see "approved secure compute environment" in this diagram: http://cyberimpact.us/dataverse-trusted-remote-storage-agent-update/
16:32 xarthisius
i.e. if they're downloading files irresponsibly
16:33 jonas47
since we have dataverse installed on a local server, why would we need a TRSA?
16:34 pdurbin
Well, the idea is that the files are so sensitive that they are not stored on the Dataverse server itself.
16:36 jonas47
how would another server be more safe than our dataverse server? we have all sensitive data in-house and not with external providers (given the small size of the data overall, we don't need advanced stuff like load balancing an dthe like)
16:39 donsizemore
@jonas47 in the impact example above, dataverse would only track the metadata. the cyberimpact project is working on webapps to handle data access (SAFE/Notary) and allow access through a central web application.
16:41 donsizemore
@jonas47 odum is experimenting with Duke's Proconsul https://github.com/carte018/Proconsul to allow researchers vetted access to our workspace (it's really a front-end to RDP/VNC) but i could see us allowing externaltool targets in this space
16:59 pdurbin
donsizemore: wow, written in C.
16:59 pdurbin
fancy
17:02 pdurbin
xarthisius: is it bad if my tale ran all weekend? How's your Amazon bill?
17:04 pdurbin
I think I'm going to kill it (maybe you should add some hover text for the "x" to indicate if you call it "kill" or not) and try to launch it again.
17:04 pdurbin
ah, it means "delete"
17:16 xarthisius
pdurbin: it's running on Jetstream so it's paid from your taxes^H^H NSF ;)
17:16 pdurbin
\o/
17:17 pdurbin
I'm getting "Please wait".
17:17 pdurbin
Maybe it should say how long I should expect to wait.
17:17 pdurbin
Because I've had three other people try this now and they all want to know.
17:18 xarthisius
while importing? as I mentioned above there's a bug on .stage
17:18 pdurbin
while doing "Compose"
17:18 pdurbin
is compose the same as import?
17:19 xarthisius
almost, the only difference for import is that it adds data on the fly
17:19 pdurbin
oh
17:20 pdurbin
Should I wait until you fix the bug on .stage before I try this again? Or should I keep waiting?
17:20 jonas48 joined #dataverse
17:20 xarthisius
pdurbin: If you could wait a couple of days, that'd be great. We have 36-mo NSF review tomorrow and I'm kinda swamped with preparations
17:21 pdurbin
Sure! I can certainly wait. Do you want to just ping me when you want me to give it another try?
17:22 xarthisius
I'll certainly do that
17:23 pdurbin
perfect, thanks!
17:26 jonas48
on my point earlier about the dataverse store: there is no way i can get the institutional it department to run a complete suite of software for research data. but if all was integrated in a store, where i can easily add components like whole tale and keep them up to date, that would be a whole different story
17:27 pdurbin
jonas48: makes sense. I would suggest creating issues in https://github.com/IQSS/dataverse-kubernetes/issues and https://github.com/IQSS/dataverse-docker/issues saying that you'd like this or that tool to be installed.
17:28 donsizemore joined #dataverse
17:29 jonas48
i like this approach: https://github.com/IQSS/dataverse/issues/5688
17:34 jonas48
sorry, i have to leave now! have a nice day :)
17:34 xarthisius
pdurbin: I'd like to note one thing with your test setup: whatever I'm doing it tries to enforce https
17:34 xarthisius
I can reproduce that outside of WT on my laptop
17:34 xarthisius
with just requests.get('http://dev2.dataverse.org/api/access/datafile/29')
17:36 pdurbin
yes, that sounds right, I'm forcing https... but I'm not sure why that matters
17:38 xarthisius
when I do http or https in python it just dies with bad handshake
17:39 pdurbin
hmm, I wonder if it's because I took out the intermediate cert. I wasn't sure if I needed it. This was late on Friday.
17:46 donsizemore
@pdurbin i'm in a meet^H^H^H^H have some quiet time and was going to start documenting jenkins. before I place an initial README, you want the repo to remain dataverse-jenkins?
17:46 pdurbin
Yeah, I'm happy with the name. I think others might try to run it some day.
17:47 pdurbin
I was thinking...
17:47 pdurbin
Did you create the first job in the GUI or by the command line?
17:52 donsizemore
i started with the GUI ; I was going to export it in XML format
17:55 pdurbin
Cool. I've never created a Jenkins job from the command line and I'm eager to learn.
17:56 pdurbin
It seems like we're always going to want "develop" from IQSS so we could automate the addition of that job in Vagrant, right?
18:15 donsizemore joined #dataverse
18:45 poikilotherm joined #dataverse
18:49 pdurbin
poikilotherm: a vote for Göttingen: http://irclog.iq.harvard.edu/dataverse/2019-04-08#i_90390
18:49 poikilotherm
Hi pdurbin :-)
18:50 pdurbin
donsizemore: shoot, are you saying Thu-Mai and Mandy tried Whole Tale too and it didn't work for them?
18:50 donsizemore
@pdurbin they got the long spinning wheel (or page load, or whatever it was)
18:52 pdurbin
bah, ok. sorry. I'll let you know when they should try again
18:54 pdurbin
donsizemore: we talked about sample data this morning. I praised you up and down.
18:55 pdurbin
poikilotherm: oh, and I suggested that perhaps you'd be receptive to adding extra stuff to dataverse-kubernetes... stuff like Data Capture Module, if you're familiar with it.
18:55 pdurbin
I'm not sure if dataverse-ansible is the right place for DCM.
18:56 poikilotherm
Sure!
18:56 poikilotherm
Just open an issue so things can be tracked
18:56 poikilotherm
I am open to pull requests
18:56 pdurbin
Cool. When Pete gets back maybe I'll ask him to open an issue.
18:57 poikilotherm
Maybe things like TRSA could get Dockerized too
18:57 poikilotherm
But as you say it: small steps...
18:57 pdurbin
sure, that's donsizemore's department
18:58 donsizemore
@poikilotherm TRSA is under heavy development(R) but a decently current Dockerfile is at https://github.com/OdumInstitute/trsa-web/blob/jee8line/Dockerfile
18:58 donsizemore
@poikilotherm it may look familiar =)
18:59 poikilotherm
LOL
18:59 poikilotherm
Indeed
18:59 poikilotherm
But why did you fork instead of basing on it
20:16 poikilotherm joined #dataverse
20:51 donsizemore joined #dataverse
20:52 donsizemore
@poikilotherm the trsa-web is a separate project from dataverse (with different requirements) so i snitch a copy of your (excellent) work to get a testable sample up and running
20:53 donsizemore
@poikilotherm if there's a smarter way you'd like for me to do that (or, now that i look at it, give you credit for the Dockerfile) i'm happy to do so
20:54 poikilotherm
Well you could simple use the upstream payara image in the FROM?
20:54 poikilotherm
And add your special things below
20:54 poikilotherm
Then you don't have to fiddle with all the stuff like tini etc
20:54 poikilotherm
Should drastically speed up build time
20:55 poikilotherm
Please don't give credit to me
20:55 poikilotherm
I signed a CLA and all credits go to Payara
20:56 donsizemore
we started with payara 4, now that we're on 5 i'll look at the upstream
20:56 pdurbin
wow, you have Dataverse running on Payara 5?
20:56 donsizemore
akio's trsa webapp
20:56 pdurbin
oh
20:57 donsizemore
we were betting that by the time it gets used anywhere dataverse 5 will be out
20:57 pdurbin
I was gonna say, https://github.com/IQSS/dataverse/issues/4172 is assigned to poikilotherm :)
20:57 donsizemore
and @poikilotherm i'll give you credit if i want to =)
20:58 poikilotherm
LOL
20:58 poikilotherm
Ok
20:58 poikilotherm
As you prefer :-D
20:58 poikilotherm
Don't say I didn't warn you when they sue you ;-)
21:01 pdurbin
They seem so nice. The founder is asking nicely about Glassfish's 3340 open issues: https://www.eclipse.org/lists/glassfish-dev/msg00046.html
21:08 donsizemore
@poikilotherm but it's apache 2.0!
21:08 poikilotherm
Yes it is. It was just a joke
21:08 donsizemore
so would be suing a state employee such as myself
21:09 poikilotherm
Sry, I didn't mean to get people crazy about possible consequences when using OSS
21:09 poikilotherm
As you said donsizemore - it is Apache 2.0, so feel free :-D
21:17 donsizemore
p.s. @pdurbin https://jenkins.dataverse.org/job/IQSS-develop/lastSuccessfulBuild/
21:28 pdurbin
Our first three builds all passed? Hooray! \o/
21:29 pdurbin
But should we call it IQSS-develop? What about dataverse-client-java for example? Please see https://github.com/IQSS/dataverse-client-java/issues/5
21:30 pdurbin
Should it be IQSS-dataverse-develop?
21:31 pdurbin
Huh, look at all these people: https://jenkins.dataverse.org/asynchPeople/ ... great!
21:40 donsizemore
i can rename it
21:41 pdurbin
thanks!
21:41 pdurbin
does having the org name first seem best?
21:43 donsizemore joined #dataverse
21:44 donsizemore
hey @pdurbin we just lost power again... i think it's time for me to head home.
21:44 pdurbin
absolutely
21:44 pdurbin
thank you!
