Time
S
Nick
Message
07:51 jri joined #dataverse
08:18 Youssef_Ouahalou joined #dataverse
08:40 jri_ joined #dataverse
10:04 Benjamin_Peuch joined #dataverse
11:48 donsizemore joined #dataverse
11:57 donsizemore
@pdurbin Shib-3.1 ahoy https://wiki.shibboleth.net/confluence/display/SP3/ReleaseNotes#ReleaseNotes-3 .1.0(April13,2020)
12:53 donsizemore
@pdurbin also: https://jenkins.dataverse.org/job/IQSS-Dataverse-Develop-PR/view/change-requests/job/PR-6822/1/testReport/
14:16 pdurbin
donsizemore: thanks for keeping an eye on shib releases
14:16 donsizemore
I got a non-answer from UNC's Identity Management so Odum will sit tight at 3.0.4 until there's a security release
14:19 pdurbin
ok
14:20 pdurbin
I'm confused by that testReport. No failures is good but it looks like the API tests didn't run.
14:20 donsizemore
oh, poop. let me look.
14:22 donsizemore
it's pulling from vagrant.yml, where they are indeed false. part of #6817 switches that to jenkins to match the -develop job.
14:22 donsizemore
in the mean time i can turn the suite back on in vagrant and re-run that job.
14:24 donsizemore
re-running.
14:26 pdurbin
Oh, it was set to false? Sounds like an easy fix. Here's an example that makes me happy. No test failures *and* the API test took 2 min 57 seconds (see "edu.harvard.iq.dataverse.api"): https://jenkins.dataverse.org/job/IQSS-dataverse-develop/403/testReport/
14:29 donsizemore
I created a jenkins.yml to pare back some of the stuff in vagrant.yml
14:30 donsizemore
speaking of, i was going to turn off prometheus and grafana in jenkins to save the CPU cost, if that's cool by you. we can turn it back on once we get the configs mapped out?
14:32 pdurbin
Sure, that sounds fine. For production installations I think Prometheus and friends are awesome but for testing via Jenkins I don't see a need for it, especially if it's expensive.
14:33 donsizemore
it's tracking basic system usage now (cpu/heap) but i thought we might do more with it than we have.
14:36 donsizemore
p.s. what i'm doing this morning, in case you want to redirect me: i've stood up payara5 with @poikilotherm's proposed s3 change to pom.xml, and am using that instance to step through localstack installation for dataverse-ansible #140
14:37 pdurbin
Oh sure. There's a MicroProfile Metrics API we can tap into for more Prometheus goodness once we get onto an app server that supports it, like Payara 5.
14:38 pdurbin
donsizemore: what you're doing sounds absolutely fantastic. Thank you!
14:39 pdurbin
I'm trying to listen to the design meeting so apologies if I'm a little scattered.
14:42 donsizemore
wooden-paneled vaulted ceilings are in now, i hear.
15:12 donsizemore
i'm 'bout to kick archive.apache.org to the curb [dataverse : download and unzip solr] ************************************* An exception occurred during task execution. To see the full traceback, use -vvv. The error was: SSLError: ('The read operation timed out',)
15:17 pdurbin
bah
15:18 pdurbin
The design meeting is over. No discussion of wood paneling.
15:26 pdurbin
poikilotherm: typo: "Chosse" in docs/day1/init-deploy.rst
15:26 pdurbin
https://github.com/IQSS/dataverse-kubernetes/compare/63-payara is looking great!
15:29 poikilotherm
Heh. Good catch. Can you add a comment?
15:29 poikilotherm
Maybe one of those nice +- again?
15:30 poikilotherm
I'm trying to get a Dev flavor for payara, too
15:32 poikilotherm
But I can't figure out how to enable JDWP via preboot script...
15:32 poikilotherm
And trying to change the default JMX config is also driving me nuts. Nothing works...
15:32 pdurbin
I don't think I can add a comment because it isn't a pull request.
15:33 poikilotherm
Hu? But I opened one...
15:33 poikilotherm
#118 sth?
15:33 poikilotherm
IIRC
15:34 poikilotherm
Or you could leave a comment on the commit...
15:34 poikilotherm
Hmm. On or at the commit?
15:39 pdurbin
Oh, you're right, there is a pull request.
15:42 pdurbin
ok, done
15:42 poikilotherm
Thx!!
15:42 poikilotherm
Maybe you can take a look at https://github.com/IQSS/dataverse/pull/6818#issuecomment-613865446 and leave a comment there, too?
15:44 pdurbin
Yeah, I saw that. No objections but I'd wait to see what the reviewer says. We can always change it later.
15:44 pdurbin
What would be a neat change some day is changing all the code from edu.harvard.iq to org.dataverse. :)
15:45 poikilotherm
There is no reviewer yet. Scolapasta sent me a mail that this looks good to him and moved it from a draft to a normal pr
15:45 poikilotherm
And djbrooke didn't say anything about a release note... :-/
15:46 pdurbin
Ah we use "assignee" to mean reviewer.
15:46 pdurbin
Assigned to leave a review.
15:46 poikilotherm
That namespace change would be a good candidate for the formatting cleanup, wouldn't it?
15:47 poikilotherm
Ah right, djbrooke assigned Gustavo
15:47 poikilotherm
But Gustavo did a non slopi reply :-D
15:47 pdurbin
Woof. I dunno. We need tooling in place before we do the formatting cleanup. Otherwise, it will drift back into chaos.
15:55 pdurbin
poikilotherm: oh, are you still there? Question about the containers page in the dev guide.
16:01 poikilotherm
I'm here, following you while building drywalls
16:03 pdurbin
Ok, I just wanted to put on your radar that in https://github.com/IQSS/dataverse/pull/6816 I put a line at the top of the containers page in the dev guide that we are open to someone deleting what's on that page and the associated openshift file in order to start over with a clean slate.
16:08 poikilotherm
Oh yeah I saw that :-)
16:11 poikilotherm
Here's a crazy idea. What about creating a PR at dataverse-k8s with the old content added to the current docs. Then delete it from the main guide, pointing to the project, the openshift issue (can't remember bthe number from the top of my head) and the PR, adding a statement about this content has been moved? The PR could even reference the already existing issue as related.
16:12 poikilotherm
I don't have capacity to work on this NOW, but that way nothing is lost.
16:14 pdurbin
Oh. Well, all that openshift content is in lots of old versions of the guides and in the git history. So I'm not worried about anything being lost. I would just point someone who wants to work on openshift stuff at the 4.20 guides or whatever.
16:15 pdurbin
That said, if you *want* all that content in the dataverse-k8s guides you are welcome to have it. It's on my mind because it references Glassfish 4 and as you well know, lots it changing (finally!) in that area. So the openshift stuff is out of date.
16:16 donsizemore
@pdurbin https://jenkins.dataverse.org/job/IQSS-Dataverse-Develop-PR/view/change-requests/job/PR-6822/3/testReport/
16:24 pdurbin
donsizemore: now *that* brings me joy. Even if there are a couple test failures. :) Thanks!!
16:27 pdurbin
Oh, by the way, I've switched to Payara for dev. I hope I don't have to switch back to Glassfish 4. :)
16:28 donsizemore
oh, it sounds like i should make arrangements to implement gustavo's patch in dataverse ansible? he said he was going to submit the PR to put it beneath installer/ correct
16:30 pdurbin
He committed it to the installer branch yesterday: https://github.com/IQSS/dataverse/blob/40076236ae3783929efa4a52e728db5c9cae6c05/scripts/installer/jsfpatch/jakarta.faces_dv.jar
16:32 donsizemore
coo
16:45 poikilotherm
Maybe scolapasta could ask Payara people to include his Mojarra patch
16:46 poikilotherm
They already ship a patched version, 2.3.14.payara-p1
16:46 poikilotherm
Looks like it didn't go into an upstream release yet...
17:23 donsizemore
@poikilotherm gustavo has been keeping a keen eye on it
17:33 pdurbin
Yeah, it would be nice not to have to patch Payara ourselves.
17:49 donsizemore
I went ahead and added it to the dataverse-ansible files/ dir. an easy switch to throw and I'm ready for Payara now
17:49 pdurbin
nice
17:50 pdurbin
all my ec2 scripts are broken on my mac, all the stuff that shells out to aws, python woes basically, not sure how to fix it
17:59 donsizemore
@pdurbin weren't they going to get you a new Mac? ;)
17:59 pdurbin
yeah, it's being ordered :)
18:00 donsizemore
@pdurbin also, just ran the API test suite on Payara 5 with Gustavo's patched JSF: Results:", "[INFO] ", "[WARNING] Tests run: 117, Failures: 0, Errors: 0, Skipped: 4"
18:00 pdurbin
Are you joining us for sprint planning?
18:00 donsizemore
@pdurbin but that's stock 4.20, not develop.
18:00 donsizemore
we're sprint planning?
18:01 pdurbin
see #dv-standup
18:01 donsizemore
@pdurbin private channel, no link preview
19:02 pdurbin
donsizemore: weird. #dv-standup doesn't have a lock next to it
21:22 pdurbin left #dataverse
