Time
S
Nick
Message
01:47 djbrooke joined #dataverse
05:31 bjonnh
pdurbin: unless that's a bug that made your project blow itself up…
05:31 bjonnh
pdurbin: ;)
05:35 bjonnh
pdurbin: http://perspectives.mvdirona.com/2017/04/at-scale-rare-events-arent-rare/ < and then you have interesting bugs
06:56 garnett joined #dataverse
06:59 andrewSC joined #dataverse
07:20 axfelix joined #dataverse
07:36 axfelix joined #dataverse
07:38 axfelix joined #dataverse
07:52 axfelix joined #dataverse
09:26 axfelix joined #dataverse
09:59 axfelix joined #dataverse
11:24 donsizemore joined #dataverse
11:43 axfelix joined #dataverse
12:16 pdurbin
huh, that is an interesting bug
13:29 andrewSC joined #dataverse
13:57 axfelix joined #dataverse
14:04 djbrooke joined #dataverse
14:10 djbrooke joined #dataverse
14:15 djbrooke joined #dataverse
14:19 djbrooke joined #dataverse
14:20 djbrooke joined #dataverse
14:38 pameyer joined #dataverse
16:16 pdurbin
knikolla: great meeting this morning! I just typed up note and action items into the same doc from yesterday. Please let me know if I missed anything. I gave you edit access so you are welcome to make corrections directly: https://docs.google.com/document/d/1JrK0pSA0eAxOkYYLC68AI_uu-DxJ_R0iEnekZKUMbig/edit?usp=sharing
16:16 knikolla
pdurbin: awesome! thanks for being so organized!
16:17 knikolla
pdurbin: i was going to look at sarah's pr, about swift. but it looks like rebases are making github show 500 files changed
16:56 pdurbin
ick
16:58 pameyer
pdurbin: do you know if there were note's from the code review on Sarah's PR?
16:58 pameyer
might be helpful for knikolla if there were
17:00 pdurbin
pameyer: I took some notes but I didn't put them anywhere.
17:01 pameyer
same with me - and most of my notes were about questions to ask
17:05 pdurbin
pameyer: Sarah just sent me the list Danny wrote up and it matches mine.
17:07 pdurbin
hmm, except there's nothing about plain text versions of tabular files
17:07 pdurbin
derivative files from a Stata file or whatever
17:12 pameyer
that was one of the things I was thinking would be a problem for compute access
17:31 donsizemore joined #dataverse
17:48 pdurbin
pameyer: right, you'd rather not have them there
17:48 pdurbin
But from the Dataverse perspective, the derivative files are still files and can be downloaded.
17:48 pdurbin
I dunno.
17:49 pdurbin
donsizemore: any news from Jan?
17:56 djbrooke joined #dataverse
17:58 djbrooke
FYI, there were notes from the code review in the PR here: https://github.com/IQSS/dataverse/pull/3723
17:59 pameyer
djbrooke: thanks
18:00 donsizemore
@pdurbin not yet, but I suspect a configuration difference between UNC and Harvard
18:01 pdurbin
yeah, me too. thanks
18:29 djbrooke joined #dataverse
18:31 donsizemore joined #dataverse
18:31 donsizemore
@pdurbin another thing i'd forgotten from last summer: "I would use SSO tracer in the browser to examine the SAML authentication requests and responses and watch the behavior of the IdP (whether or not it prompts for username/password). It is possible that SPs are making “ForceAuthn” requests which always prompt for username password, or that Harvard’s IdP session is shorter than UNC’s or that their IdP clustering does not use session
18:34 donsizemore
@pdurbin so heading for HathiTrust I see Cookie: _ga=GA1.2.692370765.1491322536; _saml_idp=dXJuOm1hY2U6aW5jb21tb246dW5jLmVkdQ%3D%3D; MDPsid=61f17f00864fbb3d1bb4ae2027f0029b; HTstatus=%7B%22authType%22%3A%22shibboleth%22%2C%22institution%22%3A%22unc%22%2C%22affiliation%22%3A%22University%20of%20North%20Carolina%20at%20Chapel%20Hill%22%2C%22displayName%22%3A%22dls%22%2C%22u%22%3A%22%22%7D; has_js=1; HT.prefs=%7B%22sdrinst%22%3A%22unc%22%7D; _sh
19:12 djbrooke joined #dataverse
19:17 djbrooke joined #dataverse
19:36 axfelix joined #dataverse
19:37 axfelix joined #dataverse
19:43 pdurbin
donsizemore: so look for ForceAuthn. Ok. Thanks!
21:18 donsizemore joined #dataverse
21:19 donsizemore
@pdurbin on shib: i think shibboleth just has to run on the same machine(s) as mod_shib (apache/proxy) but don't quote me on that
21:19 donsizemore
@pdurbin i get into trouble when people quote me on things
21:20 djbrooke joined #dataverse
21:23 djbrooke joined #dataverse
21:24 djbrooke joined #dataverse
21:25 djbrooke_ joined #dataverse
22:40 pdurbin
knikolla: I just emailed this to you but I think you're suffering from https://github.com/IQSS/dataverse/issues/3446 - "Stop distributing TestShib IdP file with old cert, better guidance on how to configure specific Shibboleth identity providers" ... the "opensaml::FatalProfileException" error, "Message was signed, but signature could not be verified"
22:51 djbrooke joined #dataverse
