Time
S
Nick
Message
07:14 jri joined #dataverse
07:34 MrK joined #dataverse
10:16 juancorr joined #dataverse
12:07 drew-jhu joined #dataverse
12:43 donsizemore joined #dataverse
13:06 MrK
pdurbin: Hi, I have a question regarding adding new Permission.
13:13 MrK
ah nvm I figured it out :P
13:14 pdurbin_m joined #dataverse
13:14 pdurbin_m
phew
14:01 stefankasberger joined #dataverse
14:02 stefankasberger
Greetings from the Dataverse Workshop at DANS. :)
14:03 pdurbin
stefankasberger: I'm so jealous!!!!1!!1!
14:04 pdurbin
I hope you're having a great time. :)
14:05 stefankasberger
We do. Slava tells funny stories about the Dataverse community. :)
14:06 pdurbin
Uh oh. Did he tell you about the salsa dancing?
14:07 stefankasberger
We will ask him that later during Dinner. ;)
14:08 stefankasberger4 joined #dataverse
14:08 pdurbin
Ok. Ask him about the Dataverse Cup too. :)
14:12 pdurbin
How many of you are coming to the Dataverse Community Meeting in June?
14:15 drew-jhu
hi, @pdurbin. I was wondering if you have a minute to discuss the reindexing issue (#5575)
14:18 pdurbin
Sure. I think I asked you if you can provide a way to reproduce it.
14:20 drew-jhu
you did indeed. while the server i can reporduce the problem on is a non-production server, the data is just a backup of our production database, so before sharing it, I'd need to see about cleaning up any non-public PII it contains.
14:22 pdurbin
That's what I was afraid of.
14:22 drew-jhu
do you know if anyone has any SQL queries for anonymizing dataverse databases or documentation about which tables & fields contain potentially problematic data?
14:23 pdurbin
I don't put please add the request to the Google doc linked from this "useful SQL queries" issue: https://github.com/IQSS/dataverse/issues/4169
14:24 pdurbin
So you have 24 datasets that are failing indexing? Is there any pattern?
14:28 drew-jhu
not that I can see on casual observation. TBH, i haven't really investigated yet. the issue was blocking my user acceptance testing of our upgrade to 4.12, and thus my queue of other tasks on other projects. i found the issue and the workaround and applied it just yesterday afternoon.
14:31 drew-jhu
the queries in that google doc are great. very useful. should totally be in a github repo, though
14:31 pdurbin
I wonder if it's failing "bean validation": https://github.com/IQSS/dataverse/issues/5753
14:32 pdurbin
On second thought, that probably doesn't make sense because I guess you wouldn't be able to index them individually.
14:41 drew-jhu
:shrugs: the errors in the log were plentiful, but not terribly informative to one without a lot of familiarity with the codebase, such as myself
14:42 pdurbin
Can you please post as much of the stacktrace as you feel comfortable with into the issue? And/or email as much of server.log as you can to support
14:45 drew-jhu
I can, but i'm afraid it'll have to be next week. i'm overdue on some other tasks. i have asked the folks in charge of our dataverse about sharing an anonymized backup of the db, as well. i can get back to you about that once i hear back from them
14:52 pdurbin
Sounds great. For what it's worth, I'm personally tracking that issue (#5575) at https://trello.com/b/YI9EbqGz/backlog-grooming-for-dataverse-pdurbin-edition :) It has an estimate at least. A "3".
14:54 drew-jhu
out of 10?
15:03 pdurbin
Our estimation system is mysterious. :)
15:03 pdurbin
it's not out of 10
15:03 pdurbin
technically, you can throw an infinity at estimation poker
15:03 pdurbin
or a question mark
15:05 drew-jhu
ah. ok. i can see your trello, but not the estimates
15:05 pdurbin
You have to go to Waffle to see the estimates: https://waffle.io/IQSS/dataverse
15:05 drew-jhu
so 3 is maybe "non-trivial, but not huge either"?
15:06 pdurbin
lot of 3's right now. one 5, a couple 1's
15:06 pdurbin
I can remember the last time we had an 8.
15:06 pdurbin
We'd probably break a would be 8 up into smaller chunks.
15:06 pdurbin
moves across the board faster
15:07 pdurbin
speaking of which, standup in 10 minutes
15:07 drew-jhu
ok. makes sense & good to know. i'm going to go knock out some tasks, so i can hopefully help out with this next week
15:08 pdurbin
Perfect. Want me to assign the issue to you? :)
15:09 drew-jhu
heh. def not. most i can realistically offer is sharing logs & testing with a hope of maybe writing the anonymization script for the backup
15:10 stefankasberger joined #dataverse
15:10 pdurbin
those are all fantastic contributions and would be very much appreciated!
15:10 drew-jhu
i will do my best. ttys
15:12 pdurbin
o/
16:08 jri joined #dataverse
17:09 donsizemore joined #dataverse
17:59 nprasai joined #dataverse
18:29 pdurbin
xarthisius: you were totally right about the bad cert on dev2. Sorry. I just fixed it. I put the intermediate cert back in. I guess I do need it.
18:38 donsizemore
@pdurbin knock knock?
18:38 pdurbin
talk to me
18:38 donsizemore
so, i see a bunch of problems editing dataset metadata, when it's been submitted for review. but all those fixes went into 4.9.3 and we're seeing this on 4.9.4
18:39 donsizemore
just checking your institutional memory
18:39 pdurbin
only when it's in review?
18:40 donsizemore
when she pulled it back we got "Dataset Submission Return Failed - Command edu.harvard.iq.dataverse.engine.command.impl.ReturnDatasetToAuthorCommand
18:40 donsizemore
only thing in server.log are a ton of "autoUpdate attribute is deprecated" warnings
18:41 donsizemore
oh, re-reading my initial description. i see a bunch of ISSUES concerning editing dataset metadata, but all fixes point to 4.9.3 and we're on 4.9.4
18:44 pdurbin
The one that sticks in my mind is https://github.com/IQSS/dataverse/issues/4139 but that was way older... fixed in 4.8.2.
18:44 pdurbin
seeing that "null" there is never good. The command should give a proper error message
18:47 pdurbin
dataset.reject.failure=Dataset Submission Return Failed - {0} at https://github.com/IQSS/dataverse/blob/v4.8.2/src/main/java/Bundle.properties#L1132
18:48 donsizemore
things look right in the database as best i can tell
18:48 pdurbin
here's where the error comes from: https://github.com/IQSS/dataverse/blob/v4.8.2/src/main/java/edu/harvard/iq/dataverse/DatasetPage.java#L1718
18:48 pdurbin
sendBackToContributor method
18:49 donsizemore
she can edit other such datasets, just not this one
18:49 pdurbin
you said "when she pulled it back" but do you mean "when she clicked return to author"?
18:50 donsizemore
she got Dataset Submission Return Failed - Command edu.harvard.iq.dataverse.engine.command.impl.ReturnDatasetToAuthorCommand
18:50 donsizemore
but now the dataset is no longer in review
18:50 donsizemore
she still can't edit it
18:51 pdurbin
hmm
18:52 pdurbin
I'd suggest getting a ticket rolling by emailing support
18:52 donsizemore
ah, okay. have a great trip!
18:52 pdurbin
thanks, still some snow in northern VT
18:54 pdurbin
I assume in server.log you see "sendBackToContributor: " followed by a message. What does the message say?
19:05 donsizemore
javax.ejb.TransactionRolledbackLocalException: Exception thrown from bean at com.sun.ejb.containers.EJBContainerTransactionManager.checkExceptionClientTx(EJBContainerTransactionManager.java:662)
19:06 pdurbin
ooooh
19:06 donsizemore
(at edu.harvard.iq.dataverse.DatasetPage.sendBackToContributor(DatasetPage.java:1784)) is buried in the middle
19:06 pdurbin
can you please email your server.log to that support address?
19:06 donsizemore
no psql errors, just transaction rollbacks
19:09 pdurbin
oh jeez, that link above was for the wrong version. I had 4.8.2 on the brain
19:10 pdurbin
https://github.com/IQSS/dataverse/blob/v4.9.4/src/main/java/edu/harvard/iq/dataverse/DatasetPage.java#L1784 is where we send ReturnDatasetToAuthorCommand to the command engine.
19:11 pdurbin
I'd need more stack trace though.
19:11 pdurbin
I know they're long but more is more. :)
19:28 donsizemore
i sent them the entire log
19:30 donsizemore
p.s. why are you on IRC instead of skiing?
19:31 pdurbin
oh, we're closing on a new (well, 80s) house tomorrow, we don't drive to VT until saturday
19:32 donsizemore
!!!
19:33 pdurbin
man, this is a lot of "autoUpdate attribute is deprecated" junk
19:35 pdurbin
Caused by: javax.persistence.PersistenceException: javax.validation.ConstraintViolationException: Bean Validation constraint(s) violated while executing Automatic Bean Validation on callback event:'preUpdate'. Please refer to embedded ConstraintViolations for details.
19:38 pdurbin
at edu.harvard.iq.dataverse.DataverseRoleServiceBean.rolesAssignments(DataverseRoleServiceBean.java:212)
19:38 pdurbin
Did we add some bean validation constraints on users or groups?
19:40 pdurbin
https://github.com/IQSS/dataverse/blob/v4.9.4/src/main/java/edu/harvard/iq/dataverse/DataverseRoleServiceBean.java#L212
19:51 pdurbin
https://github.com/IQSS/dataverse/blob/v4.9.4/src/main/java/edu/harvard/iq/dataverse/engine/command/impl/ReturnDatasetToAuthorCommand.java#L54 is talking about reviewers.
19:51 pdurbin
people with Permission.PublishDataset permission
19:51 pdurbin
donsizemore: what if you remove some reviewers?
19:52 pdurbin
it seems like it's blowing up due to a bean validation constraint being violated on a reviewer
20:20 pdurbin
larsks: hey, I'm going to spin down that ec2 instance from yesterday. I hope you don't mind.
20:21 donsizemore
@pdurbin i'll let mandy do that -- there is a long list of reviewers
20:24 donsizemore
@pdurbin mandy points out that the record has been through 3 resubmits and this is the first error, and other datasets don't throw the error. shrug
20:26 pdurbin
Are there different reviewers on this dataset? A long list, you say?
20:37 pdurbin
larsks: terminated.
20:38 pdurbin
donsizemore: hopefully, someone will reply on the ticket you opened soon. I'm out. I'll be back on Monday the 22nd.
20:38 pdurbin
Have a good weekend, everyone. It starts early for me. :)
20:38 pdurbin left #dataverse
20:46 jri joined #dataverse
21:58 pdurbin joined #dataverse
21:58 pdurbin left #dataverse
23:27 andrewSC joined #dataverse
