Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

User studies: When an install fails, still seeing a status of "installing" #38

Open
SheenaTomko opened this issue Jul 5, 2022 · 8 comments
Assignees
Labels
bug Something isn't working high priority pause

Comments

@SheenaTomko
Copy link

See slack thread: https://epvb.slack.com/archives/CBEA7PQDQ/p1657045518138049

Basically, I tried to upload the PRISM_cohort_Participants.txt file as a user study. I received no error message on import or install, but even though the logs showed the install failed within a few min, 40 min later I still saw the status as "installing". This ticket is specifically to address the bug where an install failed but was not reported as a failure.

Screen Shot 2022-07-05 at 1 21 20 PM

@dmgaldi
Copy link
Contributor

dmgaldi commented Jul 5, 2022

Error from Logs https://ws.qa.apidb.org/job/irods-handler-ClinEpiDB/76/label=irods-north/console:

[05/Jul/2022|13:41:51] WARN  - org.gusdb.wdk.model.user.dataset.event.UserDatasetEventSync:153 - Event processing failed for event 165704282300012765
org.gusdb.wdk.model.WdkModelException: Failed running command: singularity run --bind /var/tmp/wdk/ud_22345_10573499135416265678:/work --bind /usr/local/home/joeuser/jenkins-irods/label/irods-north/gus_home/config/ClinEpiDB/gus.config:/gusApp/gus_home/config/gus.config --bind /u01/app/oracle/product/12.2.0.1/db_12/network/admin:/opt/oracle/instantclient_21_6/network/admin docker://veupathdb/dataset-installer-isasimple:latest loadStudy.bash /var/tmp/wdk/ud_22345_10573499135416265678/PRISM_cohort_Participants.txt 22345 /var/tmp/wdk/ud_22345_10573499135416265678/tmp-meta.json . For details, see /var/tmp/wdk/ud_22345_10573499135416265678/stderr
	at org.gusdb.wdk.model.user.dataset.UserDatasetTypeHandler.runCommand(UserDatasetTypeHandler.java:164) ~[wdk-model-1.0.0.jar:?]
	at org.gusdb.wdk.model.user.dataset.UserDatasetTypeHandler.installInAppDb(UserDatasetTypeHandler.java:117) ~[wdk-model-1.0.0.jar:?]
	at org.gusdb.wdk.model.user.dataset.event.UserDatasetEventSyncHandler.handleInstallEvent(UserDatasetEventSyncHandler.java:223) ~[wdk-model-1.0.0.jar:?]
	at org.gusdb.wdk.model.user.dataset.event.UserDatasetEventSyncHandler.handleInstallEvent(UserDatasetEventSyncHandler.java:145) ~[wdk-model-1.0.0.jar:?]
	at org.gusdb.wdk.model.user.dataset.event.UserDatasetEventSync.handleEventList(UserDatasetEventSync.java:120) [wdk-model-1.0.0.jar:?]
	at org.apidb.irods.ProcessIrodsEvents.syncMode(ProcessIrodsEvents.java:161) [eupath-irods-controller-1.0.0.jar:?]
	at org.apidb.irods.ProcessIrodsEvents.main(ProcessIrodsEvents.java:105) [eupath-irods-controller-1.0.0.jar:?]

Event 165704282300012765 seems to be marked as failed in database according to @steve-fischer-200

@jtlong3rd
Copy link
Contributor

Just took a look at the client code. It seems like for uninstalled UDs, we always display “installing” if less than four hours have passed since the UD’s creation. If more than four hours have passed, we display an “error.”

@jtlong3rd
Copy link
Contributor

jtlong3rd commented Jul 5, 2022

It seems like the existing UD response doesn’t provide an “error” field for UDs which failed to install. Not sure why.

@danicahelb danicahelb added the bug Something isn't working label Jul 5, 2022
@dmgaldi dmgaldi transferred this issue from VEuPathDB/EdaNewIssues Jul 7, 2022
@jtlong3rd
Copy link
Contributor

Paused while UI/Infra assesses a rehaul of the User Datasets system, which would likely obviate the need to fix this big

@danicahelb
Copy link

@jtlong3rd are we ready to take another look at this ticket? please close it if the rehaul fixed the issue. thanks!

@dmfalke
Copy link
Member

dmfalke commented Sep 14, 2022

I spoke with Jamie about this issue. The backend currently uses IRODs, which is, in part, responsible for the issue. The backend will need to migrate away from IRODs to address this issue. I don't know when that might happen, but definitely not any time soon.

@danicahelb
Copy link

danicahelb commented Sep 14, 2022

ok, i am moving this out of b60. thanks
i added a high priority label so that it stays on our radar

@aurreco-uga
Copy link
Member

this will be reviewed in the context of the new VDI

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working high priority pause
Projects
None yet
Development

No branches or pull requests

6 participants