search

sciencehistory / export_archivesspace_xml

Export our archivesspace EADs to S3.
Other
0 stars 0 forks source link

UPenn diagnostic "endpoint" page is unreachable #39

Open eddierubeiz opened 2 days ago

eddierubeiz commented 2 days ago

The page UPenn uses to show us whether or not its EAD harvest worked successfully is currently unreachable.

Note that our export to S3 is working fine. We just don't know whether UPenn is harvesting those files from s3.

Because we check this page's status every day as part of our export workflow (see the heroku scheduler for the project ), the nightly task is not reporting to HoneyBadger, which in turn is letting us know it's not receiving the nightly check-ins it's expecting.

eddierubeiz commented 2 days ago

I notified Sarah N. and Holly Jo Mengel at UPenn. Will update the ticket as needed.

jrochkind commented 2 days ago

Is that UPenn page a specifically created for us “diagnostic” page? I wonder if we coudl actually check a live public page of some kind instead? They don’t have much incentive to notice when a just-for-us diagnostic page goes down, or to fix it.

Can we also (or do we already have) automated monitoring that actually checks our OWN endpoints are HTTP accessible with expected content?

A custom-for-us diagnostic page on THEIR end does not seem the optimal monitoring solution to me.