The point of this project is prevent non-curation traffic from interfering with normal operations. As a demonstration of the issue, we have previously dealt with outages caused by spiders crawling third-party sites (alliance) triggering API calls into our curation system at such a high volume that it was brought down. While this has been temporarily mitigated, there are other ways into the system. For example:
Currently, GO-CAM URIs resolve to the graph editor; we know of groups that use these as links for their work
Links from widgets that we produce lead back to a view that is drive by the curation system
The most robust solution would be to public facing replacement for these views that does not depend on the curation system (e.g. the /go-cam site or a new TBD site); all links and URIs should resolve to there.
kltm
commented
1 year ago
Project link
https://github.com/orgs/geneontology/projects/TBD
Project description
The point of this project is prevent non-curation traffic from interfering with normal operations. As a demonstration of the issue, we have previously dealt with outages caused by spiders crawling third-party sites (alliance) triggering API calls into our curation system at such a high volume that it was brought down. While this has been temporarily mitigated, there are other ways into the system. For example:
The most robust solution would be to public facing replacement for these views that does not depend on the curation system (e.g. the /go-cam site or a new TBD site); all links and URIs should resolve to there.
PI
TBD
Project owner (PO)
TBD
Technical lead (TL)
TBD
Other personnel (OP)
TBD
Technical specs
TBD
Other comments
At least part of this would involve https://github.com/geneontology/api-gorest-2021/issues/6