I have been working on some functionality that wraps robis this week. The package seemed to be working fine. However, from this morning, all requests seem to be returning errors, and the error messages seem to vary. Eg
dataset()
robis::dataset(scientificname = "mola mola")
#> Error in robis::dataset(scientificname = "mola mola"): Bad Request (HTTP 400).
But https://api.obis.org/v3/occurrence?taxonid=325567 and https://api.obis.org/v3/occurrence?taxonid=325567&after=-1&size=5000] (the underlying request) returns:
Something broke: [index_not_found_exception] no such index [occurrence], with { resource.type="index_or_alias" & resource.id="occurrence" & index_uuid="_na_" & index="occurrence" }
Not sure what the issue is and it's hard to find information on why the workflow is not working. The CI system in the robis repository doesn't seem to have detected it either (last build of master appears to be 3 months ago). Perhaps setting TRAVIS on a daily CRON job would pick something like this up which would at least give some indication to users that the service is down and through TRAVIS logs, perhaps some insight also.
Hello,
I have been working on some functionality that wraps
robis
this week. The package seemed to be working fine. However, from this morning, all requests seem to be returning errors, and the error messages seem to vary. Egdataset()
Created on 2020-03-25 by the reprex package (v0.3.0)
checklist()
Created on 2020-03-25 by the reprex package (v0.3.0)
occurrence()
Created on 2020-03-25 by the reprex package (v0.3.0)
Digging into the
occurrence
function, here's the request resultCreated on 2020-03-25 by the reprex package (v0.3.0)
A taxon get request to the API using
https://api.obis.org/v3/taxon/325567
returns a successful result:But
https://api.obis.org/v3/occurrence?taxonid=325567
andhttps://api.obis.org/v3/occurrence?taxonid=325567&after=-1&size=5000]
(the underlying request) returns:Not sure what the issue is and it's hard to find information on why the workflow is not working. The CI system in the
robis
repository doesn't seem to have detected it either (last build of master appears to be 3 months ago). Perhaps setting TRAVIS on a daily CRON job would pick something like this up which would at least give some indication to users that the service is down and through TRAVIS logs, perhaps some insight also.