Orion context broker sometimes not responsive

[Drbvr]

Our implementation of the Orion Context-LD broker does sometimes not process a POST call (create entity) but gives no errors in logging. Reproducing is difficult, since this happens only sometimes. See below for an example of the request of a failed call to the context broker.

We use the Orion context broker with MongoDB installed on Ubuntu 20.04 VM, version 4.4.14.

request details

headers: Link = <https://raw.githubusercontent.com/POORT8/poort8.dvu.datamodels/main/datamodels.context-ngsi.jsonld>; rel="http://www.w3.org/ns/json-ld#context"; type="application/ld+json"

body: {"id":"urn:ngsi-ld:EAN-lookup:*****”,”type":"EAN-lookup","ean”:”*****”,”verblijfObjectIdentificatie”:”*****”,”adresZoekId”:”*****”,”adresseerbaarObjectIdentificatie”:”*****”,”adresWeergaveNaam”:”*****”}

logging Orion context broker

2022-06-21T11:50:17.221742748Z time=Tuesday 21 Jun 11:50:17 2022.222Z | lvl=TMP | corr=N/A | trans=N/A | from=N/A | srv=N/A | subsrv=N/A | comp=Orion | op=orionldMhdConnectionInit.cpp[891]:orionldMhdConnectionInit | msg=------------------------- Servicing NGSI-LD request 005: POST /ngsi-ld/v1/entities --------------------------
2022-06-21T11:50:17.222467143Z time=Tuesday 21 Jun 11:50:17 2022.222Z | lvl=TMP | corr=N/A | trans=N/A | from=N/A | srv=N/A | subsrv=N/A | comp=Orion | op=orionldPostEntities.cpp[63]:orionldPostEntities | msg=********** Running Legacy version of orionldPostEntities

settings on the Orion context broker (all defaults)

orion:
image: fiware/orion-ld
hostname: orion
container_name: fiware-orion
expose:
- "1026"
ports:
- "1026:1026"
command: -dbhost xxx:xxx -logLevel DEBUG
healthcheck:
test: curl --fail -s http://orion:1026/version || exit 1

This issue was created as a follow-up on our earlier helpdesk request at i4trust (https://spaces.fundingbox.com/spaces/i4trust-i4trust-helpdesk/62ab2a46605ce52f65e6d851)

[kzangeli]

We've been having audios about this issue ... Is all good, or do you still have issues?

[Drbvr]

We haven't come around to switching back to this implementation with our customer yet. Will let you know if we run into it again.