docker-compose up server failed to execute workflow because of "XContentType" class is not found

[dingliu1]

Describe the bug
docker-compose up server failed to execute workflow because of "XContentType" class is not found.

Details
Conductor version: v3.15.0 or the latest
Persistence implementation: Cassandra, Postgres, MySQL, Dynomite etc: Postgres
Queue implementation: Postgres, MySQL, Dynoqueues etc
Lock: Redis or Zookeeper?
Workflow definition:
Task definition:
Event handler definition:

To Reproduce
Steps to reproduce the behavior:

1. Go to conductor-3.15.0\docker or the latest version\docker
2. Execute below command
   docker-compose -f docker-compose-postgres.yaml build
   docker-compose -f docker-compose-postgres.yaml up
3. Create some sample workflow definition and execute
   "Definition" -> "New Workflow Definition" -> set sample "uri" e.g.
   uri": "https://datausa.io/api/data?drilldowns=Nation&measures=Population",drilldowns=Nation&measures=Population",
   Save the definition -> Go to "Workbench" -> choose just created workflow name -> click execute
   Now the conductor server logs report:
   "org.springframework.web.util.NestedServletException: Handler dispatch failed; nested exception is java.lang.NoClassDefFoundError: org/elasticsearch/common/xcontent/XContentType"
4. The root cause seems that:
   The Dockerfile packaged "conductor-server.jar\BOOT-INF\lib" I can see elasticsearch-x-content-7.17.13.jar, the "XContentType" class is re-packaged.
   https://stackoverflow.com/questions/72428790/xcontenttype-is-removed-from-elasticsearch-ver-7-17-3

Expected behavior
docker-compose inside docker/server folder should work at any time

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.

[Nostras]

Can confirm we're facing the exact same issue.

[gforce81]

Same issue

[kyandaks]

Strangely the same issue has surfaced on our side too

[tianxie]

Me too. Is there any workaround?

[solomonmetta]

Me too. Is there any workaround?

Seems to be due to some changes in the conductor-community repo. A temporary work-around could be to use an older commit of the conductor community repo in the docker/server/Dockerfile.

[nadaHilal1]

same here !

[gajendrangnanasekaran]

any update regarding this issue, is this fixed ?please reply

[gajendrangnanasekaran]

we also facing the same issue exactly

[manan164]

Hi @gajendrangnanasekaran , The issue is fixed in PR Netflix/conductor-community#303. It will be released in the next release.

[Nostras]

Specifically, add
RUN git reset --hard 7e1b9122477ae0647b61d1206e49be87f0ae65d1
to line 39 in the Dockerfile https://github.com/Netflix/conductor/blob/main/docker/server/Dockerfile until v1r3n has made a new community release.

[gajendrangnanasekaran]

Thanks lot when it will get released?

[Nostras]

Thanks lot when it will get released?

Just apply my fix for now, the community version can then release when they're ready.

[gajendrangnanasekaran]

Now its working fine as per @Nostras inputs, thanks lot @Nostras and @manan164