Nuxeo starting very Slow

Hello, We have upgraded nuxeo to version 9.10. And Nuxeo starting is very slow , about 45 minutes to start ! We think that it provokes by the size of the data ( 145G) because we have a other application Nuxeo in version 9.10 with a smaller size of the data (14G) and it isn't the same problems.Thanks for your helping Regards, Eric Galant

0 votes

2 answers



I doubt the size of data (files, DB ?) has any effect on starting Nuxeo.

Do you have Elasticsearch configured and reindex on startup set to true ?

What is written into server.log during startup ?


Yes, we have Elasticsearch configured : elasticsearch.addressList=localhost:9300





The log server.log :

2018-05-04 06:14:32,021 INFO [localhost-startStop-1] [javax.enterprise.resource.webcontainer.jsf.application] jsf.core.injection.provider_generic_web_configured

2018-05-04 06:40:54,209 WARN [http-nio-] [com.sun.jersey.spi.inject.Errors] The following warnings have been detected with resource and/or provider classes: WARNING: A HTTP GET method, public java.lang.String org.nuxeo.elasticsearch.http.readonly.Main.searchWithPayload(java.lang.String,java.lang.String,, throws,org.json.JSONException, should not consume any entity. WARNING: A HTTP GET method, public java.lang.String org.nuxeo.elasticsearch.http.readonly.Main.searchWithPayload(java.lang.String,, throws,org.json.JSONException, should not consume any entity. WARNING: A HTTP GET method, public java.lang.String org.nuxeo.elasticsearch.http.readonly.Main.searchWithPayload(, throws,org.json.JSONException, should not consume any entity.


An embedded ElasticSearch with lots of documents could be a problem

How I could disabled an embedded ElasticSearch ? Thanks.

Either set elasticsearch.enabled to false in nuxeo.conf … or setup a cluster


Please check out all the add-ons installed in nuxeo and if they are compatible with each other. You can check the extensions installed using “nuxeoctl mp-list” on windows. In my case, when I installed nuxeo diff add-on and the server never started because it was not having nuxeo jsf ui compatible with nuxeo diff.

0 votes

You can find below the result at the command nuxeoctl mp-list, I don't think there are a conflict between add-on.

Local packages: studio started ubo-ubo-studio-nuxeo-uni-0 (id: ubo-ubo-studio-nuxeo-uni-0-0.0.0-SNAPSHOT) hotfix started nuxeo-9.10-HF01 (id: nuxeo-9.10-HF01-1.0.1) hotfix started nuxeo-9.10-HF02 (id: nuxeo-9.10-HF02-1.0.0) hotfix started nuxeo-9.10-HF03 (id: nuxeo-9.10-HF03-1.0.1) hotfix started nuxeo-9.10-HF04 (id: nuxeo-9.10-HF04-1.1.0) hotfix started nuxeo-9.10-HF05 (id: nuxeo-9.10-HF05-1.1.0) hotfix started nuxeo-9.10-HF06 (id: nuxeo-9.10-HF06-1.1.0) addon started cas2-authentication (id: cas2-authentication-1.5.3) addon started nuxeo-dam (id: nuxeo-dam-6.3.3) addon started nuxeo-drive (id: nuxeo-drive-1.7.4) addon started nuxeo-jsf-ui (id: nuxeo-jsf-ui-9.10.0-HF01) addon started nuxeo-platform-importer (id: nuxeo-platform-importer-1.8.3) addon started nuxeo-showcase-content (id: nuxeo-showcase-content-1.2.3)


I am not very sure but seems like problem lies in dependency between DAM and JSF UI. My opinion will be:

  1. Uninstall all jsf add-ons.
  2. Then install "nuxeo-jsf-ui (id: nuxeo-jsf-ui-9.3.0)"
  3. Start the server.

See if it works


Is it getting timed out while start?

0 votes

No, because we set this value : launcher.start.max.wait=3600 launcher.stop.max.wait=3600

I moved Elasticsearch in the other machine (cluster) but the problem is the same. The startup of Nuxeo is very slow. We use nuxeo version 9.10 and the size of data is 140G.