Support Questions

Find answers, ask questions, and share your expertise
Announcements
Celebrating as our community reaches 100,000 members! Thank you!

NIFI : processor not the most up-to-date

avatar
Rising Star

hi all,

I don't know why there are these logs and the cluster goes down. thanks for help

/var/opt/hosting/nifi_root/logs/nifi-user_2017-03-14.log:2017-03-14 20:21:00,398 INFO [NiFi Web Server-8928] o.a.n.w.a.c.InvalidRevisionExceptionMapper org.apache.nifi.web.InvalidRevisionException: [1, null, ff913814-9d14-1f85-bd62-e150653f6c39] is not the most up-to-date revision. This component appears to have been modified. Returning Bad Request response.
/var/opt/hosting/nifi_root/logs/nifi-user_2017-03-14.log:2017-03-14 20:21:00,457 INFO [NiFi Web Server-6246] o.a.n.w.a.c.IllegalClusterStateExceptionMapper org.apache.nifi.cluster.manager.exception.IllegalClusterStateException: Node nifi001.net:9443 is unable to fulfill this request due to: [1, null, ff913814-9d14-1f85-bd62-e150653f6c39] is not the most up-to-date revision. This component appears to have been modified. Returning Conflict response.

1 REPLY 1

avatar
Super Mentor

@mayki wogno

Are you issuing commands against the rest api or are you trying to make a change within the UI when this occurs?

Sounds like multiple changes being made against the same component at the same time. Each component has a revision number so that two people can't make changes to the exact same component at the same time. So when the second change is applied using the same revision as the first request which was successful, you get these responses.

Two ways this can occur...

1. Two authenticated users are making a change to the configuration of the same processor. User 1 hits apply and that change is applied. User 2 then hits apply and a conflict response occurs from the first node that receives the request.

2. Multiple rest api call are being made against the same component without updating the revision number in the subsequent rest api calls.

As far as node going down.... Do you mean you lose the api and have to refresh the browser? or does the cluster completely go down forcing you to restart nodes to get nodes to rejoin cluster?

Thanks,

Matt