nodetool netstats fail to reach jmx endpoint

[glommer]

This was seen by a user today:

For some reason we can't get to the proper REST endpoint to serve a JMX request, and as a result we print the following message instead:

Read Repair Statistics:
Attempted: 220
Mismatch (Blocking): 0
Mismatch (Background): 0
Pool Name                    Active   Pending      Completed   Dropped
nodetool: Scylla API server HTTP GET to URL '/messaging_service/messages/pending' failed: Not found
See 'nodetool help' or 'nodetool help <command>'.

@amnonh please advise ASAP.

[tzach]

What is the Scylla version? Could it be a mismatch between nodetool Scylla?

[glommer]

For this case, enterprise-2018.1.4

[amnonh]

I'm on vacation with limited connectivity, is that the only api the nodetool command uses that doesn't work?

It's very rare that only a specific api call would fail, it can happen at startup if scylla didn't boot completely, I don't think this is the case

On Tuesday, September 11, 2018, Glauber Costa [email protected] wrote:

For this case, enterprise-2018.1.4

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/scylladb/scylla-jmx/issues/56#issuecomment-420360956, or mute the thread https://github.com/notifications/unsubscribe-auth/ACBRv_xqcY5uVUiUlgCjvqanPYcOUzitks5uZ_dtgaJpZM4WjaAU .