Bug #5979: librados: imposes internal tooling expectations on users - Ceph - Ceph

Actions

Copy link

Bug #5979

closed

librados: imposes internal tooling expectations on users

Added by Greg Farnum over 10 years ago. Updated over 10 years ago.

Status:

Resolved

Priority:

Normal

Assignee:

Greg Farnum

Category:

librados

Target version:

% Done:

Source:

Development

Tags:

Backport:

Regression:

Severity:

3 - minor

Reviewed:

Affected Versions:

ceph-qa-suite:

Pull request ID:

Crash signature (v1):

Crash signature (v2):

Description

gregf@kai:~/ceph/examples/librados [(no branch)]$ ./librados_hello_world -c ../../src/ceph.conf
we just set up a rados cluster object
unable to find any monitors in conf. please specify monitors via -m monaddr or -c ceph.conf
couldn't connect to cluster! error -2

This is from the program I'm working on for #5905, and while I wrote "we just set up a rados cluster object" and "couldn't connect to cluster! error -2", the output "unable to find any monitors in conf. please specify monitors via -m monaddr or -c ceph.conf" is generated in the depths of our init code from a simple call to rados.connect(). That's not really appropriate for other applications which might use totally different option syntax.

I'm not sure if this is best handled as a single error string that needs to be dealt with, or if we need to re-examine the librados interfaces systematically for suitability with users outside of the Ceph core projects.