Hi Ernesto,
That makes sense - thanks for explaining why the changes were made.
As we don't use cephadm, being able to customise the data source would
be fantastic.
I've put a tiny pull request to update the docs mentioning "The data
source must be named "Dashboard1"
best regards,
Jake
On 11/26/21 9:32 AM, Ernesto Puerta wrote:
Hi Jake,
AFAIC there were a couple of reasons why we changed that:
* Now, with Cephadm, you can easily deploy a highly-available
monitoring stack (needs a bit of polishing yet though). If, for
example, you extend the Prometheus instance count to 2 or 3, Cephadm
will automatically configure extra data sources accordingly (named
Dashboard2, Dashboard3, etc.).
* The other one is that using "default" basically broke migrations
from pre-Cephadm, and in environments where the "default" datasource
was other than Prometheus. I remember that a Grafana issue prevented
users from renaming the default data source under some circumstances
(I think, when created via the provisioning API).
That said, in the Ceph Dashboard the embedded Grafana URL is hardcoded
to "Dashboard1", so we could add a setting to customize that if required.
Kind Regards,
Ernesto
--
Dr Jake Grimmett
Head Of Scientific Computing
MRC Laboratory of Molecular Biology
Francis Crick Avenue,
Cambridge CB2 0QH, UK.
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
