We need a clear indicator of how full is a storage domain, and alerts when a storage domain becomes too full (e.g. 90% of max lvs). The actual limit on number of logical volumes is currently 1947 volumes. This limit is the size of the leases volume (2GiB), which currently is never extended. If we add support for extending this volume when needed, we can support more logical volumes. We also need a supported way to look up how many LVs are in use, either via a script which is part of engine and maintained by engine developers, or adding this info in the UI. Here is more detail. Every snapshot consumes: - one logical volume for every disk included in the snapshot - one logical volume for memory snapshot if memory snapshot was enabled but the memory volume may be created on another storage domain - logical volume for configuration data (not sure it is still used) The limit of 1000 volumes is just an arbitrary number we invented. We could find the number of LVs in use by cooking up a database query, but accessing the database directly is a bad idea. The query will break when we change the schema, and we must have the flexibility to change the schema whenever we like to.
Created attachment 1447858 [details] Screenshot
Added the number of images for each data domain as seen in the attachment
Verified. Version: Engine - 4.3.0-0.0
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHEA-2019:1085
BZ<2>Jira Resync