+++ This bug was initially created as a clone of Bug #1891999 +++ Description of problem: Current storage efficiency card, shows the ratio of `ceph_bluestore_bluestore_compressed_original` and `ceph_bluestore_bluestore_compressed` instead of `ceph_bluestore_bluestore_compressed_original` and `ceph_bluestore_bluestore_compressed_allocated`.`ceph_bluestore_bluestore_compressed` query tells us the new size of the data after compression and not the size required to store the compressed data, as it also depends on min_blob_size of bluestore. Version-Release number of selected component (if applicable): 4.6 How reproducible: 2/2 Steps to Reproduce: 1. Navigate to Storage class creation view 2. Create a compressed pool and a storage class using it. 3. Attach PVC to this storage class and add some data to it. 4. Observe the Storage efficiency card and capacity breakdown card. Actual results: The Storage efficiency card will show high amount of efficiency, but the storage saved will be lesser. Expected results: Storage saved due to compression should match the ratio from storage efficiency card. Additional info:
POOL ID STORED OBJECTS USED %USED MAX AVAIL QUOTA OBJECTS QUOTA BYTES DIRTY USED COMPR UNDER COMPR p-rep3comp 19 10 GiB 2.63k 22 GiB 3.52 199 GiB N/A N/A 2.63k 8.2 GiB 16 GiB The ratio shows which is got as we see 16GiB that where compressed into 8.2GiB But I have a problem with the saving that shows only 5.48 GiB The client sent 10g which with replica 3 should be 30g and now it is 22g so the saving should like 8G. Also we can see that 16g where comporessed to 8.2g so something is wrong with the saving value. How it is calculated?
The query for saving and ratio are same (one gives the ratio and other gives the difference) ratio: sum(ceph_bluestore_bluestore_compressed_original) / clamp_min(sum(ceph_bluestore_bluestore_compressed_allocated),1) saved: (sum(ceph_bluestore_bluestore_compressed_original) - sum(ceph_bluestore_bluestore_compressed_allocated)) I will suggest to open a different bug for saving, as this bug is for compression ratio.
(In reply to Kanika Murarka from comment #8) > The query for saving and ratio are same (one gives the ratio and other gives > the difference) > ratio: sum(ceph_bluestore_bluestore_compressed_original) / > clamp_min(sum(ceph_bluestore_bluestore_compressed_allocated),1) > saved: (sum(ceph_bluestore_bluestore_compressed_original) - > sum(ceph_bluestore_bluestore_compressed_allocated)) > > I will suggest to open a different bug for saving, as this bug is for > compression ratio. Verifying and will open another BZ for the saving issue
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 (OpenShift Container Platform 4.6.6 bug fix update), 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/RHBA-2020:5115