Bug 2318399
| Summary: | [4.16.2] CephFS - Creating large PV from volumesnapshcontent takes a long time | ||
|---|---|---|---|
| Product: | [Red Hat Storage] Red Hat OpenShift Data Foundation | Reporter: | David Vaanunu <dvaanunu> |
| Component: | ceph | Assignee: | Venky Shankar <vshankar> |
| ceph sub component: | CephFS | QA Contact: | Elad <ebenahar> |
| Status: | NEW --- | Docs Contact: | |
| Severity: | high | ||
| Priority: | unspecified | CC: | bniver, muagarwa, sostapov |
| Version: | 4.16 | ||
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | x86_64 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | Type: | Bug | |
| Regression: | --- | Mount Type: | --- |
| Documentation: | --- | CRM: | |
| Verified Versions: | Category: | --- | |
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
| Cloudforms Team: | --- | Target Upstream Version: | |
| Embargoed: | |||
Can see the time gap between the PVC and PV # oc get pv | grep datagen-1pod-3000g-fs pvc-25682db5-1da6-46f9-8e48-13332eaca35e 4000Gi RWO Delete Bound datagen-1pod-3000g-fs/pvc-busy-data-fs-1pod-3000g-1 ocs-storagecluster-cephfs <unset> 29h # oc get pvc -ndatagen-1pod-3000g-fs NAME STATUS VOLUME CAPACITY ACCESS MODES STORAGECLASS VOLUMEATTRIBUTESCLASS AGE pvc-busy-data-fs-1pod-3000g-1 Bound pvc-25682db5-1da6-46f9-8e48-13332eaca35e 4000Gi RWO ocs-storagecluster-cephfs <unset> 45h Can it stay open until the fix? It is impacting my test cases and a relevant bug is needed. Moving the non-blocker bzs out of ODF-4.17.0. If this is a blocker, please feel free to propose it as a blocker for ODF-4.17.0 with justification note. |
Created attachment 2051837 [details] csi-cephfsplugin-provisioner-5677fd74c7-62q2c pod log Description of problem (please be detailed as possible and provide log snippets): Create a single namespace with 1 pod using a large PV size(500g/1T/2T/4T) and usage (100g/500g/1.5T/3T) on CephFs. Using OADP to create CSI snapshot backup for the single namespace. (volumesnapshot & volumesnapshotcontent were created during the backup. volumesnapshot - Delete while backup complete) While restoring the single namespace using the volumesnapshotcontent, the PVC creates fast and the PV creation takes a long time. Version of all relevant components (if applicable): OCP - 4.16.2 ODF - 4.16.2-rhodf Does this issue impact your ability to continue to work with the product (please explain in detail what is the user impact)? Is there any workaround available to the best of your knowledge? Rate from 1 - 5 the complexity of the scenario you performed that caused this bug (1 - very simple, 5 - very complex)? Can this issue reproducible? Yes Can this issue reproduce from the UI? maybe If this is a regression, please provide more details to justify this: Steps to Reproduce: 1. Using OADP operator to backup and restore 2. Create a single namespace with 1 pod 3. Using OADP - Create CSI snapshot backup 4. Using OADP - Restore from CSI snapshot backup Actual results: PV creation (cephFS) takes a long time Expected results: PV creation (cephFS) takes a few seconds (like CephRBD) Additional info: CephFS vs. CephRBD restore duration: CephFS CephRBD PV Size 500G, Usage 100g 0:33:43 0:00:13 PV Size 1000G, Usage 500g 2:44:43 0:00:13 PV Size 2000G, Usage 1500g 8:13:32 0:00:23 PV Size 4000G, Usage 3000g 16:32:42 0:00:23