Back to bug 2149676
| Who | When | What | Removed | Added |
|---|---|---|---|---|
| RHEL Program Management | 2022-11-30 15:04:39 UTC | Target Release | ODF 4.12.0 | --- |
| Mudit Agarwal | 2022-11-30 15:04:58 UTC | Depends On | 2149677 | |
| Mudit Agarwal | 2022-11-30 15:05:24 UTC | Status | NEW | POST |
| Nishanth Thomas | 2022-12-01 11:38:25 UTC | Assignee | nthomas | jolmomar |
| Harish NV Rao | 2022-12-13 13:47:53 UTC | QA Contact | hnallurv | vkathole |
| RHEL Program Management | 2022-12-13 13:48:02 UTC | Target Release | --- | ODF 4.11.5 |
| Sunil Kumar Acharya | 2022-12-13 14:45:02 UTC | CC | sheggodu | |
| Flags | needinfo?(jolmomar) | |||
| Juan Miguel Olmo | 2022-12-14 12:44:10 UTC | Link ID | Github red-hat-storage/ceph-csi/pull/152 | |
| Status | POST | ON_QA | ||
| Flags | needinfo?(jolmomar) | |||
| Juan Miguel Olmo | 2022-12-14 12:46:17 UTC | Doc Type | If docs needed, set a value | Bug Fix |
| Doc Text | Cause: When a PVC uses CephFs as storage backend, the inodes metrics (kubelet_volume_stats_inodes_free, kubelet_volume_stats_inodes, kubelet_volume_stats_inodes_used) are not correct because in CephFs inodes get allocated on-demand. Note: (CephFS is a filesystem with dynamic inode allocation, this is by design) Consequence: False alerts around usage in PVs can be raised because the involved metrics to raise this alerts are providing a storage capacity status which can be dynamically changed without any intervention if it is required more storage space. Fix: metrics for kubelet_volume_stats_inodes_free, kubelet_volume_stats_inodes, kubelet_volume_stats_inodes_used are not provided for Cephfs backed PVCs Result: False alarms for storage capacity based in inodes metrics won't be raised. |
|||
| Sunil Kumar Acharya | 2022-12-14 16:20:26 UTC | Status | ON_QA | MODIFIED |
| Red Hat Bugzilla | 2022-12-31 19:07:16 UTC | QA Contact | vkathole | hnallurv |
| Red Hat Bugzilla | 2022-12-31 20:04:36 UTC | QA Contact | hnallurv | |
| Red Hat Bugzilla | 2022-12-31 23:45:11 UTC | CC | jolmomar | |
| Assignee | jolmomar | nthomas | ||
| Red Hat Bugzilla | 2023-01-01 00:45:37 UTC | Assignee | nthomas | nobody |
| CC | nthomas | |||
| Red Hat Bugzilla | 2023-01-01 08:37:56 UTC | CC | ndevos | |
| Alasdair Kergon | 2023-01-04 04:42:08 UTC | Assignee | nobody | jolmomar |
| Alasdair Kergon | 2023-01-04 04:54:40 UTC | QA Contact | vkathole | |
| Alasdair Kergon | 2023-01-04 05:00:20 UTC | CC | jolmomar | |
| Alasdair Kergon | 2023-01-04 05:20:21 UTC | CC | ndevos | |
| Alasdair Kergon | 2023-01-04 05:23:51 UTC | CC | nthomas | |
| Sunil Kumar Acharya | 2023-01-10 05:28:56 UTC | Fixed In Version | odf-4.11.5-8 | |
| Status | MODIFIED | ON_QA | ||
| Raúl Fernández | 2023-01-20 13:42:11 UTC | CC | rauferna | |
| Vishakha Kathole | 2023-01-30 11:02:24 UTC | Status | ON_QA | VERIFIED |
| Harish NV Rao | 2023-02-02 14:52:53 UTC | CC | hnallurv | |
| Kusuma | 2023-02-13 10:27:35 UTC | CC | kbg | |
| Doc Text | Cause: When a PVC uses CephFs as storage backend, the inodes metrics (kubelet_volume_stats_inodes_free, kubelet_volume_stats_inodes, kubelet_volume_stats_inodes_used) are not correct because in CephFs inodes get allocated on-demand. Note: (CephFS is a filesystem with dynamic inode allocation, this is by design) Consequence: False alerts around usage in PVs can be raised because the involved metrics to raise this alerts are providing a storage capacity status which can be dynamically changed without any intervention if it is required more storage space. Fix: metrics for kubelet_volume_stats_inodes_free, kubelet_volume_stats_inodes, kubelet_volume_stats_inodes_used are not provided for Cephfs backed PVCs Result: False alarms for storage capacity based in inodes metrics won't be raised. | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this release, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result false alarms for storage capacity based on inodes metrics are not raised. |
||
| Kusuma | 2023-02-13 10:35:47 UTC | Doc Text | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this release, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result false alarms for storage capacity based on inodes metrics are not raised. | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this release, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result, false alarms for storage capacity based on inodes metrics are not raised. |
| Kusuma | 2023-02-13 10:38:22 UTC | Doc Text | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this release, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result, false alarms for storage capacity based on inodes metrics are not raised. | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this fix, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result, false alarms for storage capacity based on inodes metrics are not raised. |
| Kusuma | 2023-02-13 12:12:16 UTC | Doc Text | Previously, false alarms for storage capacity based on inodes metrics were raised because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand as CephFS is a filesystem with dynamic inode allocations by design. With this fix, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result, false alarms for storage capacity based on inodes metrics are not raised. | Previously, false alarms for storage capacity based on inodes metrics were raised. This was because the metrics to raise the alerts provided a storage capacity status that could be dynamically changed without any intervention when more storage space was required. When a PVC uses CephFs as storage backend, the inodes metrics such as `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, `kubelet_volume_stats_inodes_used` are not correct because in CephFs inodes get allocated on-demand. By design, CephFS is a filesystem with dynamic inode allocations. With this fix, metrics for `kubelet_volume_stats_inodes_free`, `kubelet_volume_stats_inodes`, and `kubelet_volume_stats_inodes_used` are not provided for CephFS backed PVCs. As a result, false alarms for storage capacity based on inodes metrics are not raised. |
| errata-xmlrpc | 2023-02-14 12:34:03 UTC | Status | VERIFIED | RELEASE_PENDING |
| errata-xmlrpc | 2023-02-14 16:58:10 UTC | Resolution | --- | ERRATA |
| Status | RELEASE_PENDING | CLOSED | ||
| Last Closed | 2023-02-14 16:58:10 UTC | |||
| errata-xmlrpc | 2023-02-14 16:58:30 UTC | Link ID | Red Hat Product Errata RHBA-2023:0764 | |
| Sonigra Saurab | 2023-03-08 08:32:27 UTC | Link ID | Red Hat Knowledge Base (Solution) 7001761 | |
| Elad | 2023-08-09 16:37:41 UTC | CC | odf-bz-bot |
Back to bug 2149676