Bug 2300310
| Summary: | [cee/sd] mclock_scheduler slow backfill | ||
|---|---|---|---|
| Product: | [Red Hat Storage] Red Hat Ceph Storage | Reporter: | Tomas Petr <tpetr> |
| Component: | RADOS | Assignee: | Sridhar Seshasayee <sseshasa> |
| Status: | CLOSED DUPLICATE | QA Contact: | skanta |
| Severity: | medium | Docs Contact: | |
| Priority: | medium | ||
| Version: | 6.1 | CC: | bhubbard, ceph-eng-bugs, cephqe-warriors, kelwhite, mcaldeir, nojha, pdhiran, rsachere, sseshasa, trchakra, vumrao |
| Target Milestone: | --- | Flags: | sseshasa:
needinfo?
(tpetr) sseshasa: needinfo? sseshasa: needinfo? (tpetr) kelwhite: needinfo? (tpetr) |
| Target Release: | 6.2 | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2025-03-28 07:26:15 UTC | 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: | |||
Please specify the severity of this bug. Severity is defined here: https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity. Putting need info on Tomas for c#9 |
Description of problem: There is an issue with slow backfill when mclock_scheduler is used, the osd_mclock_override_recovery_settings=true + osd_max_backfills=8 does not seem to have any effect on amount of PGs being in backfilling state: --- data: volumes: 1/1 healthy pools: 42 pools, 10488 pgs objects: 710.17M objects, 2.1 PiB usage: 5.4 PiB used, 16 PiB / 21 PiB avail pgs: 21327554/4573948158 objects misplaced (0.466%) 10341 active+clean 68 active+remapped+backfill_wait 61 active+clean+scrubbing+deep 14 active+clean+scrubbing 4 active+remapped+backfilling io: client: 551 MiB/s rd, 23 MiB/s wr, 501 op/s rd, 611 op/s wr recovery: 21 MiB/s, 6 objects/s --- ceph_config_dump:osd advanced osd_max_backfills 8 ceph_config_dump:osd advanced osd_mclock_override_recovery_settings true --- The data moved are cephfs data pool EC profile = 8+3, the reason for data movement is to move data from old HDD OSDs to new HDD_ECC OSDs (both with block.db on SSD) Version-Release number of selected component (if applicable): RHCS 6.1z6 ceph version 17.2.6-216.el9cp How reproducible: Always in this environment Steps to Reproduce: 1. 2. 3. Actual results: Expected results: Additional info: