Releases: ceems-dev/ceems
Releases · ceems-dev/ceems
0.12.0 / 2025-11-08
- [FEAT] System level logging for
cacct#436 (@mahendrapaipuri) - [BUGFIX] bug(fix): Use configured time zone when SLURM does not include time offsets #433 (@mahendrapaipuri)
- [MAINT] Make multiple update calls to eBPF coll in unit tests #424 (@mahendrapaipuri)
- [MAINT] chore: Update go to 1.25.x #420 (@mahendrapaipuri)
- [MAINT] Always return error responses in JSON for CEEMS LB #414 (@mahendrapaipuri)
- [MAINT] Bump dependencies #412, #415, #416, #417, #419, #421, #422, #423, #425, #426, #427, #428, #429, #430, #434 (@dependabot)
0.11.2 / 2025-09-12
- [MAINT] Update static emission factors for OWID and world average #411 (@mahendrapaipuri)
- [MAINT] Bump dependencies #410 (@dependabot)
0.11.1 / 2025-09-06
- [BUGFIX] Allow GPU fetching to fail for libvirt collector #409 (@mahendrapaipuri)
0.11.0 / 2025-09-02
Breaking Changes
CEEMS Exporter
- Collector
raplis disabled by default now and to enable it add--collector.raplto CLI arguments. - Collector
ipmi_dcmihas been renamed toipmias more functionality beyond DCMI has been added to the collector. - Following metric labels have been renamed to be more consistent with Prometheus naming convention:
ceems_ipmi_dcmi_current_watts->ceems_ipmi_dcmi_power_current_wattsceems_ipmi_dcmi_min_watts->ceems_ipmi_dcmi_power_min_wattsceems_ipmi_dcmi_max_watts->ceems_ipmi_dcmi_power_max_wattsceems_ipmi_dcmi_avg_watts->ceems_ipmi_dcmi_power_avg_wattsceems_redfish_current_watts->ceems_redfish_power_current_wattsceems_redfish_min_watts->ceems_redfish_power_min_wattsceems_redfish_max_watts->ceems_redfish_power_max_wattsceems_redfish_avg_watts->ceems_redfish_power_avg_watts
CEEMS tool
- The relabel configs generated by subcommand
create-relabel-configsare obsolete as the relabelling of metrics directly handled inside the recording rules. Please
regenerate recording rules with new version and remove existing relabel configs on Prometheus server. - Several minor bugs in recording rules have been fixed. Please regenerate the recording rules with new version of
ceems_tool. - GPU profiling metrics have been renamed to have
profin the metric label. For instance,uuid:ceems_gpu_sm_active:ratiobecame
uuid:ceems_gpu_prof_sm_active:ratio. - NVIDIA profiling metrics suffix has been corrected to use
suminstead ofratiofor NVLink, PCIe traffic metrics. Thus, metrics
have been renamed as follows:uuid:ceems_gpu_pcie_tx_bytes:ratio->uuid:ceems_gpu_prof_pcie_tx_bytes:sumuuid:ceems_gpu_pcie_rx_bytes:ratio->uuid:ceems_gpu_prof_pcie_rx_bytes:sumuuid:ceems_gpu_nvlink_tx_bytes:ratio->uuid:ceems_gpu_prof_nvlink_tx_bytes:sumuuid:ceems_gpu_nvlink_rx_bytes:ratio->uuid:ceems_gpu_prof_nvlink_rx_bytes:sum
List of PRs
- [FEAT] Add rules for IO and network metrics #406 (@mahendrapaipuri)
- [FEAT] Support runtime XML directory for libvirt collector #404 (@mahendrapaipuri)
- [MAINT] Bump golanglint-ci to 2.4 #399 (@mahendrapaipuri)
- [BREAKING] Updates and fixes to recording rules subcommand of
ceems_tool#397 (@mahendrapaipuri) - [BREAKING] Support exporting metrics of IPMI sensors #395 (@mahendrapaipuri)
- [MAINT] Bump dependencies #394, #398, #400, #405, #407, #408 (@dependabot)
0.10.2 / 2025-08-07
- [BUGFIX] Fix bpf code to work with LLVM 20 #393 (@mahendrapaipuri)
- [BUGFIX] Fix k8s resource manager #392 (@mahendrapaipuri)
- [MAINT] Bump dependencies #389, #390, #387 (@dependabot)
0.10.1 / 2025-07-22
- [BUGFIX] Fix parsing nvidia-smi XML output #388 (@mahendrapaipuri)
- [MAINT] Bump dependencies #387 (@dependabot)
0.10.0 / 2025-07-20
- [CI] Free up disk space for crossbuild jobs #386 (@mahendrapaipuri)
- [DOCS] Add CONTRIBUTING.md file #385 (@mahendrapaipuri)
- [FEAT] Migrate repo to ceems-dev org #384 (@mahendrapaipuri)
- [FEAT] Filter SLURM cgroups to remove stale ones #382 (@mahendrapaipuri)
- [FEAT] K8s support for CEEMS API server #381 (@mahendrapaipuri)
- [FEAT] Add systemd-less mode for Libvirt collector #377 (@wtripp180901)
- [MAINT] Bump dependencies #375, #376, #378, #383 (@dependabot)
0.9.1 / 2025-07-02
- [FEAT] Support gzip compression #374 (@mahendrapaipuri)
- [MAINT] Bump dependencies #372, #373 (@dependabot)
0.9.0 / 2025-06-27
Breaking Changes
CEEMS LB
- Undocumented Resource-based LB strategy has been removed. Deployments using this strategy must use Prometheus' remote read feature to achieve the same functionality.
CEEMS Exporter
- The configuration of Redfish collector must be under the section
redfish_collectorinstead ofredfish_web. More details in docs. - CLI flag
--collector.redfish.web-confighas been deprecated in the favour of--collector.redfish.config.file. - CLI flag
--collector.k8s.kube-config-filehas been deprecated in the favour of--collector.k8s.kubeconfig.file. - CLI flag
--collector.k8s.kubelet-socket-filehas been deprecated in the favour of--collector.k8s.kubelet-podresources-socket.file.
Redfish Proxy
- The configuration of Redfish proxy must be under
redfish_proxyinstead ofredfish_proxy.web. More details in docs.
List of PRs
- [FEAT] Support env vars in config files #369 (@mahendrapaipuri)
- [FEAT] Add k8s admission controller #367 (@mahendrapaipuri)
- [MAINT] refactor: Rename config section names to be consistent across package #364 (@mahendrapaipuri)
- [BREAKING] breaking: Remove resource-based LB strategy #361 (@mahendrapaipuri)
- [FEAT] Native eBPF profiler #360 (@mahendrapaipuri)
- [MAINT] Bump dependencies #359, #362, #365, #366, #368, #371 (@dependabot)
0.8.0 / 2025-05-20
- [FEAT] Harden redfish proxy app #357 (@mahendrapaipuri)
- [MAINT] Several maintenance changes #354 (@mahendrapaipuri)
- [FEAT] Add k8s collector in the exporter #349 (@mahendrapaipuri)
- [MAINT] Bump dependencies #345, #346, #347, #348, #351, #353, #355, #356, #358 (@dependabot)