From c99e0052d137936f9406ea0db83ad762d3e3a4b5 Mon Sep 17 00:00:00 2001 From: Nicolas Perrin <nicolas.perrin@inria.fr> Date: Thu, 23 Jan 2025 09:34:17 +0100 Subject: [PATCH 1/7] [sophia][esterel1] add basic configuration generated by rake mass:create --- .../sophia/clusters/esterel1/esterel1.json | 20 +++ .../clusters/esterel1/nodes/esterel1-1.json | 128 ++++++++++++++++++ .../sophia/clusters/esterel1/esterel1.yaml | 41 ++++++ .../sophia/clusters/esterel1/nodes.yaml.erb | 80 +++++++++++ 4 files changed, 269 insertions(+) create mode 100644 data/grid5000/sites/sophia/clusters/esterel1/esterel1.json create mode 100644 data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json create mode 100644 input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml create mode 100644 input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json new file mode 100644 index 00000000000..d66d9081b69 --- /dev/null +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -0,0 +1,20 @@ +{ + "boot_type": "uefi", + "created_at": "Thu, 01 Jan 1970 00:00:00 GMT", + "exotic": false, + "kavlan": false, + "manufactured_at": "1970-01-01", + "metrics": [ + + ], + "model": "Cluster Model", + "priority": 197001, + "queues": [ + "admin", + "testing" + ], + "redfish": true, + "type": "cluster", + "uid": "esterel1", + "warranty_end": "1970-01-01" +} \ No newline at end of file diff --git a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json new file mode 100644 index 00000000000..08284bd9150 --- /dev/null +++ b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json @@ -0,0 +1,128 @@ +{ + "architecture": { + "cpu_core_numbering": "contiguous", + "nb_cores": 72, + "nb_procs": 1, + "nb_threads": 72, + "platform_type": "x86_64" + }, + "bios": { + "release_date": "01/01/2000", + "vendor": "Unknown", + "version": 1 + }, + "bmc_version": "v1", + "chassis": { + "manufactured_at": "1970-01-01", + "manufacturer": "Unknown", + "name": "Unknown", + "warranty_end": "1970-01-01" + }, + "exotic": false, + "main_memory": { + "ram_size": 8 + }, + "management_tools": { + "bmc_vendor_tool": "ipmitool", + "ipmitool": { + "retries": 5 + } + }, + "memory_devices": [ + { + "device": "dimm_proc 1 dimm 1", + "size": 8, + "technology": "dram" + } + ], + "network_adapters": [ + { + "device": "eth0", + "driver": "mlx_core", + "enabled": true, + "interface": "Ethernet", + "ip": "172.16.131.1", + "ip6": "2001:660:4406:800:4::1", + "kavlan": false, + "mac": "18:66:da:ed:2c:44", + "management": false, + "mountable": true, + "mounted": true, + "name": "enp1s0f0np0", + "network_address": "esterel1-1.sophia.grid5000.fr", + "rate": 10000000000, + "switch": null, + "switch_port": null + }, + { + "device": "bmc", + "enabled": true, + "interface": "Ethernet", + "ip": "172.17.131.1", + "kavlan": false, + "mac": "18:fb:7b:a8:b6:a1", + "management": true, + "mountable": false, + "mounted": false, + "network_address": "esterel1-1-bmc.sophia.grid5000.fr" + } + ], + "nodeset": "esterel1", + "operating_system": { + "cstate_driver": "unknown", + "cstate_governor": "unknown", + "ht_enabled": true, + "pstate_driver": "unknwon", + "pstate_governor": "unknown", + "turboboost_enabled": true + }, + "performance": { + "core_flops": 128, + "node_flops": 9216 + }, + "processor": { + "cache_l1": null, + "cache_l1d": 8, + "cache_l1i": 8, + "cache_l2": 8, + "cache_l3": 8, + "clock_speed": 8, + "ht_capable": true, + "instruction_set": "x86-64", + "microarchitecture": "Haswell", + "microcode": "0xd000001", + "model": "Unknown", + "other_description": "description", + "vendor": "vendor", + "version": "vendor" + }, + "redfish": true, + "software": { + "forced-deployment-timestamp": 202007300948, + "postinstall-version": "1.2025011610", + "standard-environment": "debian11-x64-std" + }, + "storage_devices": [ + { + "by_id": "", + "by_path": "/dev/disk/by-path/dummy", + "id": "disk0", + "interface": "SAS", + "model": "unknown", + "size": 8, + "storage": "SSD", + "vendor": "Unknown" + } + ], + "supported_job_types": { + "besteffort": true, + "deploy": true, + "max_walltime": 0, + "queues": [ + "admin", + "testing" + ] + }, + "type": "node", + "uid": "esterel1-1" +} \ No newline at end of file diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml new file mode 100644 index 00000000000..9469497d91f --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml @@ -0,0 +1,41 @@ +--- +model: Cluster Model # TODO: change this value. +created_at: 1970-01-01 # TODO: change this value +kavlan: false +boot_type: uefi # TODO: specify if 'uefi' (ideally) or 'bios' (legacy, if no other choice) +exotic: false # TODO: specify if 'true' or 'false' +queues: + - admin + - testing +nodes: + esterel1-1: + chassis: + manufactured_at: 1970-01-01 # TODO: Put date. + warranty_end: 1970-01-01 # TODO: Put date. + supported_job_types: + deploy: true + besteffort: true + max_walltime: 0 + processor: + microarchitecture: Haswell # TODO: replace with microarch name. + clock_speed: 8 # TODO: Replace with clock speed. + network_adapters: + bmc: + interface: Ethernet + enabled: true + mountable: false + mounted: false + eth0: + enabled: true + mountable: true + mounted: true + storage_devices: + disk0: # This field will have to be renamed later. + id: disk0 + interface: SAS + by_path: "/dev/disk/by-path/dummy" # this path will have to change later. + software: + standard-environment: debian11-x64-std # TODO: check that architecture is OK + management_tools: + bmc_vendor_tool: ipmitool # TODO: replace with bmc_vendor_tool (ipmitool, racadm) + nodeset: esterel1 diff --git a/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb b/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb new file mode 100644 index 00000000000..c4cf8cc0922 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb @@ -0,0 +1,80 @@ +<% + # File generated by 'rake mass:create SRC=doc/sophia-clusters-mass-create-tmp.csv + # If changes are needed, it might be better to edit the source data and regenerate using 'rake mass:create' + + cluster_name = "esterel1" + nodes_number = 1 # Size of the cluster (number of nodes) + # MAC addresses declaration + mac_eth0_list = %w( +18:66:da:ed:2c:44 + ) + mac_bmc_list = %w( +18:fb:7b:a8:b6:a1 + ) +%> +--- +nodes: +<% (1..nodes_number).each { |i| %> + <%= cluster_name %>-<%= i %>: + architecture: + nb_procs: 1 # Fake data, will be replaced by g5k-checks + nb_cores: 72 # Fake data, will be replaced by g5k-checks + nb_threads: 72 # Fake data, will be replaced by g5k-checks + platform_type: x86_64 # Fake data, will be replaced by g5k-checks + cpu_core_numbering: contiguous # Fake data, will be replaced by g5k-checks + bios: + release_date: 01/01/2000 # Fake date, will be replaced by g5k-checks + vendor: Unknown # Fake vendor, will be replaced by g5k-checks + version: 1 # Fake version, will be replaced by g5k-checks + bmc_version: v1 # Fake version, will be replaced by g5k-checks + chassis: + manufacturer: Unknown # Fake manufacturer, will be replaced by g5k-checks + name: Unknown # Fake name, will be replaced by g5k-checks + main_memory: + ram_size: 8 # Fake size, will be replaced by g5k-checks + memory_devices: + dimm: + size: 8 # Fake size, will be replaced by g5k-checks + technology: dram # Common memory technology, will be replaced by g5k-checks + processor: + model: Unknown # Fake model name, will be replaced by g5k-checks + other_description: description # Fake description, will be replaced by g5k-checks + vendor: vendor # Fake vendor, will be replaced by g5k-checks + version: vendor # Fake version, will be replaced by g5k-checks + cache_l1d: 8 # Fake cache, will be replaced by g5k-checks + cache_l1i: 8 # Fake cache, will be replaced by g5k-checks + cache_l2: 8 # Fake cache, will be replaced by g5k-checks + cache_l3: 8 # Fake cache, will be replaced by g5k-checks + instruction_set: x86-64 # Common instruction set, will be replaced by g5k-checks + microcode: "0xd000001" # Fake microcode, will be replaced by g5k-checks + ht_capable: true # Default ht capable value, will be replaced by g5k-checks + main_memory: + ram_size: 8 # Fake ram size, will be replaced by g5k-checks + memory_devices: + dimm_proc 1 dimm 1: # Fake dimm name, will be replaced by g5k-checks + size: 8 # Fake dimm size, will be replaced by g5k-checks + technology: dram # Default dimm technology, will be replaced by g5k-checks + operating_system: + cstate_driver: unknown # Fake driver, will be replaced by g5k-checks + cstate_governor: unknown # Fake governor, will be replaced by g5k-checks + ht_enabled: true # common value for hyper threading, will be replaced by g5k-checks + pstate_driver: unknwon # Fake driver, will be replaced by g5k-checks + pstate_governor: unknown # Fake driver, will be replaced by g5k-checks + turboboost_enabled: true # Default value for turboboost, will be replaced by g5k-checks + network_adapters: + bmc: + management: true + mac: <%= mac_bmc_list[i - 1] %> + eth0: + interface: Ethernet + management: false + driver: mlx_core # Fake data, will be replaced by g5k-checks + name: enp1s0f0np0 # Fake data, will be replaced by g5k-checks + rate: 10000000000 # Fake data, will be replaced by g5k-checks + mac: <%= mac_eth0_list[i - 1] %> + storage_devices: + disk0: + storage: SSD # Fake data, will be replaced by g5k-checks + model: unknown # Fake data, will be replaced by g5k-checks + size: 8 # Fake data, will be replaced by g5k-checks +<% } %> -- GitLab From 6fa44205b69d659dad0848a22d4ee1bdbb06ce26 Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Wed, 12 Mar 2025 15:20:32 +0100 Subject: [PATCH 2/7] [sophia][esterel1] Add disk by-path --- .../sites/sophia/clusters/esterel1/esterel1.json | 2 ++ .../sophia/clusters/esterel1/nodes/esterel1-1.json | 12 +++++++++++- .../sites/sophia/clusters/esterel1/esterel1.yaml | 6 +++++- .../sites/sophia/clusters/esterel1/nodes.yaml.erb | 4 ++++ 4 files changed, 22 insertions(+), 2 deletions(-) diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json index d66d9081b69..2fc681f9576 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -8,6 +8,8 @@ ], "model": "Cluster Model", + "nodes_count": 1, + "nodes_description": "1 CPU Unknown vendor, 72 cores/CPU, 0GB RAM, 2x0GB SSD, 1 x 10Gb Ethernet", "priority": 197001, "queues": [ "admin", diff --git a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json index 08284bd9150..ce5bd2d7cfb 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json @@ -105,13 +105,23 @@ "storage_devices": [ { "by_id": "", - "by_path": "/dev/disk/by-path/dummy", + "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0", "id": "disk0", "interface": "SAS", "model": "unknown", "size": 8, "storage": "SSD", "vendor": "Unknown" + }, + { + "by_id": "", + "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0", + "id": "disk1", + "interface": "SAS", + "model": "unknown", + "size": 8, + "storage": "SSD", + "vendor": "Unknown" } ], "supported_job_types": { diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml index 9469497d91f..a907cb04de3 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml @@ -33,7 +33,11 @@ nodes: disk0: # This field will have to be renamed later. id: disk0 interface: SAS - by_path: "/dev/disk/by-path/dummy" # this path will have to change later. + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" # this path will have to change later. + disk1: + id: disk1 + interface: SAS + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" software: standard-environment: debian11-x64-std # TODO: check that architecture is OK management_tools: diff --git a/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb b/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb index c4cf8cc0922..6048d7e052a 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb +++ b/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb @@ -77,4 +77,8 @@ nodes: storage: SSD # Fake data, will be replaced by g5k-checks model: unknown # Fake data, will be replaced by g5k-checks size: 8 # Fake data, will be replaced by g5k-checks + disk1: + storage: SSD # Fake data, will be replaced by g5k-checks + model: unknown # Fake data, will be replaced by g5k-checks + size: 8 # Fake data, will be replaced by g5k-checks <% } %> -- GitLab From 3c1fbf165eae48831abeaab087dfaa1ef8b35f40 Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Wed, 12 Mar 2025 15:43:28 +0100 Subject: [PATCH 3/7] [sophia][esterel1] net name mapping --- lib/refrepo/net_names_mapping.yaml | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/lib/refrepo/net_names_mapping.yaml b/lib/refrepo/net_names_mapping.yaml index e36217d4b4f..c5a2a9d743c 100644 --- a/lib/refrepo/net_names_mapping.yaml +++ b/lib/refrepo/net_names_mapping.yaml @@ -378,6 +378,12 @@ vianden: eno8303: eth4 eno8403: eth5 enp158s0np0: eth6 +esterel1: + enp1s0f0np0: eth0 + eno2: eth1 + eno3: eth2 + eno4: eth3 + ibp131s0: ib0 esterel5: eno1: eth0 eno2: eth1 -- GitLab From f40f47b89b31b9b05ec31d3fd861beeca6c8a671 Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Wed, 12 Mar 2025 16:16:46 +0100 Subject: [PATCH 4/7] [sophia][esterel1] g5k checks import and IB interface --- .../sophia/clusters/esterel1/esterel1.json | 2 +- .../clusters/esterel1/nodes/esterel1-1.json | 172 +++++++++++++----- input/grid5000/ipv4.yaml | 1 + .../sophia/clusters/esterel1/esterel1.yaml | 4 +- .../sophia/clusters/esterel1/esterel1_ib.yaml | 10 + .../sophia/clusters/esterel1/nodes.yaml.erb | 84 --------- .../clusters/esterel1/nodes/esterel1-1.yaml | 132 ++++++++++++++ 7 files changed, 275 insertions(+), 130 deletions(-) create mode 100644 input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml delete mode 100644 input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb create mode 100644 input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json index 2fc681f9576..b9b8568cd59 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -9,7 +9,7 @@ ], "model": "Cluster Model", "nodes_count": 1, - "nodes_description": "1 CPU Unknown vendor, 72 cores/CPU, 0GB RAM, 2x0GB SSD, 1 x 10Gb Ethernet", + "nodes_description": "2 CPUs Intel Xeon E5-2623 v4, 4 cores/CPU, 32GB RAM, 2x372GB HDD, 1 x 1Gb Ethernet, 1 x 40Gb InfiniBand", "priority": 197001, "queues": [ "admin", diff --git a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json index ce5bd2d7cfb..4a69e5440b9 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json @@ -1,26 +1,27 @@ { "architecture": { - "cpu_core_numbering": "contiguous", - "nb_cores": 72, - "nb_procs": 1, - "nb_threads": 72, + "cpu_core_numbering": "round-robin", + "nb_cores": 8, + "nb_procs": 2, + "nb_threads": 8, "platform_type": "x86_64" }, "bios": { - "release_date": "01/01/2000", - "vendor": "Unknown", - "version": 1 + "release_date": "12/12/2023", + "vendor": "Dell Inc.", + "version": "2.19.0" }, - "bmc_version": "v1", + "bmc_version": "2.86", "chassis": { "manufactured_at": "1970-01-01", - "manufacturer": "Unknown", - "name": "Unknown", + "manufacturer": "Dell Inc.", + "name": "PowerEdge R730", + "serial": "88Y9RG2", "warranty_end": "1970-01-01" }, "exotic": false, "main_memory": { - "ram_size": 8 + "ram_size": 34359738368 }, "management_tools": { "bmc_vendor_tool": "ipmitool", @@ -30,29 +31,111 @@ }, "memory_devices": [ { - "device": "dimm_proc 1 dimm 1", - "size": 8, + "device": "dimm_a1", + "size": 8589934592, + "technology": "dram" + }, + { + "device": "dimm_a2", + "size": 8589934592, + "technology": "dram" + }, + { + "device": "dimm_b1", + "size": 8589934592, + "technology": "dram" + }, + { + "device": "dimm_b2", + "size": 8589934592, "technology": "dram" } ], "network_adapters": [ { "device": "eth0", - "driver": "mlx_core", + "driver": "tg3", "enabled": true, + "firmware_version": "FFV22.61.8 bc 5720-v1.39", "interface": "Ethernet", "ip": "172.16.131.1", "ip6": "2001:660:4406:800:4::1", "kavlan": false, "mac": "18:66:da:ed:2c:44", "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", "mountable": true, "mounted": true, "name": "enp1s0f0np0", "network_address": "esterel1-1.sophia.grid5000.fr", - "rate": 10000000000, - "switch": null, - "switch_port": null + "rate": 1000000000, + "sriov": false, + "sriov_totalvfs": 0, + "switch": "sw-6", + "switch_port": "1/1/13", + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "eth1", + "driver": "tg3", + "firmware_version": "FFV22.61.8 bc 5720-v1.39", + "interface": "Ethernet", + "kavlan": false, + "mac": "18:66:da:ed:2c:45", + "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "name": "eno2", + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "eth2", + "driver": "tg3", + "firmware_version": "FFV22.61.8 bc 5720-v1.39", + "interface": "Ethernet", + "kavlan": false, + "mac": "18:66:da:ed:2c:46", + "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "name": "eno3", + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "eth3", + "driver": "tg3", + "firmware_version": "FFV22.61.8 bc 5720-v1.39", + "interface": "Ethernet", + "kavlan": false, + "mac": "18:66:da:ed:2c:47", + "management": false, + "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "name": "eno4", + "sriov": false, + "sriov_totalvfs": 0, + "vendor": "Broadcom Inc. and subsidiaries" + }, + { + "device": "ib0", + "driver": "mlx4_core", + "enabled": true, + "firmware_version": "2.36.5000", + "guid": "0x7cfe900300bf0d41", + "interface": "InfiniBand", + "ip": "172.18.131.1", + "kavlan": false, + "mac": "7c:fe:90:03:00:bf:0d:41", + "management": false, + "model": "MT27500 Family [ConnectX-3]", + "mountable": true, + "mounted": true, + "name": "ib0", + "netmask": "255.255.240.0", + "network_address": "esterel1-1-ib0.sophia.grid5000.fr", + "rate": 40000000000, + "vendor": "Mellanox Technologies" }, { "device": "bmc", @@ -69,32 +152,32 @@ ], "nodeset": "esterel1", "operating_system": { - "cstate_driver": "unknown", - "cstate_governor": "unknown", - "ht_enabled": true, - "pstate_driver": "unknwon", - "pstate_governor": "unknown", + "cstate_driver": "intel_idle", + "cstate_governor": "menu", + "ht_enabled": false, + "pstate_driver": "intel_cpufreq", + "pstate_governor": "performance", "turboboost_enabled": true }, "performance": { "core_flops": 128, - "node_flops": 9216 + "node_flops": 1024 }, "processor": { "cache_l1": null, - "cache_l1d": 8, - "cache_l1i": 8, - "cache_l2": 8, - "cache_l3": 8, + "cache_l1d": 32768, + "cache_l1i": 32768, + "cache_l2": 262144, + "cache_l3": 10485760, "clock_speed": 8, "ht_capable": true, "instruction_set": "x86-64", "microarchitecture": "Haswell", - "microcode": "0xd000001", - "model": "Unknown", - "other_description": "description", - "vendor": "vendor", - "version": "vendor" + "microcode": "0xb000040", + "model": "Intel Xeon", + "other_description": "Intel(R) Xeon(R) CPU E5-2623 v4 @ 2.60GHz", + "vendor": "Intel", + "version": "E5-2623 v4" }, "redfish": true, "software": { @@ -104,24 +187,26 @@ }, "storage_devices": [ { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x61866da07c7090001fcad360885bea08", "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0", + "firmware_version": 4.3, "id": "disk0", "interface": "SAS", - "model": "unknown", - "size": 8, - "storage": "SSD", - "vendor": "Unknown" + "model": "PERC H730 Mini", + "size": 399431958528, + "storage": "HDD", + "vendor": "Dell" }, { - "by_id": "", + "by_id": "/dev/disk/by-id/wwn-0x61866da07c7090001fcad58ba96aaf8a", "by_path": "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0", + "firmware_version": 4.3, "id": "disk1", "interface": "SAS", - "model": "unknown", - "size": 8, - "storage": "SSD", - "vendor": "Unknown" + "model": "PERC H730 Mini", + "size": 399431958528, + "storage": "HDD", + "vendor": "Dell" } ], "supported_job_types": { @@ -131,7 +216,8 @@ "queues": [ "admin", "testing" - ] + ], + "virtual": "ivt" }, "type": "node", "uid": "esterel1-1" diff --git a/input/grid5000/ipv4.yaml b/input/grid5000/ipv4.yaml index 11bac905e7d..ca71c539b03 100644 --- a/input/grid5000/ipv4.yaml +++ b/input/grid5000/ipv4.yaml @@ -126,6 +126,7 @@ ipv4: sophia mercantour6 ib0 0 0 2 45 sophia mercantour7 eth0 0 0 2 46 sophia esterel1 eth0 0 0 3 0 + sophia esterel1 ib0 0 0 3 0 sophia esterel2 eth0 0 0 3 1 sophia esterel3 eth0 0 0 3 2 sophia esterel4 eth0 0 0 3 3 diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml index a907cb04de3..8999f20a787 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml @@ -30,11 +30,11 @@ nodes: mountable: true mounted: true storage_devices: - disk0: # This field will have to be renamed later. + pci-0000:03:00.0-scsi-0:2:0:0: # This field will have to be renamed later. id: disk0 interface: SAS by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" # this path will have to change later. - disk1: + pci-0000:03:00.0-scsi-0:2:1:0: id: disk1 interface: SAS by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml new file mode 100644 index 00000000000..74090dcbcc2 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml @@ -0,0 +1,10 @@ +nodes: + esterel1-1: + network_adapters: + ib0: + enabled: true + mountable: true + mounted: true + management: false + interface: InfiniBand + netmask: 255.255.240.0 diff --git a/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb b/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb deleted file mode 100644 index 6048d7e052a..00000000000 --- a/input/grid5000/sites/sophia/clusters/esterel1/nodes.yaml.erb +++ /dev/null @@ -1,84 +0,0 @@ -<% - # File generated by 'rake mass:create SRC=doc/sophia-clusters-mass-create-tmp.csv - # If changes are needed, it might be better to edit the source data and regenerate using 'rake mass:create' - - cluster_name = "esterel1" - nodes_number = 1 # Size of the cluster (number of nodes) - # MAC addresses declaration - mac_eth0_list = %w( -18:66:da:ed:2c:44 - ) - mac_bmc_list = %w( -18:fb:7b:a8:b6:a1 - ) -%> ---- -nodes: -<% (1..nodes_number).each { |i| %> - <%= cluster_name %>-<%= i %>: - architecture: - nb_procs: 1 # Fake data, will be replaced by g5k-checks - nb_cores: 72 # Fake data, will be replaced by g5k-checks - nb_threads: 72 # Fake data, will be replaced by g5k-checks - platform_type: x86_64 # Fake data, will be replaced by g5k-checks - cpu_core_numbering: contiguous # Fake data, will be replaced by g5k-checks - bios: - release_date: 01/01/2000 # Fake date, will be replaced by g5k-checks - vendor: Unknown # Fake vendor, will be replaced by g5k-checks - version: 1 # Fake version, will be replaced by g5k-checks - bmc_version: v1 # Fake version, will be replaced by g5k-checks - chassis: - manufacturer: Unknown # Fake manufacturer, will be replaced by g5k-checks - name: Unknown # Fake name, will be replaced by g5k-checks - main_memory: - ram_size: 8 # Fake size, will be replaced by g5k-checks - memory_devices: - dimm: - size: 8 # Fake size, will be replaced by g5k-checks - technology: dram # Common memory technology, will be replaced by g5k-checks - processor: - model: Unknown # Fake model name, will be replaced by g5k-checks - other_description: description # Fake description, will be replaced by g5k-checks - vendor: vendor # Fake vendor, will be replaced by g5k-checks - version: vendor # Fake version, will be replaced by g5k-checks - cache_l1d: 8 # Fake cache, will be replaced by g5k-checks - cache_l1i: 8 # Fake cache, will be replaced by g5k-checks - cache_l2: 8 # Fake cache, will be replaced by g5k-checks - cache_l3: 8 # Fake cache, will be replaced by g5k-checks - instruction_set: x86-64 # Common instruction set, will be replaced by g5k-checks - microcode: "0xd000001" # Fake microcode, will be replaced by g5k-checks - ht_capable: true # Default ht capable value, will be replaced by g5k-checks - main_memory: - ram_size: 8 # Fake ram size, will be replaced by g5k-checks - memory_devices: - dimm_proc 1 dimm 1: # Fake dimm name, will be replaced by g5k-checks - size: 8 # Fake dimm size, will be replaced by g5k-checks - technology: dram # Default dimm technology, will be replaced by g5k-checks - operating_system: - cstate_driver: unknown # Fake driver, will be replaced by g5k-checks - cstate_governor: unknown # Fake governor, will be replaced by g5k-checks - ht_enabled: true # common value for hyper threading, will be replaced by g5k-checks - pstate_driver: unknwon # Fake driver, will be replaced by g5k-checks - pstate_governor: unknown # Fake driver, will be replaced by g5k-checks - turboboost_enabled: true # Default value for turboboost, will be replaced by g5k-checks - network_adapters: - bmc: - management: true - mac: <%= mac_bmc_list[i - 1] %> - eth0: - interface: Ethernet - management: false - driver: mlx_core # Fake data, will be replaced by g5k-checks - name: enp1s0f0np0 # Fake data, will be replaced by g5k-checks - rate: 10000000000 # Fake data, will be replaced by g5k-checks - mac: <%= mac_eth0_list[i - 1] %> - storage_devices: - disk0: - storage: SSD # Fake data, will be replaced by g5k-checks - model: unknown # Fake data, will be replaced by g5k-checks - size: 8 # Fake data, will be replaced by g5k-checks - disk1: - storage: SSD # Fake data, will be replaced by g5k-checks - model: unknown # Fake data, will be replaced by g5k-checks - size: 8 # Fake data, will be replaced by g5k-checks -<% } %> diff --git a/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml new file mode 100644 index 00000000000..e1eb832b0d1 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml @@ -0,0 +1,132 @@ +# Generated by g5k-checks (g5k-checks -m api) +--- +esterel1-1: + architecture: + cpu_core_numbering: round-robin + nb_cores: 8 + nb_procs: 2 + nb_threads: 8 + platform_type: x86_64 + bios: + release_date: 12/12/2023 + vendor: Dell Inc. + version: 2.19.0 + bmc_version: '2.86' + chassis: + manufacturer: Dell Inc. + name: PowerEdge R730 + serial: 88Y9RG2 + main_memory: + ram_size: 34359738368 + memory_devices: + dimm_a1: + size: 8589934592 + technology: dram + dimm_a2: + size: 8589934592 + technology: dram + dimm_b1: + size: 8589934592 + technology: dram + dimm_b2: + size: 8589934592 + technology: dram + network_adapters: + bmc: + ip: 172.17.131.1 + mac: 18:fb:7b:a8:b6:a1 + management: true + eth0: + driver: tg3 + firmware_version: FFV22.61.8 bc 5720-v1.39 + interface: Ethernet + ip: 172.16.131.1 + mac: 18:66:da:ed:2c:44 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: enp1s0f0np0 + rate: 1000000000 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + eth1: + driver: tg3 + firmware_version: FFV22.61.8 bc 5720-v1.39 + interface: Ethernet + mac: 18:66:da:ed:2c:45 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: eno2 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + eth2: + driver: tg3 + firmware_version: FFV22.61.8 bc 5720-v1.39 + interface: Ethernet + mac: 18:66:da:ed:2c:46 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: eno3 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + eth3: + driver: tg3 + firmware_version: FFV22.61.8 bc 5720-v1.39 + interface: Ethernet + mac: 18:66:da:ed:2c:47 + management: false + model: NetXtreme BCM5720 2-port Gigabit Ethernet PCIe + name: eno4 + sriov: false + sriov_totalvfs: 0 + vendor: Broadcom Inc. and subsidiaries + ib0: + driver: mlx4_core + firmware_version: 2.36.5000 + guid: '0x7cfe900300bf0d41' + interface: InfiniBand + ip: 172.18.131.1 + mac: 7c:fe:90:03:00:bf:0d:41 + management: false + model: MT27500 Family [ConnectX-3] + name: ib0 + rate: 40000000000 + vendor: Mellanox Technologies + operating_system: + cstate_driver: intel_idle + cstate_governor: menu + ht_enabled: false + pstate_driver: intel_cpufreq + pstate_governor: performance + turboboost_enabled: true + processor: + cache_l1d: 32768 + cache_l1i: 32768 + cache_l2: 262144 + cache_l3: 10485760 + ht_capable: true + instruction_set: x86-64 + microcode: '0xb000040' + model: Intel Xeon + other_description: Intel(R) Xeon(R) CPU E5-2623 v4 @ 2.60GHz + vendor: Intel + version: E5-2623 v4 + storage_devices: + pci-0000:03:00.0-scsi-0:2:0:0: + by_id: "/dev/disk/by-id/wwn-0x61866da07c7090001fcad360885bea08" + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" + firmware_version: 4.3 + model: PERC H730 Mini + size: 399431958528 + storage: HDD + pci-0000:03:00.0-scsi-0:2:1:0: + by_id: "/dev/disk/by-id/wwn-0x61866da07c7090001fcad58ba96aaf8a" + by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" + firmware_version: 4.3 + model: PERC H730 Mini + size: 399431958528 + storage: HDD + supported_job_types: + virtual: ivt -- GitLab From a85400cbd0bcd1866900d967ae24f496dd250768 Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Thu, 13 Mar 2025 09:55:01 +0100 Subject: [PATCH 5/7] [sophia][esterel1] Add informations about the cluster --- .../sophia/clusters/esterel1/esterel1.json | 10 +++--- .../clusters/esterel1/nodes/esterel1-1.json | 27 +++++++++----- .../sophia/clusters/esterel1/esterel1.yaml | 36 ++++++++++++------- .../sophia/clusters/esterel1/esterel1_ib.yaml | 2 -- .../clusters/esterel1/nodes/esterel1-1.yaml | 2 +- 5 files changed, 47 insertions(+), 30 deletions(-) diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json index b9b8568cd59..af21f8ff3b2 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -1,16 +1,16 @@ { "boot_type": "uefi", - "created_at": "Thu, 01 Jan 1970 00:00:00 GMT", + "created_at": "Thu, 13 Mar 2025 00:00:00 GMT", "exotic": false, "kavlan": false, - "manufactured_at": "1970-01-01", + "manufactured_at": "2016-11-01", "metrics": [ ], - "model": "Cluster Model", + "model": "Dell PowerEdge R730", "nodes_count": 1, "nodes_description": "2 CPUs Intel Xeon E5-2623 v4, 4 cores/CPU, 32GB RAM, 2x372GB HDD, 1 x 1Gb Ethernet, 1 x 40Gb InfiniBand", - "priority": 197001, + "priority": 201611, "queues": [ "admin", "testing" @@ -18,5 +18,5 @@ "redfish": true, "type": "cluster", "uid": "esterel1", - "warranty_end": "1970-01-01" + "warranty_end": "2021-11-01" } \ No newline at end of file diff --git a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json index 4a69e5440b9..00ccd88379d 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.json @@ -11,20 +11,20 @@ "vendor": "Dell Inc.", "version": "2.19.0" }, - "bmc_version": "2.86", + "bmc_version": "2.86.86.86", "chassis": { - "manufactured_at": "1970-01-01", + "manufactured_at": "2016-11-01", "manufacturer": "Dell Inc.", "name": "PowerEdge R730", "serial": "88Y9RG2", - "warranty_end": "1970-01-01" + "warranty_end": "2021-11-01" }, "exotic": false, "main_memory": { "ram_size": 34359738368 }, "management_tools": { - "bmc_vendor_tool": "ipmitool", + "bmc_vendor_tool": "racadm", "ipmitool": { "retries": 5 } @@ -78,12 +78,15 @@ { "device": "eth1", "driver": "tg3", + "enabled": false, "firmware_version": "FFV22.61.8 bc 5720-v1.39", "interface": "Ethernet", "kavlan": false, "mac": "18:66:da:ed:2c:45", "management": false, "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "mountable": false, + "mounted": false, "name": "eno2", "sriov": false, "sriov_totalvfs": 0, @@ -92,12 +95,15 @@ { "device": "eth2", "driver": "tg3", + "enabled": false, "firmware_version": "FFV22.61.8 bc 5720-v1.39", "interface": "Ethernet", "kavlan": false, "mac": "18:66:da:ed:2c:46", "management": false, "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "mountable": false, + "mounted": false, "name": "eno3", "sriov": false, "sriov_totalvfs": 0, @@ -106,12 +112,15 @@ { "device": "eth3", "driver": "tg3", + "enabled": false, "firmware_version": "FFV22.61.8 bc 5720-v1.39", "interface": "Ethernet", "kavlan": false, "mac": "18:66:da:ed:2c:47", "management": false, "model": "NetXtreme BCM5720 2-port Gigabit Ethernet PCIe", + "mountable": false, + "mounted": false, "name": "eno4", "sriov": false, "sriov_totalvfs": 0, @@ -160,8 +169,8 @@ "turboboost_enabled": true }, "performance": { - "core_flops": 128, - "node_flops": 1024 + "core_flops": 16000000000, + "node_flops": 128000000000 }, "processor": { "cache_l1": null, @@ -169,10 +178,10 @@ "cache_l1i": 32768, "cache_l2": 262144, "cache_l3": 10485760, - "clock_speed": 8, + "clock_speed": 2000000000, "ht_capable": true, "instruction_set": "x86-64", - "microarchitecture": "Haswell", + "microarchitecture": "Sandy Bridge", "microcode": "0xb000040", "model": "Intel Xeon", "other_description": "Intel(R) Xeon(R) CPU E5-2623 v4 @ 2.60GHz", @@ -182,7 +191,7 @@ "redfish": true, "software": { "forced-deployment-timestamp": 202007300948, - "postinstall-version": "1.2025011610", + "postinstall-version": "1.2025021810", "standard-environment": "debian11-x64-std" }, "storage_devices": [ diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml index 8999f20a787..f10d9bb4157 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1.yaml @@ -1,24 +1,24 @@ --- -model: Cluster Model # TODO: change this value. -created_at: 1970-01-01 # TODO: change this value +model: Dell PowerEdge R730 +created_at: 2025-03-13 kavlan: false -boot_type: uefi # TODO: specify if 'uefi' (ideally) or 'bios' (legacy, if no other choice) -exotic: false # TODO: specify if 'true' or 'false' +boot_type: uefi +exotic: false queues: - admin - testing nodes: esterel1-1: chassis: - manufactured_at: 1970-01-01 # TODO: Put date. - warranty_end: 1970-01-01 # TODO: Put date. + manufactured_at: 2016-11-01 + warranty_end: 2021-11-01 supported_job_types: deploy: true besteffort: true max_walltime: 0 processor: - microarchitecture: Haswell # TODO: replace with microarch name. - clock_speed: 8 # TODO: Replace with clock speed. + microarchitecture: Sandy Bridge + clock_speed: 2000000000 network_adapters: bmc: interface: Ethernet @@ -29,17 +29,27 @@ nodes: enabled: true mountable: true mounted: true + eth1: + enabled: false + mountable: false + mounted: false + eth2: + enabled: false + mountable: false + mounted: false + eth3: + enabled: false + mountable: false + mounted: false storage_devices: - pci-0000:03:00.0-scsi-0:2:0:0: # This field will have to be renamed later. + pci-0000:03:00.0-scsi-0:2:0:0: id: disk0 interface: SAS - by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:0:0" # this path will have to change later. pci-0000:03:00.0-scsi-0:2:1:0: id: disk1 interface: SAS - by_path: "/dev/disk/by-path/pci-0000:03:00.0-scsi-0:2:1:0" software: - standard-environment: debian11-x64-std # TODO: check that architecture is OK + standard-environment: debian11-x64-std management_tools: - bmc_vendor_tool: ipmitool # TODO: replace with bmc_vendor_tool (ipmitool, racadm) + bmc_vendor_tool: racadm nodeset: esterel1 diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml index 74090dcbcc2..550cb30e48a 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_ib.yaml @@ -5,6 +5,4 @@ nodes: enabled: true mountable: true mounted: true - management: false - interface: InfiniBand netmask: 255.255.240.0 diff --git a/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml b/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml index e1eb832b0d1..0857a243955 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/nodes/esterel1-1.yaml @@ -11,7 +11,7 @@ esterel1-1: release_date: 12/12/2023 vendor: Dell Inc. version: 2.19.0 - bmc_version: '2.86' + bmc_version: '2.86.86.86' chassis: manufacturer: Dell Inc. name: PowerEdge R730 -- GitLab From f20565a7cb8c57abc038a0a6f1268873ff91a0f9 Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Thu, 13 Mar 2025 14:07:19 +0100 Subject: [PATCH 6/7] [sophia][esterel1] Add metrics --- .../sophia/clusters/esterel1/esterel1.json | 73 ++++++++++++++++++- .../clusters/esterel1/esterel1_metrics.yaml | 60 +++++++++++++++ 2 files changed, 132 insertions(+), 1 deletion(-) create mode 100644 input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json index af21f8ff3b2..4671086449e 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -5,7 +5,78 @@ "kavlan": false, "manufactured_at": "2016-11-01", "metrics": [ - + { + "description": "Default subset of metrics from Prometheus Node Exporter", + "name": "prom_default_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "id": [ + "node_boot_time_seconds", + "node_cpu_scaling_frequency_hertz", + "node_cpu_seconds_total", + "node_filesystem_free_bytes", + "node_filesystem_size_bytes", + "node_load1", + "node_load15", + "node_load5", + "node_memory_Buffers_bytes", + "node_memory_Cached_bytes", + "node_memory_MemAvailable_bytes", + "node_memory_MemFree_bytes", + "node_memory_MemTotal_bytes", + "node_memory_Shmem_bytes", + "node_memory_SwapFree_bytes", + "node_memory_SwapTotal_bytes", + "node_network_receive_bytes_total", + "node_network_receive_packets_total", + "node_network_transmit_bytes_total", + "node_network_transmit_packets_total", + "node_procs_blocked", + "node_procs_running", + "kwollect_custom" + ], + "port": 9100, + "protocol": "prometheus" + } + }, + { + "description": "All metrics from Prometheus Node Exporter", + "name": "prom_all_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "port": 9100, + "protocol": "prometheus" + } + }, + { + "description": "Default subset of metrics from Prometheus Nvidia DCGM Exporter", + "name": "prom_nvgpu_default_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "id": [ + "DCGM_FI_DEV_SM_CLOCK", + "DCGM_FI_DEV_MEM_CLOCK", + "DCGM_FI_DEV_GPU_TEMP", + "DCGM_FI_DEV_POWER_USAGE", + "DCGM_FI_DEV_MEM_COPY_UTIL" + ], + "port": 9400, + "protocol": "prometheus" + } + }, + { + "description": "All metrics from Prometheus Nvidia DCGM Exporter", + "name": "prom_nvgpu_all_metrics", + "optional_period": 15000, + "period": 0, + "source": { + "port": 9400, + "protocol": "prometheus" + } + } ], "model": "Dell PowerEdge R730", "nodes_count": 1, diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml new file mode 100644 index 00000000000..e2714b591b6 --- /dev/null +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml @@ -0,0 +1,60 @@ +--- +metrics: + - name: prom_default_metrics + description: Default subset of metrics from Prometheus Node Exporter + period: 0 + optional_period: 15000 + source: + protocol: prometheus + port: 9100 + id: + - node_boot_time_seconds + - node_cpu_scaling_frequency_hertz + - node_cpu_seconds_total + - node_filesystem_free_bytes + - node_filesystem_size_bytes + - node_load1 + - node_load15 + - node_load5 + - node_memory_Buffers_bytes + - node_memory_Cached_bytes + - node_memory_MemAvailable_bytes + - node_memory_MemFree_bytes + - node_memory_MemTotal_bytes + - node_memory_Shmem_bytes + - node_memory_SwapFree_bytes + - node_memory_SwapTotal_bytes + - node_network_receive_bytes_total + - node_network_receive_packets_total + - node_network_transmit_bytes_total + - node_network_transmit_packets_total + - node_procs_blocked + - node_procs_running + - kwollect_custom + - name: prom_all_metrics + description: All metrics from Prometheus Node Exporter + period: 0 + optional_period: 15000 + source: + protocol: prometheus + port: 9100 + - name: prom_nvgpu_default_metrics + description: Default subset of metrics from Prometheus Nvidia DCGM Exporter + period: 0 + optional_period: 15000 + source: + protocol: prometheus + port: 9400 + id: + - DCGM_FI_DEV_SM_CLOCK + - DCGM_FI_DEV_MEM_CLOCK + - DCGM_FI_DEV_GPU_TEMP + - DCGM_FI_DEV_POWER_USAGE + - DCGM_FI_DEV_MEM_COPY_UTIL + - name: prom_nvgpu_all_metrics + description: All metrics from Prometheus Nvidia DCGM Exporter + period: 0 + optional_period: 15000 + source: + protocol: prometheus + port: 9400 -- GitLab From ae72036a78d34039c32dfa14eca4fd81b651a50e Mon Sep 17 00:00:00 2001 From: Hugo Dominois <hugo.dominois@inria.fr> Date: Fri, 14 Mar 2025 09:25:33 +0100 Subject: [PATCH 7/7] [sophia][esterel1] Remove GPU metrics for prometheus --- .../sophia/clusters/esterel1/esterel1.json | 27 ------------------- .../clusters/esterel1/esterel1_metrics.yaml | 20 -------------- 2 files changed, 47 deletions(-) diff --git a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json index 4671086449e..a203cbfc115 100644 --- a/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json +++ b/data/grid5000/sites/sophia/clusters/esterel1/esterel1.json @@ -49,33 +49,6 @@ "port": 9100, "protocol": "prometheus" } - }, - { - "description": "Default subset of metrics from Prometheus Nvidia DCGM Exporter", - "name": "prom_nvgpu_default_metrics", - "optional_period": 15000, - "period": 0, - "source": { - "id": [ - "DCGM_FI_DEV_SM_CLOCK", - "DCGM_FI_DEV_MEM_CLOCK", - "DCGM_FI_DEV_GPU_TEMP", - "DCGM_FI_DEV_POWER_USAGE", - "DCGM_FI_DEV_MEM_COPY_UTIL" - ], - "port": 9400, - "protocol": "prometheus" - } - }, - { - "description": "All metrics from Prometheus Nvidia DCGM Exporter", - "name": "prom_nvgpu_all_metrics", - "optional_period": 15000, - "period": 0, - "source": { - "port": 9400, - "protocol": "prometheus" - } } ], "model": "Dell PowerEdge R730", diff --git a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml index e2714b591b6..729e9beb582 100644 --- a/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml +++ b/input/grid5000/sites/sophia/clusters/esterel1/esterel1_metrics.yaml @@ -38,23 +38,3 @@ metrics: source: protocol: prometheus port: 9100 - - name: prom_nvgpu_default_metrics - description: Default subset of metrics from Prometheus Nvidia DCGM Exporter - period: 0 - optional_period: 15000 - source: - protocol: prometheus - port: 9400 - id: - - DCGM_FI_DEV_SM_CLOCK - - DCGM_FI_DEV_MEM_CLOCK - - DCGM_FI_DEV_GPU_TEMP - - DCGM_FI_DEV_POWER_USAGE - - DCGM_FI_DEV_MEM_COPY_UTIL - - name: prom_nvgpu_all_metrics - description: All metrics from Prometheus Nvidia DCGM Exporter - period: 0 - optional_period: 15000 - source: - protocol: prometheus - port: 9400 -- GitLab