How to find out the container's memory usage reported by operation system in VMware Tanzu Application Service (TAS) for VMs

Products

• VMware Tanzu Application Service for VMs

Issue/Introduction

The container's resource usage can be found by running the command "cf app app_name".

However, some user might be interested in how these values are retrieved. This Knowledge Base (KB) article shows where the memory usage comes from.

Environment

Product Version: 2.8

Resolution

The container's memory usage is reported by cgroup and can be retrieved using the following procedures.

1. Retrieve the guid of the app.

######MBP:Downloads ###### cf app cook --guid
aee32c4b-####-####-a74d-695676528000

2. Find out the diego_cell where this app instance is running.

#####MBP:Downloads ###### cf curl /v2/apps/aee32c4b-####-####-a74d-695676528000/stats
{
   "0": {
      "state": "RUNNING",
      "isolation_segment": null,
      "stats": {
         "name": "cook",
         "uris": [
            "cookie.apps.####.apj-###.#####.com"
         ],
         "host": "10.##.##.##",
         "port": 61068,
         "uptime": 2180,
         "mem_quota": 1073741824,
         "disk_quota": 1073741824,
         "fds_quota": 16384,
         "usage": {
            "time": "2020-07-03T02:06:12+00:00",
            "cpu": 0.008487803282072376,
            "mem": 258871171,
            "disk": 145375232
         }
      }
   }
}

3. bosh ssh into this diego_cell.

4. Find out the instance_id of this app's container with a filter of the guid.

diego_cell/8b9fbe9f-####-####-a5af-525df19d920e:~# cfdot actual-lrps |grep aee32c4b-642c-####-####-695676528000|jq .
{
  "process_guid": "aee32c4b-####-####-a74d-695676528000-647e08f1-####-####-ae9e-7a4719552a60",
  "index": 0,
  "domain": "cf-apps",
  "instance_guid": "3b09f328-####-####-6781-ad69",
  "cell_id": "8b9fbe9f-####-####-####-525df19d920e",
  "address": "10.###.##.##",
  "ports": [
    {
      "container_port": 8080,
      "host_port": 61068,
      "container_tls_proxy_port": 61001,
      "host_tls_proxy_port": 61070
    },
    {
      "container_port": 2222,
      "host_port": 61069,
      "container_tls_proxy_port": 61002,
      "host_tls_proxy_port": 61071
    }
  ],
  "instance_address": "10.###.###.153",
  "preferred_address": "HOST",
  "crash_count": 1,
  "crash_reason": "CELL/SSHD: Exited with status 137",
  "state": "RUNNING",
  "since": 1593739792034621000,
  "modification_tag": {
    "epoch": "6e4db726-####-####-7dc0-871cdc210b72",
    "index": 5
  },
  "presence": "ORDINARY"

5. Go to the cgroup folder of this app instance with the instance_guid, /sys/fs/cgroup/memory/system.slice/garden.service/garden/instance_guid.

diego_cell/8b9fbe9f-####-####-####-525df19d920e:~# cd /sys/fs/cgroup/memory/system.slice/garden.service/garden/3b09f328-0053-####-####-ad69

6. Print out the memory usage, memory.usage_in_bytes.

diego_cell/8b9fbe9f-e883-####-####-525df19d920e:/sys/fs/cgroup/memory/system.slice/garden.service/garden/3b09f328-0053-####-####-ad69# more memory.usage_in_bytes
281792512

7. Observe that It matches the output of the cf app output:

DavidZhouMBP:Downloads dzhou$ cf app cook
Showing health and status for app cook in org system / space system as admin...
 
 
name:              cook
requested state:   started
routes:            cookie.apps.####.apj-###.#####.com
last uploaded:     Tue 30 Jun 12:07:01 CST 2020
stack:             cflinuxfs3
buildpacks:        client-certificate-mapper=1.11.0_RELEASE container-security-provider=1.16.0_RELEASE
                   java-buildpack=v4.26-offline-https://github.com/cloudfoundry/java-buildpack.git#e06e00b java-main java-opts java-security
                   jvmkill-agent=1.16.0_RELEASE open-jdk...
 
 
type:           web
instances:      1/1
memory usage:   1024M
     state    since                  cpu    memory        disk          details
#0   running   2020-07-03T01:29:52Z   0.9%   246.9M of 1G   138.6M of 1G