Using IBM Cloud Monitoring with IBM Cloud Object Storage

Working with Metrics

According to the Monitoring documentation, "A metric is a quantitative measure that has one or more labels to define its characteristics." When you configure your buckets to forward data to a Monitoring instance, that data is automatically collected and available for analysis through the web UI.

Instantiating and Provisioning IBM Cloud Object Storage

If you have not already done so, set up and provision your instance of Object Storage. If you already have an instance of Monitoring to work with, verify you have either the manager or writer role to work with this guide.

Configure a bucket for metrics

Metrics are only sent to Monitoring every 24 hours. Any alert thresholds exceeded on metrics will not trigger until the next time the metrics are sent to Monitoring.

When creating a bucket, you can configure your Monitoring instance at the same time.

You can also configure an existing bucket to use the Monitoring instance by selecting configuration in the console.

Note that in each case the region for your new instance of Monitoring is automatically tied to the region of the bucket.

Note that if the Monitoring instance is created by COS, Platform metrics must be set on the Monitoring instance.

To set Platform metrics on the instance:

1. Go to your dashboard.
2. Click on "Configure platform metrics."
3. Select the region where the Monitoring instance was created.
4. Click "Configure."
5. Your Monitoring instance is now set for Platform metrics.

We will use cURL to connect to the Resource Configuration API and modify our bucket's configuration.

curl -X PATCH -k  \
  -H "authorization: Bearer $TOKEN" \
  https://config.cloud-object-storage.cloud.ibm.com/v1/b/$1 \
  -H 'cache-control: no-cache' \
  -d '{"metrics_monitoring": {
    "usage_metrics_enabled": true,
    "request_metrics_enabled": true,
    "metrics_monitoring_crn": "crn:v1:bluemix:public:sysdig-monitor:us-east:a/9xxxxxxxxxb1xxxc7fdxxxxxxxxxx5:7xxxxxxxx0-xx7x-xdx8-9fxx-123456789012::"
    }
   }'

In the code sample, replace the placeholders with the appropriate values.

Launch your instance of Monitoring

When you have created your instance of Monitoring, click on the "View Monitoring" button in the "View dashboard" column, as shown in Figure 4.

Choose a pre-built dashboard for Monitoring

This guide has skipped right past the installation of agents that is typical of most monitoring configurations. In addition, you can also skip the manual creation of building a dashboard by using a pre-built solution. As shown in Figure 2, you can choose how to present your data using one of the options given.

View your data in Monitoring

Once you've configured your dashboard, you can view your data. Figures 3-5 show different views of your usage.

Usage metrics

There are a set of basic metrics that track usage:

• ibm_cos_bucket_used_bytes
• ibm_cos_bucket_object_count
• ibm_cos_bucket_hard_quota_bytes

Request metrics

There are metrics that report the aggregates for different classes of HTTP requests:

• ibm_cos_bucket_all_requests
• ibm_cos_bucket_get_requests
• ibm_cos_bucket_put_requests
• ibm_cos_bucket_delete_requests
• ibm_cos_bucket_post_requests
• ibm_cos_bucket_list_requests
• ibm_cos_bucket_head_requests

Errors are also collected, with server-side (5xx) errors broken out:

• ibm_cos_bucket_4xx_errors
• ibm_cos_bucket_5xx_errors

The minimum, maximum, and average bytes transferred by network type are reported:

• ibm_cos_bucket_bytes_download_public_min
• ibm_cos_bucket_bytes_download_public_max
• ibm_cos_bucket_bytes_download_public_avg
• ibm_cos_bucket_bytes_download_private_min
• ibm_cos_bucket_bytes_download_private_max
• ibm_cos_bucket_bytes_download_private_avg
• ibm_cos_bucket_bytes_download_direct_min
• ibm_cos_bucket_bytes_download_direct_max
• ibm_cos_bucket_bytes_download_direct_avg
• ibm_cos_bucket_bytes_upload_public_min
• ibm_cos_bucket_bytes_upload_public_max
• ibm_cos_bucket_bytes_upload_public_avg
• ibm_cos_bucket_bytes_upload_private_min
• ibm_cos_bucket_bytes_upload_private_max
• ibm_cos_bucket_bytes_upload_private_avg
• ibm_cos_bucket_bytes_upload_direct_min
• ibm_cos_bucket_bytes_upload_direct_max
• ibm_cos_bucket_bytes_upload_direct_avg

Latency metrics (first byte and general) for requests are broken down by request type:

• ibm_cos_bucket_first_byte_latency_read_min
• ibm_cos_bucket_first_byte_latency_read_max
• ibm_cos_bucket_first_byte_latency_read_avg
• ibm_cos_bucket_first_byte_latency_write_min
• ibm_cos_bucket_first_byte_latency_write_max
• ibm_cos_bucket_first_byte_latency_write_avg
• ibm_cos_bucket_first_byte_latency_misc_min
• ibm_cos_bucket_first_byte_latency_misc_max
• ibm_cos_bucket_first_byte_latency_misc_avg
• ibm_cos_bucket_request_latency_read_min
• ibm_cos_bucket_request_latency_read_max
• ibm_cos_bucket_request_latency_read_avg
• ibm_cos_bucket_request_latency_write_min
• ibm_cos_bucket_request_latency_write_max
• ibm_cos_bucket_request_latency_write_avg
• ibm_cos_bucket_request_latency_misc_min
• ibm_cos_bucket_request_latency_misc_max
• ibm_cos_bucket_request_latency_misc_avg

All metrics are reported as float64 numeric values:

Global Attributes

The following attributes are available for segmenting all of the metrics listed above

Additional Attributes

The following attributes are available for segmenting one or more attributes as described in the reference above. Please see the individual metrics for segmentation options.