Cloud Storage Gateway:Alerts

Procedure

1. Method 1: View alerts for all gateways.

   1. Go to the Alerts page in the CSG console.

   2. On the Alerts page, filter alerts by Alert Time, Gateway ID, Share Name, or Alert Status.

2. Method 2: View alerts for a single gateway.

   1. Go to the Gateways page in the CSG console. Find and click the target file gateway, and then click the Alerts tab.

   2. On the Alerts page, filter alerts by Alert Time, Share Name, or Alert Status.

Insufficient metadata cache space

Cause

A cache disk is composed of data space and metadata cache space. The metadata cache space stores metadata, such as directory structures, for an SMB or NFS share. The more files a share contains, the more metadata cache space it consumes. By default, 20% of the cache disk is allocated for metadata. This alert is triggered when the available metadata cache space drops below 3 GB or 20% of the total metadata cache space.

Solution

To prevent write failures, expand the share's cache disk to increase its metadata cache space. For more information, see Expand a cache.

Throttled

Cause

This alert is triggered when the amount of data written concurrently to the cache disk exceeds the high-water mark, which is 80% of the data space on the cache disk. This can cause subsequent write operations to fail.

Solution

1. Investigate any data upload issues that might prevent cache space from being released. Uploads may fail in the following situations:

   • The OSS bucket does not exist.

   • An OSS bucket policy restricts access from the gateway.

   • The storage class of the OSS bucket is Archive or Cold Archive.

   • The total size of files written concurrently exceeds the available cache disk capacity, preventing files from being closed and uploaded promptly.

2. If applicable, adjust your business workload to reduce the amount of data written concurrently to the cache disk.

3. Expand the share's cache disk to increase the data space. For more information, see Expand a cache.

Backlogged Upload Queue

Cause

Possible causes include:

• The OSS bucket does not exist.

• An OSS bucket policy restricts access from the gateway.

• The storage class of the OSS bucket is Archive or Cold Archive.

Solution

Follow these steps to troubleshoot the issue:

1. Verify that the OSS bucket exists. If not, configure the share to use an existing OSS bucket.

2. Check for OSS bucket policies that prevent gateway access.

3. Verify that the storage class of the OSS bucket is not Archive or Cold Archive. Unless you have specific requirements, we recommend that you use the Standard storage class for your OSS bucket.

4. If the upload queue length is less than 1,000, or if it shows a decreasing trend after the alert occurs, the gateway is likely operating normally.

   Note

   To view upload queue details:

   • For gateway versions 1.9.2 and later, CloudMonitor provides upload queue metrics for file gateways, which you can view in the CloudMonitor console. For more information, see CloudMonitor integration.

   • If your gateway version is earlier than 1.9.2, you can view the upload queue length in the CSG console.

     • Go to the Gateways page in the CSG console and click the target gateway.

     • On the Shares tab, find the target share and click the + icon. The Upload Queue length is displayed in the expanded details.

If the issue persists after you perform these troubleshooting steps or if you have further questions, submit a ticket.