Storage UI 2 Release

Changes in Storage UI 2.3

This release features support for Swarm's new S3 Backup feed type and several new settings for managing SwarmFS behavior and performance in different implementations.

S3 Backup Feeds — Swarm allows tiering to public cloud services for easy off-premises storage for disaster recovery (DR). Amazon S3 has the widest support in the industry, so S3-compatible endpoints are the first cloud destination from Swarm. (UIS-1027) 

• On the Feeds page of the Swarm UI, S3 Backup feed can be added, which targets an existing S3 bucket. See S3 Backup Feeds.
  

  

  
  

  

• The S3 feed behaves like a Replication feed, and it populates the same real-time dashboard charts and feed reports for monitoring backups. (UIS-1027)
  

  

SwarmFS Tuning — For SwarmFS exports, several new Advanced settings are available to adjust SwarmFS for the environment: 

• Read buffer size (ReadaheadSize) allows matching the expected workload on a specific share, lowering for small and non-sequential reads, increasing for large and sequential ones. (UIS-1007)

• Parallel read buffer requests (ReadaheadCount) allows tuning the performance of large object reads; the default of 4 reflects the optimal number of threads, per performance testing. (UIS-1007)

• Maximum part size (MaxPartSize) allows increasing the part size for large (multipart) uploads to improve the throughput when applications are writing huge files. (UIS-1018)

• Collector sleep time (CollectorSleepTime) allows minimizing object consolidation by sending fewer and larger sets of data to Swarm (at the expense of both RAM and read performance) if an implementation is sensitive to how quickly the Swarm health processor consolidates objects, which cannot be guaranteed. (UIS-1018)

• Elasticsearch buffer refresh time (ESBufferRefreshTime) allows tuning how rapidly non-SwarmFS object updates are reflected in SwarmFS listings. Lower to reduce the wait for consistency, at the cost of increased load on Elasticsearch. (UIS-1037)

In addition, these issues are fixed:

• Issues existed with feeds defined to use a non-default admin password. (UIS-759)

• Clicking Add Export on the NFS page caused an immediate 500 error if a cluster had no search feed defined. (UIS-441)

Watch Items and Known Issues

• Feeds can be created but they cannot be updated via Swarm UI 2.3 because of changes to handling of feed definitions when using Gateway 5.4. (UIS-1033)

• On the Chassis Details page, Advanced tab, the data for Health Data and View the raw JSON may not display. (UIS-1048)

• Turn off the identify function before removing the disk from the chassis. Failure to perform this can result in the need to restart the chassis when using the Swarm UI to identify volumes. (UIS-564)

• The UI does not convey pausing cannot begin until the feed backlog is cleared, which can be a long delay when pausing a feed. (UIS-437)

Changes in Storage UI 2.2

This release features improvements to the handling of Swarm Storage settings and how they are accessed in the UI.

• Swarm settings include both cluster-wide and node-specific options, which can vary from chassis to chassis. On the Cluster Settings page, the Swarm UI now prevents erroneous changes by hiding the node-specific settings when Platform Server is not implemented and handling those nodes. (UIS-1000)

• The Cluster Settings page now includes an option to Show advanced settings. This option reveals all advanced (unpublished) settings that are dynamic (persisted) when enabled. They can be updated on a running cluster without a reboot. As before, bold fonts and Default information displays alert to settings with custom values in the cluster. (UIS-998)

Watch Items and Known Issues

• Issues exist with feeds defined to use a non-default admin password. (UIS-759)

• Turn off the identify function before removing the disk from the chassis. Failure to perform this can result in the need to restart the chassis when using the Swarm UI to identify volumes. (UIS-564)

• Clicking Add Export on the NFS page causes an immediate 500 error if a cluster has no search feed defined. (UIS-441)

• Turn off the identify function before removing the disk from the chassis. Failure to perform this can result in the need to restart the chassis when pausing a feed, which can be a long delay. (UIS-437)

Changes in Storage UI 2.1

• Improved: Swarm settings include both cluster-wide and node-specific options. On the Cluster Settings page, the Swarm UI now prevents erroneous changes by hiding the node-specific settings when Platform Server is not implemented and handling those nodes. (UIS-798)

• Improved: Turn off the identify function before removing the disk from the chassis. Failure to perform this so can result in the need to restart the chassis when access tokens expire or are deleted during an active session. (UIS-975)

• Fixed: The UI refreshes and reselects them all if deselecting nodes from the queue without Restarting when editing the Rolling Restart Queue. (UIS-957)

• Fixed: On the Details tab of the Chassis Details page, the Actions menu for each disk erroneously but harmlessly showed a testing-only command to Fail the disk. (UIS-955)

Watch Items and Known Issues

• A node may become temporarily unresponsive when attempting to view Health Data (the raw JSON of the health report) on the Advanced tab of the Chassis Details page. (SWAR-8349)

• A storage node may be reported to be in an unknown state rather than in maintenance mode while a reboot is in progress. (SWAR-8348)

• Issues exist with feeds defined to use a non-default admin password. (UIS-759)

• Turn off the identify function before removing the disk from the chassis when using the Swarm UI to identify volumes. Failure to perform this can result in the need to restart the chassis. (UIS-564)

• Clicking Add Export on the NFS page causes an immediate 500 error if a cluster has no search feed defined. (UIS-441)

• The UI does not convey pausing cannot begin until the feed backlog is cleared when pausing a feed, which can be a long delay. (UIS-437)

Changes in Storage UI 2.0

Hardware Management

Aligning with the new architecture of Swarm Storage 10, Storage UI 2.0 is extensively expanded to support the monitoring and administration of Swarm implementations, replicating the rich functionality of the legacy Admin Console (which is deprecated) and adding visibility in to Swarm's management API.

Rolling Restarts — With Platform Server installed, choose to perform a Rolling Restart, so the cluster remains fully operational, with chassis going offline one at a time to avoid service interruption. The Rolling Restart Queue allows reordering and removing chassis from the queue, monitor the progress, and cancel queued restarts. (UIS-588)

Chassis Details page features:

• Details Tab

  • The main Details tab includes counts of each disk's streams and the size of the largest stream. Watching these counts helps monitor the progress of disks being retired. (UIS-533)

  • Disks can be retired individually as well as retiring the machine (chassis) as a whole. Disk-level retires are useful for targeting bad (slow) disks and for working around having too limited capacity for retires of entire chassis. (UIS-544)

  • The affected chassis now shows the status when one or more disks is in the process of retiring, for improved visibility and tracking. (UIS-749)

• Logs Tab

  • The Logs tab allows clearing out the logs when no longer needed. (UIS-616)

• Driver Message Tab (new)

  • Driver Message displays the output from the dmesg command, which prints the message buffer of the kernel. (UIS-799)

• Hardware Info Tab (new)

  • Hardware Info displays the output of the hwinfo hardware detection tool. (UIS-799)

• Memory Tab (new)

  • The Memory tab reports details of memory usage on the specific machine, to help with capacity planning and analysis. (UIS-723)

• Statistics Tab (new)

  • The Statistics tab rolls up a detailed, expandable report combining Health Processor (HP), Communications (cluster network), and Memory usage counts and values, to help with analysis and troubleshooting. (UIS-886)

• Advanced Tab (new)
  

  

  
  

  • View the cluster's Health Data, which is the raw JSON content of the health report a cluster sends to DataCore Support. (UIS-838)

  • View and change the Log Level Settings dynamically, to simplify machine-level debugging. (UIS-835)

  • Verify the Software Revision of Swarm Storage running on the specific machine, which is useful for managing rolling upgrades. (UIS-802)

  • Use an embedded HAL Browser to explore the complete API for Swarm storage management dynamically. (UIS-836)

  • Access the Swagger.io UI. This API visualization tool also allows exploring Swarm's API for managing the storage cluster. (UIS-837)
    

    

• Dashboard

  • The dashboard now dynamically reports the amount of space available as well as space used in the storage cluster. (UIS-765)

Replication Feeds with SSL

With 2.0, Replicate via direct POST now supports SSL/TLS network encryption and standard proxy servers for replication feeds, which eliminates the need for separate VPN tunnels between clusters. This capability streamlines deployments where encrypted communications are needed over wide-area, untrusted networks. See Replicating Feeds over Untrusted Networks.

• Swarm now supports using SSL for remote replication data transfer. This configuration requires an SSL offload proxy (such as HAProxy) in the target cluster environment. (SWAR-7826)

• For sites using SSL with remote replication, Swarm allows the establishment of be self-signed trusted certificates (public keys). (SWAR-8080) See Adding a Trusted Certificate to Swarm.

• Swarm allows placing a forward proxy to the source cluster into the replication path. (SWAR-8025)

Feed Control and Monitoring

• The statuses reported on the Cluster > Feeds and Reports > Feeds pages now refresh automatically. (UIS-796)

• Feed table — To support feed troubleshooting, the Feed Settings page for a given feed now includes a command to View feed table, which displays the the SNMP repository dump for the selected node. (UIS-787)

• Domain filtering — Filter which domains to include, exclude, or both when defining new replication feeds, and specify whether to replicate any unnamed objects not tenanted in any domain. The domain filters support wildcard matching for ease of maintenance. (UIS-709) 
  

  

Elasticsearch Cluster Status

Research the ES cluster status on the Elasticsearch Reports page if the Elasticsearch panel on the Dashboard shows a problem.

These reports generate on demand and allow drilling in to details spanning the ES nodes, thread pools, indices, and shards. See Using Cluster Reports.

Additional Changes

• Improved: The cluster-level Health Report now totals stream counts for individual machines (chassis) and for the entire cluster. (UIS-716)

• Improved: The Swarm UI now supports being served on ports other than 91, as governed by the cluster_admin bind port set in the Gateway configuration. This change allows use of binding needed for the environment, such as for a proxy or in a Docker environment.  (UIS-934)

• Fixed: Storage UI did not display search feed configuration if the Elasticsearch service was paused. (UIS-766)

Watch Items and Known Issues

• The UI reselects all nodes if deselecting them from the queue but do not launch the Restart when editing the Rolling Restart Queue. (UIS-957)

• A node may become temporarily unresponsive when attempting to view Health Data (the raw JSON of the health report) on the Advanced tab of the Chassis Details page. (SWAR-8349)

• A storage node may be reported to be in an unknown state rather than in maintenance mode while a reboot is in progress. (SWAR-8348)

• On the Details tab of the Chassis Details page, the Actions menu for each disk erroneously shows a testing-only command to Fail the disk. Selecting the command causes errors to display but does not affect the disk (UIS-955)

• Issues exist with feeds defined to use a non-default admin password. (UIS-759)

• Turn off the identify function before removing the disk from the chassis when using the Swarm UI to identify volumes. Failure to perform this can result in the need to restart the chassis. (UIS-564)

• Clicking Add Export on the NFS page causes an immediate 500 error if a cluster has no search feed defined. (UIS-441)

• The UI does not convey pausing cannot begin until the feed backlog is cleared when pausing a feed, which can be a long delay. (UIS-437)