Is your Dell EMC Unity system out of storage? This video will show you how to identify if you've run out of storage, tell you what to do if you already have run out of storage, and teach you some tips and tricks for planning and maximizing your Unity storage in the future.
In this video, we'll show you how to identify if you've run out of storage for a polar resource. Tell you what to do if you already have run out of storage and teach you some tips and tricks for capacity planning and maximizing your unity storage in the future. In a typical environment, several storage resources will share capacity from the same pool when you have thinly provision resources. And IO for one or more of those resources ramps up quickly or unexpectedly, this can lead to pool over subscription when a pool is oversubscribed for a long period of time, storage resources can go read only.
So what should you do if you've already run out of space? And one of your resources is read only first. Let's review the alerts to see which storage resources are not in a healthy state. We can see from this alert that file system fs one is in a read only mode because the pool it belongs to is run out of space. We can see from the alerts page that pool one is over the capacity threshold. Let's go to the pools page to check out the storage resources for pool one and figure out what's going on on the usage tab for pool one, we can see overall and used capacity of the pool.
The used capacity is currently over the threshold and the pool is in an unhealthy state. The maroon segment of the bar graph represents the non base space used which is pool space used by snapshots and thin clones. This space accounts for a significant amount of the overall pool space on the storage resources sub ta we can see which storage resources are used in the pool and how much space they are using file system. FS one is one of the largest resources in the pool. So what do we do next to free? Some space? Here are some options. First, let's see if we can delete unus snapshots and thin clothes. We already know file system.
FS one won't read only likely because the pool is out of space. This file system is also taking up a large amount of space in pool one, let's go see if there are any old snapshots of that file system. We can delete under file, file systems, click on FS one to open its properties and review its snapshots. The general tab shows that the snapshot space used for this file system is relatively high compared to the overall size of the file system and of the pool on the snapshot sub tab, we can see a lot of snapshots if you have many older outdated snapshots and you have more recent snapshots of your resource.
You can freeze some space by deleting the older snapshots. Let's go ahead and do that for FS one. Since we have many more recent snaps now that the snaps are deleted, we can see that FS One is back in a healthy state on the pools page. Pool. One is also back in a healthy state since it now has enough free space. Another option for restoring the pool to a healthy state is to move a lun consistency group or VMFS data store to another pool that has more space. And this example will move a lun using the lun move operation. It's always a good idea to perform a move during periods of planned downtime or low host activity.
If there has already been an IO disruption because the pool is out of space, you can do a lun move to help alleviate the issue. Select a lun that is taking up a lot of space in the unhealthy pool. We'll move lun seven since it's one of the largest lines in pool one select lens seven. And from the more actions menu select, move, select another pool from the drop down that meets the minimum performance requirements to support the lun and also has enough free space.
Move lun seven to pool two, optionally, you can select the data reduction and advanced de Dulic application check boxes during one move, selecting these options will apply data reduction to already written data in the lun during the move process, further reducing space taken up by the lun on the system and increasing the Lu's overall storage efficiency. Lastly, you can expand the pool by adding physical drives. For this step, you may need additional drives. Consider this option last since you may need to actually order more drives which will take time to ship and then need to be physically added to your system before the pool can be expanded.
After you've received your new drives from the pools page, select pool one and click expand pool, select an available storage tier and change the raid configuration to support the new drives. Confirm your selections and wait for the system to expand the pool. We now see that po one has more storage available to help prevent your unity system from running out of storage in the future. Follow these best practices. First, set up oversubscription alerts for your pools, oversubscription alerts, let you know when your pool is almost out of capacity by triggering an alert.
When it reaches a certain space threshold, you should also set up an automatic snapshot deletion schedule. Make sure your deletion schedule is in alignment with any data protection backup or replication schedules. So that snapshots are replicated before deletion when necessary. If your system is connected to support, you can use IQ to monitor your capacity trends. Cloud IQ can also help you check for potentially reclaimable space on your unity system. You can subscribe to IQ alerts which notify you when the system has a significant health status change because it is approaching or exceeding capacity limitations.
For more information about cloud IQ, go to dell E MC dot com slash cloud IQ. For more information on how to perform the tasks outlined in this video or to learn more about how storage provisioning works. Refer to the Un Unisphere online help for more information and help with other aspects of your unity system. Visit the unity documentation page at dell dot com slash U docs.
