How to Avoid System Outages & Most Common Severity Level 1 Issues: Part 2
In Part 1, we explored how a Severity 1 issue can cause more than down time, it can also drag your organization’s reputation down. But Severity 2 is no joke either. DataCore takes outages very seriously and our teams work hard to continually update our documentation and products to help you avoid outages at every level. Let’s take a look at what those are and how you can avoid them.
The second most common site outage is because the disk pools on both sides of mirrored virtual disks become full at the same time. This can be avoided by adding tasks that alert the administrators, and of course monitoring the disk pools, or by configuring the pools to have different amounts of storage space allocated to the pool. When pools have the same storage space allocation, they will fill at the same time due to the way writes are mirrored to those pools.
When a disk pool’s storage capacity is completely used up, it will be marked as ‘Full’ and all virtual disks that have storage sources from that disk pool will be marked as ‘failed’. No more writes can be accepted for storage sources from this disk pool. If the virtual disks are mirrored and the mirror partner’s disk pool is not yet full, then all IO will be redirected via the DataCore server’s mirror paths to the mirror partner. The virtual disk will be marked as ‘Failed Redundancy’. If the virtual disks are mirrored and the mirror partner’s disk pool is also full then the storage sources will be in ‘double failure’.
Read the entire article here, How to Avoid System Outages & Most Common Severity Level 1 Issues: Part 2
Via the fine folks at DataCore Software.