How Symantec Broke Backup Exec
Symantec’s Backup Exec 2012, released in February of this year, is causing a lot of frustration in the IT community and a lot of backup admins want to go back to the previous version.
Assuming, of course, they don’t look elsewhere for a backup solution.
Here’s a summary of the changes causing frustration with Backup Exec 2012:
Changing From a “Job-Centric” to a “Server-Centric” Approach
Going from a “job-centric” approach to a “server-centric” one isn’t inherently bad, but it’s a mismatch for how backups were being done, by a lot of people. Specifically people using tape.
It seems like BE did this for two reasons. First, to allow for job parallelization. Basically, allowing all servers to back up at once, instead of working down the list of servers in a single job in series. Second, to allow for better reporting – so that error reports wouldn’t read, “something in this job failed” but instead, “something on Server X failed.”
Both of these changes seem logical, but admins in BE’s huge installed base, who”™ve used it for years, had actually had spent a lot of time creating custom jobs and now they have to go back to square one and re-configure all the backups — even ones they were perfectly happy with.
Removal of Specific Configurations
The product team at BE must have thought “Well, NOBODY does THAT anymore!” when they decided to remove a bunch of features that people are still using. Things like B2D for device libraries, backing up multiple servers in a single job to tape, or using a folder on your local disk as a backup target (it wants the whole partition).
Generally Very Buggy
This is not totally unexpected given the breadth of the changes. They changed the agent, the overall agent taxonomy, the way backups are configured, the way restores are executed, the way the storage media is configured, and the UI. Phew.
The worst part is that it really feels like this product was not ready. What’s important to understand is the foundation of this product cut its code tree in the 1980’s, and some of that still lives down deep. And so with each release, things tend to get slower, more bogged down, and with a larger footprint.
The really unforgivable thing here is that, from an anecdotal perspective, the support folks at Symantec seem to have thrown up their hands and are saying, “You’d better start from scratch and re-configure your backups all over again. The conversion process [from “job” to “server”] doesn’t work.”
Extensive Changes To The UI
Some folks simply do not like the extent to which the UI has been consumerized. Though others are focused on how it’s missed a few use cases completely, “The system doesn’t help arbitrate or setup the jobs in a way that prevents serious clashes for resource” according to a poster on Spiceworks(registration required).
To compound the problem, simply declining to upgrade from an older version of BE isn’t a long-term solution. This is because, as Microsoft customers know well, you can’t count on patches and other tech support continuing for older versions indefinitely, or at all.
A lot has changed in the world of backup in the past year — the move from tape to disk to online, from silo backups to multi-purpose snapshot & replication, more data than ever, and, of course, incessant trimming of CapEx and OpEx from the budget.
So, as long as you’re being “asked” to move your backup to a significantly different paradigm, take a step back and ask, “What do we want from our backup? How can it play a greater role in our business? And how can we simplify backup from the IT perspective?”
Even if you’re not using Symantec Backup Exec, and even if you haven’t migrated to the 2012 version, you may have been wondering if there’s a better way to do things. This is a great time to check out alternatives.