Is it just me, or does anyone else absolutely HATE the redesign of Backup Exec? I have worked with BE since version 8, and I have become acutely familiar with the menus, where everything is, and how it works.
This redesign of the UI reminds me of the differences between Microsoft Office 2003 and Office 2007, only much worse. Menus are now hidden behind other menus, and everything has a completely counter-intuitive feel.
At first, I thought that the feeling would pass as I grew more familiar with the product, but in fact my dislike has grown as I have found more issues.
Does anyone else feel the same way?
I have a 16 slot Quantum autoloader that is completely useless because every tape gets an end marker unreadable errors. Backup to disk speed is significantly slower than BE2010. Inability to run simple one-time backups with options like choosing a last modified file date. Inability to create a recurring job that runs a differential every night - some of my servers simply don't need a full backup on a regular schedule.
I could live with the server centric nature of the product, but I can't live with the lost flexibility in terms of choosing custom backup options and I certainly can't live with a product that's obviously incompatible with my Quantum Superloader.
It's flawed and to make matters worse should never have been released since it clearly isn't ready for production.
Edit: Since I can only back up to disk, and since I can't run simple differential backups I had to spring for an 8TB NAS just to get basic data protection. There are probably free products out there that are better than this. Does Spiceworks have a backup feature, lol?
I just wanted to give you an update on my request for user feedback.
I had 9 participants who looked at the mock-ups of the new workflow. Everyone gave me a lot of great feedback which I’m taking back to the developers.
I wanted to thank everyone who participated! I really appreciate everyone’s input!
Thought I would post this here because this aggravating to me. I recently downgraded from 2012 back to 2010. Afterwards I thought I would write detailed instructions for other users who might want to do this. I created an article and then as a precaution also made a new forum post (reply if they take down my post - I've already mirrored it elsewhere :) in case the article didnt get approved.
Today I got the following message from someone at Symantec.
The message states
Hope you are doing good. I've come across your recent article "How to downgrade from BE2012 to BE2010R3". I'd like to inform you that the article will not be published. Also, as a gentle request, please let us know if we can un-publish the forum posting too.
Just thought I would share :)
Wow... really... wow...
This is probably from the same product team that came up with the new BE interface...
I guess the users of BE 2010 will have to go underground for community based support. Maybe Veeam could sponser us.
Great, someone else with the same complete show-stopping issue as me (End Marker Unreadable). My 16 slot loader also has a Quantum drive. I've been away from work for 2 days and the robotics spent the full 48 hours loading and unloading the same tape which was not overwritable. HOW COULD THAT BE A GOOD IDEA? If I haven't inserted a new one and re-scanned the slots, then what is the point of loading and unloading ad infinitum?
Anyway, I have been asked to try disabling hardware encryption on all jobs to see if that's where the problem lies. I'll report back with my findings.
@zoobadger, are you using hardware encryption?
If this is the source of the problem, what I would like to know, is HOW did Symantec get through multiple alphas, RCs and even the damn beta test with no one enabling encryption? Seriously.
Received the replacement LTO drive Tuesday and re-ran my full backups last night. I'm happy to report no errors. I had one DFS job that was just under 1TB and it dup'd to tape without issue.
I applied SP1a yesterday....
I also had success with a test job of one of my MS Hyper-V HA VM's on my SAN Cluster.
I'll be running the 1.5TB HV HA VM job over the weekend over the weekend to tape and see how that fairs.
29 servers dedup'd then duplicate to tape ran last night without any errors.....Yippee!
For anyone interested about my setup:
Dell R710 with a TL2000 Autoloader - "1 IBM ULT3580-TD3 drive"
OS - Windows Server 2008 Ent x64 - SP2
Backup Exec 2012 SP1a
I can't address the Quantum autoloader issues you're experiencing -- and I realize that's the biggest issue you're facing.
But, regarding the loss of backup flexibility you've mentioned, I wonder if you've found or tried a few things:
- One-Time Backup using at last modified date: you'll find this option in the selections detail. If you create a backup (either recurring or one-time), click "Edit" on the selection summary, and then click on the "Selection Details" tab. You'll then be able to insert a file selection with all the options you've had in previous releases of Backup Exec:
In the recurring backup dialog, on the "Schedule" page, you will find an option on the Full backup to "Run now with no recurring schedule". If you use that option, and then schedule a Differential every day in the same Backup Definition, you're almost where you want to be.
In previous releases, we did allow Differential or Incremental-only standalone (recurring) backup jobs, but technically the first one had to be equivalent to a Full...we've made that fact explicit with 2012 (and the first backup sets are accurately marked as Full sets).
I'm curious: what is your recovery plan using a Differential-only backup scheme?
I have two servers that rarely need full backups. So in BE2010 I'd run a one-time manual full backup to tape when the backup window on the daily differential to disk became excesive. This could be two months, or six months depending on circumstances. Maybe this violates the purity of the best practices but it was a reasonable approach and worked for us.
But that's nothing compared to the fact that the Superloader just doesn't work with BE2012. There's the end marker unreadable problem and also I'm getting about 700MB per minute throughput which is unacceptable. And all kinds of read errors in spite of the fact that the device is brand new as are all the tapes. Even erasing a tape generates all kinds of complex errors. It's maddening.
I don't have any criticism of your full/diff rotation strategy -- just wanted to understand it. The scheduling options in 2012 should satisfy any requirements you have (except you'll have to run at least one full after upgrading). I have heard stories of customers doing their full with one product and their diff/incremental backups with BE. That seems like a recipe for trouble -- and it doesn't sound like you're doing that.
I'm trying to track down someone who could help with the autoloader issues you're seeing; that's beyond my scope of expertise.
Never before we had to invest so much time to get all of our customers which we upgraded to 2012 back on the run and dealing with working backups. After now 5 weeks of bug and failure treating we've got tired and lost all confidence.
We hoped to get a straightforwared product and got a sick dinasaur. Slow, buggy and tricky.
I really hope the symantec reads all this from hereand other blogs such as spiceworks...
I'm trough with it.
I started this thread because I, like many of you, was extremely dissatisfied with my initial impression of BackupExec 2012. My particular areas of concern were the new GUI and the "server-centric" paradigm shift.
This thread has gotten pretty large with most of you echoing my sentiments and adding many of your own.
Let me tell you, unequivocally, that Symantec has read these comments and has heard our pain loud and clear. They just flew some of us down to Orlando to meet with the team personally, to voice our greivances.
People, just hang in there. Some positive changes have taken place already with the release of Service Pack 1 earlier this week. Many, many more are in the works now - things like a return of the familiar "Job Monitor" page that we all rely on.
The server-centric concept is not going away. However, a system is being developed that lets you handle a sequence of backup jobs like a single entity - pretty much how we used to.
I came to Orlando fairly disgruntled, but I am leaving confident that Symantec is on the right track, and that new releases of BackupExec are once again going to be better than they were before.
For anyone that's still ticked off, if you get a chance, I would recommend that you participate in the upcoming beta for the next revision. I guarantee you will be much happier with it than with what you saw in the original release.
Who are you and what have you done with the real Bulbous?! What did they feed you in Orlando?
Just kidding! Couldn't resist. I guess BE 2012 R2 will be upon us fairly soon.
In other news, the intermittent End Marker Unreadable on tapes does now appear to be caused by a glitch with BE 2012's handling of hardware AES encryption support. By temporarily running the jobs with software encryption I haven't had a recurrence. Unfortunately pre-existing encrypted tapes written using BE 2010 R3 seem to be detected with unreadable end markers for the time being. Apparently an orphan fix is due out soon, with a proper hotfix to follow later once it's been through testing. How it took three whole weeks to get to this realisation is another matter...
EDIT - There's now a technote for this issue.
What I don't get though is that most decent tape drives do AES in hardware (i.e. there's no performance hit to having it enabled). Many customers will be sending their media for off-site storage with a 3rd party, and those tapes are often collected from relatively insecure reception desks or post rooms from which they could be stolen. So surely it follows that many customers will be encrypting their data. So why on earth is Symantec presumably running their QA testing with encryption disabled?
I, too, attended the User Feedback Summit last week in Orlando, and have great hopes for the future of Backup Exec again. As Bulbous mentioned above, they are listening to this, and all forum posts, and are making great attempts to correct the functionality issues we've listed. With the latest release of SP1a, some of the items have already been corrected, and many more useability features that we've come to love, and sometimes hate, will be returning soon.
My biggest pain points were the server-centic mode we were forced to use, and the inability to exclude dates on a per job level.
As Bulbous explained above, they are still moving forward with the server-centric mode, but will be giving us a way to senquence the jobs for ease of monitoring and setup. This will allow us to manage our jobs, near the way we used too, but allow for enhanced features that only a sever-centic mode can provide.
The Exclude Date per Job functionality is back!
And, there a many more changes coming that will enhance the UI's and job performance.
I would also encourage you to keep a look out for the next beta release. It should resolve 99% of everything that has been mentioned in this forum post that has not yet been corrected, while also forging the next-generation backup and disaster recovery application.
Thank you to everyone at Symantec for bringing us down and taking the time to listen to us.