VMware Cloud Community
jarsenea
Contributor
Contributor

Anyone else having these VDR issues?

I love the fact that VMware is providing VDR as part of the vSphere package. It's definitely a step in the right direction, albeit I'm still inclined to think this software hasn't been put through the ringer in terms of proper QA. I'm just trying to put out a feeler to see how many others have experienced some of the same issues I'm having.

To start, I'm backing up my VMs via a network share on a standalone Windows 2003 server that has a NAS attached to it.

Some of the issues I've noticed:

1) Backups take an inordinate amount of time. I can understand the first backup, but my VMs don't change very much from day to day. Most of the data being manipulated is located on RDMs are these are backed up using Tivoli, not VDR (I use VDR solely for the OS partitions). Each partition is approximately 25GB, there are 15 VMs and my backup window (10pm - 6pm) isn't sufficient to complete the process.

2) Integrity checks for the backups are taking a crazy amount of time and will usually stop due to my window being closed (see point #1)

3) I'm getting inconsistent "failures" for certain VMs (the report will simply state that a VM failed to backup, not much else). It also varies per night and not always the same VMs (not exactly sure if this is related to #1 where the window is closing while VDR is executing)

4) I had the most difficult time setting up the remote share from the VDR appliance in vSphere. The username and password would never be accepted (even though if I tried the same share with the same user/pass on a Windows machine, it would work fine). I finally narrowed down the problem to the simple fact that the VDR appliance can't handle passwords that have special characters in them (this password had an "@" and a ","). Looking at the console while attempting to mount the share would spit out a CIFS error -22. Changing the password to include only numbers and letters was sufficient to work around this issue.

5) Snapshots not being created for no apparent reason and thus failing the VDR process. I'm fully able to do a manual snapshot with or without the memory state, so I'm not sure why VDR can't do it. This issue is very intermittent. I had it often when I first setup VDR, but now it only happens every so often (without any type of consistency).

I think that's all I can think about for now..

Tags (1)
Reply
0 Kudos
252 Replies
Kiemo
Contributor
Contributor

Just to update my concerns above. The reclaim process is working and removing backups each day that match my removal criteria.

If its still working this well in early Jan 2010 I'll be a happy man!

Reply
0 Kudos
adisarro
Contributor
Contributor

I've been using 1.1 since it came out and i'm very impressed by this update.

I would now trust this in a production environment.

Reply
0 Kudos
chrisaug
Contributor
Contributor

I've been testing it since it's 1.0.x release and while I can say the new version is immensely better, I'm still having strange issues. I find I need to restart the device every few days, and today I was backing up 4 test servers and I noticed that the VDR VM just rebooted itself in the middle of a backup. The report said "Task terminated unexpectedly, possibly due to a power failure or system crash".

Reply
0 Kudos
Stevester
Contributor
Contributor

Well the 1.1 verison is what is out now and from all reports it is stable, however i still feel though that VDR in general still lacks the nice robust features that its counterparts have such as email notification, an ability to perform FULL/Incremental backups on a cycle basis. The VDR 1.1 is a step in right direction as must "crawl before you can walk". Until that happens, I think I will stick with GhettoVCB.

Thanks

Steve

Reply
0 Kudos
adisarro
Contributor
Contributor

I am using VDR just for daily backups. Just as something extra to have and certainly easier for file level restore.

For the main backups i'm using vcb weekly/monthly to tape.

Reply
0 Kudos
alexlie
Contributor
Contributor

Had to reboot my VDR 1.1 yesterday 'cause of one stuck integrity check

process (got stuck already a week ago without getting noticed - there

needs to be some work done considering monitoring and notification in case

of errors).

But it's now working since the official release date without any strange

issues.

Still I am not relying on it as my only backup solution but I hope for are

stable system... finally.

Reply
0 Kudos
Stevester
Contributor
Contributor

While VDR 1.1 is clearly more stable than the previous verisons, I am still not convinced it's ready for primetime/production time. VDR 1.1 is a step in the right direction, but its still not as robust as Veeam Backup or any of the other 3rd party VM backup products. For example, can you peform Full -Incremental backup in a cycle against the current way rentention policies can be implemented? There is no email notification system..etc. I think I will stick with GhettoVCB until VDR's true time is at hand. Hopefully in the 2.0 verison we will truly have a reliable product. However you have to "crawl before you can walk" so this is to be expected. Those are just my opinions....LOL!!!!!

Thanks :smileyblush:

Stevester

Reply
0 Kudos
jjewett
Enthusiast
Enthusiast

For all you guys saying "vDR is not ready for primetime", you are all a bunch of chickens. Smiley Happy

This early adopter has been using vDR since v1.0 and has upgrading incrementally thru version 1.1. I have nothing but good things to say about the VMware Data Recovery appliance.

I realize that many have experienced some serious problems and had difficulty implementing vDR. Of course, I have had some issues getting vDR and vCenter to connect and the occasional snapshot failure. However, after reading the admin guide and proven practices document I was able to get a reliable deployment. It does work as advertised and is very good.

Our main environment has 24 ESX servers and 280 VMs. But we are using vDR at a campus location with 2 ESX servers that are being managed by the same vCenter. This location has 12 VMs being backed up onto a NexentaStor Dev Edition DIY NAS. We have 1.5TB of storage presented as an NFS share. The share is attached as a datastore for the vDR appliance to use as a dedup store. Performance is very good. We typically see inital backups of 20-40GB run at 2.5GB/min which take about a half hour and restores are faster at about 3.5GB/min. The version 1.1 file-level restore .exe is a now GUI and works splendidly.

I think VMware has done a great job in keeping this appliance simple to implement, efficient with resources and effective as a backup. It still has a lot of room for improvement and isn't an enterprise-class backup solution by any means. But for the small-to-midsize environment it is a very appealing, cost-effective (free!) option.

Jonathan

Reply
0 Kudos
Stevester
Contributor
Contributor

Ouch JJewett!!!!!!!! That hurts.....lol!!!!! Well to each its own but in my opinion, VDR still hasn't truly passed the test of time. It's still crawling learning how to walk. It's missing functions such as email notification, performance monitoring, etc. However this is normal in software development. VDR is getting there but i am still of the opinion its still not ready for primetime.

Thanks

Stevester

Reply
0 Kudos
KBuchanan
Enthusiast
Enthusiast

Stevester: You are absolutely correct. VDR is not ready for production use...beware any that uses it for production!!

As for the sw dev lifecycle, you are missing something. There should have been a clear understanding of how the product should work (to meet the end user needs), and clearly - VDF misses the mark! Secondly, the product obviously wasn't tested adequately as nearly EVERYONE can attest - it was, and in my opinion, still is, a flawed product. Granted, this release seems to be more stable, but I have had it to freeze/lock-up one time.

Kevin

Reply
0 Kudos
abbasi
Enthusiast
Enthusiast

Still not ready for Production!!!!!

VDR 1.1 has the some of the same issue as previous versions. It created snapshots of VM's then for for some reason did not go back and delete the snapshots. So the snapshots keep growing. It did this for 5 machines.

Why is there no a simple check in this program that if it has created a snapshot to delete the snapshot after the backup is done or the backup has failed or delete it after X amount of time. This seems like an obvious component. The danger of snapshots growing out of control is pretty serious.

I had to shut VDR down. Go to each machine and manually delete the snapshots, then remove the drives from the VDR machine

I cannot use this. I will wait for ver 2.0

Reply
0 Kudos
XavierE
Enthusiast
Enthusiast

Everyone should be using VDR in a test environment unless someone wants to play with the production env.

Indeed VDR is not ready yet but it's getting closer. First of all it should be 99.99 % clean from bugs and failures, without mentioning lack of features like notifications, better integration with vCenter, and more.

I bet VMware is working on this, but they are WAY behind their schedule. In the mean time they already got the money in the pocket and their customers are not being paid/reimbursed for the time spent debugging and testing.

Reply
0 Kudos
einstein-a-go-g
Hot Shot
Hot Shot

vDR is working fine for me here.

The Snapshot issue, I don't believe is a vDR issue, vDR only calls the snapshot function. This could be a vSphere 4.0 issue.

What error messages are being seen in the vDR logs, and have you reported this to VMware and raised an SR.

e

Reply
0 Kudos
einstein-a-go-g
Hot Shot
Hot Shot

I disagee.

When I first used pHd Espress and Vizioncore ESX Ranger product, when they first came to market they had bugs, and didn't have notifications.

e

Reply
0 Kudos
KBuchanan
Enthusiast
Enthusiast

VMWARE IS NOT LISTENING...they need to recall v1.x and notify their customers the product has serious flaws that could result in the inability to recover a VM.

What's to say that "v2.0" would be any better? So far - VMware is not giving me the confidence that they have a clear understanding of the problems (call tech support a few times - you'll see what I mean). I have consistently been told that these problems are isolated to a few clients...THAT IS BOGUS. They are disillusioned if they really believe that. Also - they overall functionality and features of VDR are seriously lacking. Heck - even my "homegrown script" can send me a status e-mail!!

They aren't listening to customers and they are losing credibility to get this product back on track. I have personally spoken with the VDR product manager (a few 3-4 months ago) and I do believe they are committed to making VDR work; however, they shouldn't have released it as a "working product" when it is characterized (by the community) as unreliable.

I insist...if someone from VMware is reading this article...SAY SO! Tell us (YOUR customers) when we can expect a working and reliable release? Stop hiding behind the lame of excuse: "We can't comment on future release/functionality BS" storyline. Give us useful and meaningful information. We ARE your customers - give us some respect!

Kevin

Reply
0 Kudos
KBuchanan
Enthusiast
Enthusiast

Amen.

Kevin

Reply
0 Kudos
tlyczko
Enthusiast
Enthusiast

I have used esXpress since its early days in 2007 and it has always had

email notifications, quite detailed ones.

Reply
0 Kudos
KBuchanan
Enthusiast
Enthusiast

My belief, (and intent I was trying to convey), is that any/all software companies should take a responsibility of their software and "do what is right". Let their customers know there are known problems with their product.

Ok - consider this: What if you had to have a implantable pacemaker installed...but the manufacture of the pacer knows that there are problems with it - the pacer has stopped working or didn't work properly all the time. Would you still want to just sit back and ignorantly believe that "all is well"?

My point is this: They need to take responsibility and admit it shouldn't be used in production. There are people out here using it and ignorantly relying on VDR to backup their VMs. There have been enough problems (just in this thread) raised about the reliability of VDR. The problem is that probably not everyone is reading the forum threads (I don't read ALL of them)...but VMware could notify all of their customers. It is their choice. Take the high road, or hide behind the legalese that probably warns you they can't be held responsible for damages caused by their product.

I would give them more respect if they could be honest and transparent. Instead - we just sit back, test it, and wait for another release. I have ABSOLUTELY no confidence about what to expect and when to expect it. In short, that translates into me losing confidence in VMware as whole.

Kevin

Reply
0 Kudos
jjewett
Enthusiast
Enthusiast

Wow. A pacemaker comparison, really Kevin? Not that backups aren't serious business, but I think that's a little too much hyperbole.

VMware vDR 1.1 can be a a viable PRODUCTION option for a small deployment. I've never advocated it for mission critical or large environments (25+ guests). It could use some refining, but it is still probably the least painful backup product I have ever used. PHD Virtual's esXpress, Symantec's NetBackup, vcbGhetto script etc., all require a decent amount of configuration and playing with. vDR is the plug and play of backup appliances. OVF deploy, power up and attach disk. Then get image-level and file-level backups in one shot. Plus all the magic of deduping speed and space savings!

VMotion, SVMotion, HA/DRS, were all revolutionary features that were a little scary to use at first, but the bugs were worked out and now most of us can't imagine doing our jobs without them. This product is barely 6 months old. Give vDR a little more time and we might feel the same way about this little gem. As for your complaints about VMware support, well that's a whole 'nother discussion.

Jonathan

Reply
0 Kudos
Stevester
Contributor
Contributor

Good Day Everyone,

The VDR 1.1 is NOT reliable. I think that's Kevin's point. More importantly, the product was advertised as reliable and clearly it isn't. I agree that responsibility should be taken by VMware in the form of a refund of money or some other form of compensation. However you have to "crawl before you can walk, so time is needed to polish the product as nothing is perfect.

Stevester

Reply
0 Kudos