A week ago I posted a list of Veeam Backup & Replication 7 bugs that I found critical: http://t.co/PRFzEky7Vq (do read my post to become familiar with my point of view).
Regarding critical bugs: in our definition, any bug that results in unrecoverable backup produced, or production data corruption. Only one such bug is reported on v7 so far (in 1 month), and it has a very small scope. Nevertheless, the hotfix is already available. We document critical bugs in our support KB, and there is a dedicated Veeam forum sticky too.
Any other bugs, we don’t normally differentiate between them. ALL issues our customers come to support are hotfixed (unless there is an easy workaround, or extremely minor). The hotfix is provided to a customer, and included in the patch release that we do once every few months (kind of service pack). At this time, we have about 60-70 change sets included in the upcoming patch, however not all of those are bug fixes – part of those are enhancements in the new functionality based on the first feedback.
I’ve been with Veeam since version 1, and v7 has been the most solid release we’ve ever had (especially considering the amount of new functionality). I guess I can say we disagree on the definition of “total disaster” in the blog post title to me, this was the most successful major release in history of Veeam.
Regarding specific items you listed:
1. The issue is not specific to v7. There is no case ID, so I cannot be sure – but this might be one of the issues we are working with Microsoft on right now.
2. Not a product issue. There are no complaints in this thread, rather proper deployment considerations dialog. Faster WAN requires faster hardware involved in WAN acceleration. I do disagree with “useless gizmo” assessment on the feature that saves almost 100x WAN bandwidth, as per numbers OP himself has posted in this topic.
3. Not a product issue. Verification speed is determined by backup storage speed. Performance reported is not poor, but rather expected for a lower end NAS on 1Gb LAN, especially if there are other workloads hitting the same NAS. Faster backup storage would result in faster verification.
4. Not a product issue. The problem looks to be specific to storage used (RDX), no similar reports from users using other backup storage. Moreover, we are using the same functionality internally.
5. Not a product issue. Why: we’ve run a poll among our users before making this decision, the results speak for themselves http://forums.veeam.com/viewtopic.php?f=2&t=13400. Note that the poll was run over 1 year ago, so you can adjust results even further towards more latest Windows versions.
Hope this answers your questions and concerns at least partly.
Now let’s go through his answer together while I comment on some of his (and Veeam’s) assumptions and statements concerning all their clients:
‘Regarding critical bugs: in our definition, any bug that results in unrecoverable backup produced, or production data corruption.’ – I am sorry, but this is total BS and it means that they relabeled critical bugs into non-critical for purposes of PR, I guess? Pardon me, but I, as a client, think that if my backup doesn’t work, I am exposed to risk and this is critical for me, even if Veeam does not consider it risky.
‘to me, this was the most successful major release in history of Veeam.’ – This is not for Veeam to decide, but for its clients. When, in the last screen of Backup Wizard window, I see only the ‘Finish’ button enabled, and ‘Back’ and ‘Cancel’ are grayed out, I wouldn’t call it ‘successful release’, that’s ‘buggy release, but that’s just me. Or, rather, it’s not just me: people on forums are saying ‘Clearly Veeam 7 has issues’ (proof: http://communities.quest.com/message/89849#89849 ). Overall view after numerous discussions on forums and among my colleagues is that it is not ‘most successful’, maybe it is the case for enterprise customers, but I am a simple dude with a handful SMB clients (a typical IT consultant) and I am worse off =(((
‘I do disagree with “useless gizmo” assessment on the feature that saves almost 100x WAN bandwidth,’ – If I do compressed incrementals of several different servers, what WAN acceleration I will get? Probability of hits in cache is almost zero and acceleration will be close to zero. Doing full backups of the same servers to get more hits and nice acceleration numbers is what I call ‘useleless gizmo’. I would like to warn everyone: in your project planning, do not count on achieving any WAN acceleration. If you count on it and will not get it because the data was too dissimilar, you will be in the world of trouble. I would say that this new feature is purely theoretical and probably added for marketing purposes, because 100X you would only get in the lab, and in the real world you gain nothing from it.
‘Not a product issue. The problem looks to be specific to storage used (RDX), no similar reports from users using other backup storage. Moreover, we are using the same functionality internally.” – Awful response that means that Veeam is not going to help me. They never mentioned anywhere that RDX is not supported, while it should be clear that RDX is NOT supported. RDX is quite popular, my clients use it and I need it to work, whose problem it is I don’t care.
‘Why: we’ve run a poll among our users before making this decision’ – 109 users of Veeam forum decided on behalf of all clients?! (poll: http://forums.veeam.com/viewtopic.php?f=2&t=13400) That is a very small number compared to the total client base of Veeam. Adding tape support needs OS drivers for it. Microsoft removed tape support in Windows 2008 R2 NTbackup and there are way less tape drivers for 2008 R2 then for 2003. So all the potential clients with tapes 5-15 years old will not be happy because their tapes work only on 2003 and these drives are not cheap – some cost up to 6-number figures in USD. Maybe someone from forums is ok with this, but I am not and there are many customers like me. I was misled. I spent lots of time and even if I get refund, who will pay me for the time I spent on this?
PS BTW, I’ve never gotten an answer from Veeam on my original question: just how many bugs were there in VBR7 so far? It sure seems like at least couple of hundred by the volume of bug reports on their own forums