While using incremental forever, every so often Unitrends backup is supposed to roll up your incrementals into a new master on the server side. Awesome, right? Well we get failures for this process on a fairly regular basis (about weekly). I would usually chalk this up to being relatively low on storage, (which was another extremely frustrating issue) but I’m still having trouble after reorganizing and eliminating well over a terabyte of data. My most recent failure was of a client utilizing a whole 40GB of the storage.
Under settings and storage I’m told I have 1.5TB of storage remaining (which is lame seeing as how that’s about what it told me before eliminating the previously mentioned terabyte) and zero retention set. We’re using a Recovery-813, and I have a separate network with a Recovery-712 with about 1.7TB remaining where this happens occasionally as well.
Is anyone else having regular failures, or did everyone just spend substantially more money on extra storage they can’t use? Or am I just doing something terribly wrong?
Support kinda sorta helps with this, but I usually get some sort of political answer that answers me without answering me.
7 Spice ups
Jake - I’m sorry that you’re having these issues. It’s a bit hard to understand what may be causing this issue, and even though I know that you’ve spoken with support about this in the past, I’d like to escalate your case within the support team to see if we can wholly resolve this issue for you. Can you private message me your contact information, including asset tag(s) so that I can look into this for you?
Will do. I wasn’t contacting support at this point because it’s such a vague issue. Just wanted to see if this is ‘normal’ behavior for anyone else.
I ran into issues with my 813 for the first year or year and a half of owning the unit. All of the drives have been replaced, and I had to re-image the unit several times due to a corrupt database from a drive failure or when the rollups failed and caused corruption. This all seemed to go away once I stopped upgrading every time there was new version available and when I started to just back up VM’s. When I had all of my problems I was backing up only the data on most of my servers because they were on a Hyper-V 2008 host. When I bought the unit sales did not tell me that Hyper-V 2008 was not supported. Now I am on Hyper-V 2012r2 with version: 7.5.0-5.2014082214 on the Unitrends device and all has been working great. Within the first six months of using the 813 I was ready to throw it off the building because of all of the problems. Since I have resolved those issues, it has worked as advertised and I would probably consider purchasing and upgrade when I outgrow the 813. My case may not help you but I would suggest that you keep on support until you have a resolution and involve your sales rep if you are not getting proper attention to your problem.
3 Spice ups
Holy cow, and I thought I was getting frustrated. I haven’t really noticed the updates until after they’ve been out for a while (or support suggests that I update).
Generally I like the product, but these little gotchas get annoying when it seemed like the Spiceworks community was touting it as the greatest thing ever when we were making our purchasing decision.
Jake - Thanks for sending me your contact information via private message. I know that our support team will do their best to get everything sorted out. If there’s ever any feedback that you want to share directly with our developers, or with our management team, just go in your interface to About > Feedback.
Interconnect6907 - I’m so glad to hear that things have improved for you. If there’s ever anything I can help with, please let me know.
1 Spice up
jrondo4
(jrondo4)
7
Hey Jake, it’s Jake…
We had the exact same issue on each of our three Recovery-813 series appliances after we upgraded to version 8. I had to fight with support about this issue for quite some time, but we did get it sorted out. Is your appliance fully up to date? If not, you’ll have to get it updated and you’ll probably need to schedule new master/full backups of all your protected clients.
2 Spice ups
Hey Jake.
My appliance is up to date, maybe I just need to run new masters manually on everything that I haven’t had to do that on to clear this up. I hate having to do that. Oh well haha.
jrondo4
(jrondo4)
9
Are all of your agents fully updated as well, or did you just update the appliance/data protection unit (DPU)?
1 Spice up
Every single one is at 8.0.0-6
This is a great question as this is a best practice for all software updates.
jrondo4
(jrondo4)
12
Well, you’re further along than I am. I think I only have one agent that’s fully up to date, and all the rest are still 8.0.0. That said, the synthesis issue in our case was traced to the DPU and the agent version (8.x) didn’t seem to matter. Once the DPU was fully up to date and new full/master backups were run, the issue pretty much went away.
2 Spice ups
Going to re-run my masters one by one and see how it goes.
Lololol. Support totally e-mailed an address 100% different than I provided. They’ll get it right some day. I guess they never updated their database like they told me last time I carefully explained where to reach me.
I guess I’m a bit confusing being listed for two companies with some overlapping info though 
jrondo4
(jrondo4)
14
As I think on it some more, support did a few other things in our environment…
They tuned the hashing and deduplication…
They ran smartvac on each of our databases…
There was an autosynth fix that was needed as well, but that was when we were on version 8.0.0-4. I’m pretty it was rolled into the 8.0.0-6 update…
There was also a separate “rubber banding” fix they tried, but it actually made the issue worse so they rolled it back…
All of that is to say there were a lot of different factors the support team wanted to rule out as potentially contributing to the synth issue. Honestly, the best bet is to push this with support and stay with it until it gets resolved. Be as patient as you can be, because it could take some time to sort out fully. I opened our case in the middle of December and we were still working at it until well into March.
2 Spice ups
jrondo4
(jrondo4)
15
One last thing to note - I didn’t really make a lot of progress on fixing the synth issue until my case was escalated to L3, and the escalation didn’t happen until I reached out to the support team manager directly.
I think one time they ran some patch like that on my 712 and called it a day. The lame thing about this is that is somewhat transient. So I don’t like to spend all day fine tuning something that we feel should, for the most part, “just work”.
Jake - We completely agree. I know that the team that’s working with you from support now will be able to get your appliance to it’s promised set & forget management.
Jake,
I have an 813 and have the same issues. They just showed up for me. I actually put a stop on a new 3 year contract with unitrends until I get this sorted out. I had 40% of my space available but with bad synthesis backups my incremental forever plan was only keeping a few days of backups…I was told by support that there was a problem with synthesis backups in the early versions of of 8.x code. Latest version it is suppose to be fixed.
2 Spice ups
Justin - I’m sorry to hear that you’re also having this problem. Do you mind sharing your latest support case number so that I can look into this for you?
I had the same problem with my 713. 8.0.0-6 fixed it.
2 Spice ups
