Solaris VM supported?
We are still evaluating the product. I have attempted to make a backup of a Solaris VM (containing NexentaStor) today. Below are the error messages we are seeing.
Basically it says the OS is not recognized.
Please advise.
Betriebssystem konnte nicht ermittelt werden.
Zusätzliche Info:
--------------------
Fehlercode: 8
Module: 57
Zeileninfo: f5e7a82671985292
Felder:
Nachricht: Betriebssystem konnte nicht ermittelt werden.
--------------------
Fehlercode: 7
Module: 57
Zeileninfo: 5186ab5c3563a7c3
Felder:
Nachricht: GRUB-Loader konnte nicht ermittelt werden.
--------------------
Fehlercode: 7
Module: 57
Zeileninfo: 85de7f4c6a577814
Felder:
Nachricht: Verarbeitung von GRUB auf Laufwerk '\comp_emu(vm://01102AF8-CDCC-4573-B6B4-2199D336B3C8/42110672-d29e-de22-bb70-de9dca449882?host=host-436&type=vmwesx)\hd_emu(1)' fehlgeschlagen.
--------------------
Fehlercode: 0
Module: 96
Zeileninfo: 5186ab5c3563a8c8
Felder:
Nachricht: Mounten von Volume '\comp_emu(vm://01102AF8-CDCC-4573-B6B4-2199D336B3C8/42110672-d29e-de22-bb70-de9dca449882?host=host-436&type=vmwesx)\hd_emu(1)\part_type(BF)num(1)' fehlgeschlagen.
--------------------
Hello dev-anon,
it is reported as a warning, however i dont think anything is being backed up. The entire VM is 1TB, and the backup finishes after 30 secs or so.
The archive on the file server share is about 2GB.
I do understand that file-by-file restore is not supported, and that is fine.
Why is a sector-by-sector backup not being performed?
Hello dev-anon, can you help us to resolve this issue?
We have updated vmprotect to the latest version today. Things have improved a bit, but it is still not working properly.
The Solaris VM consists of 1 small boot disk and 4 data disks. From what i can see, it appears only the first (boot) disk is being backed up, the backup of the remaining disks is failing. With the previous version, none of the disks were being backup up.
I am attaching the log.
Can someone please comment?
regards
can you help us to resolve this issue?
No, because I know nothing more about VMprotect than listed in the userguide. But if you still are on the (15 days iirc?) trial, you may open a support ticket. What do you get if you try to recover this backup - a VM with the only boot disk without other four?
Same issue with debian - this time we have three disks assigned to the VM, first one is the disk the system boots from, the other two disks are aggregated as a Volume using LVM.
The boot disk gets backed up, the other two disks get skipped with a warning.
Since the situation on the Solaris VM is very similar (the 4 disks which are not being backed up are combined to a large Volume) my conclusion is that Volumes are not being supported, or this is a bug.
Let's see if we can get support ticket opened for that.
hmm
for lvm it seems for restore it is not supported would this purhaps mean lvm is totally not support?
I meant, not supported for file recover. It should be able to restore them all as-is... if it could back up them.
Btw, @zel - how's your support ticket?
well we have opened a support ticket more then one week ago for the problem described above.
The way this is being handled by Acronis support staff is a bit frustrating, to say the least.
First, we really had to convince the woman on the support chat to escalate the issue. I dont think she understood the problem at all, and was trying to convince us that we are not using the product properly.
We persisted, and submitted the logs and everything she asked for.
now more then a week later she replied essentially saying this was by design (or to that effect) and that only sector-by-sector backup is supported.
Since i do not see an option to force or select sector-by-sector backup i dont see how her reply is helping in any way.
So we keep running around in circles here.
My impression is that she did not even look at the submitted information at all.
The way this support case is being handled is making me really question if Acronis is ready to support the intended audience for such a product professionally enough.
From other comments on this forum i can tell we are not the only ones seeing that issue, so one would think that Acronis should have an natural interest obtaining more detailed information about such problems. After all, the configuration we use with Debian and LVM isn't really that excotic.
Hello Zel,
Thank you for bringing this to our attention.
We're sincerely sorry for the inconvenience you've faced with support. We would like to review the situation, and help you with the problem, but unfortunately I cannot find any cases tied to your account. I assume you've used another e-mail address. Could you please let me know the case number?
Thank you in advance for cooperation.
Btw, if you didn't collect http://kb.acronis.com/content/1637 from inside the Debian VM and just any more or less comprehensive report about partitions from the Solaris one - it was a hint.
Hi all,
zel, thank you for contacting me via PM. dev-anon, thanks a lot for your help.
I decided I update this thread with the most recent information.
Zel kindly provided me with the case number, and we've managed to check it with the vmProtect development team. For now there's no information on the cause of the problem, and the issue requires reproducing in our lab. So in case you've faced this problem, please provide support with the following information:
1) Logs from vmProtect
2) ESX/ESXi Virtual Appliance Logs
3) Logs of the ESX host: connect via the vSphere Client Infrastructure, then go to Administration-> Export System Logs
4) Screenshots of all options of the backup plan
5) Screenshots of the configuration of all disks on ESX for these virtual machines
6) Please update vmware tools (and let us know the version you're using)
7) What is the disks configuration in Solaris?
This information will help us to reproduce the problem, and find a solution for it.
Thank you in advance for cooperation.
Thanks for your reply, Yana.
As mentioned previously it took more then a week for the support to respond to our initial request, and our test license has expired by now.
I will try to obtain a new test license and send you the logs as soon as i find a minute.
kind regards
Hello Zel,
Thank you for the update.
In order to extend the trial period, please contact the support in the case with the request, they will contact the Sales department for the new license. I will also send a message to them in a minute, so there should be no delay with the response from their side.
If there is, please, let me know, we'll speed up the process.
Thank you.
i also have an linux vm with an extended LVM volume with a second disk and this this is also skipped.
Yana,
i have been trying really hard to obtain the logs you are requesting but i am having not much joy.
The link you have indicated here:
2) ESX/ESXi Virtual Appliance Logs
is actually for another product. Once again, we are using vmprotect.
So i have been trying to turn debugging on according to this KB article: http://kb.acronis.com/content/30926
I managed to get into the console, and even to open the file with vi. But then i got totally stuck because of the unusual keyboard mapping which is active within the appliance. I even looked at a standard russian keyboard mapping trying to find where the colon is located (this character --> :) but i was not successful. Without a colon, it is impossible to quit vi.
Since i was stuck in vi i decided to reboot the appliance. I managed to get back to the graphical screen (the one which shows the ip address etc) and eventually restarted the appliance.
Now i have the effect that the logging does not seem to work at all anymore. In the web management interface when i go to the logs section all it show is "loading" but never finishes.
I think i am stuck and i dont know what to do. I am thinking about reinstalling the appliance but i am afraid of losing all configuration settings.
What i have done is that i have made some screen shots which i am attaching.
First one is a screen shot of the spinning wheel when trying to display the logs.
The remaining screen shots are the settings of the Solaris VM we were trying to backup.
The Solaris VM consists of 5 virtual disks. First disk is a small boot disk. The remaining 4 disks have been aggregated in a ZFS zpool.
The first disk gets backed up from what i can see, the 4 disks which are in a pool simply get skipped.
Here are some suggestions:
- make the debug mode easier accessible, without having to edit a system file with vi
- make it easier to reboot / shutdown the appliance, since the appliance has no vmware tools installed it is very hard to navigate the screen with a mouse as it jumps around like crazy!
Further, i do not understand why it is so hard to recreate the problem in your labs - just install a Debian VM, add two virtual disks and bind them together as one volume using LVM - shouldnt really be a big problem....
Please let me know your suggestions and if there is anything else i can help with.
Kind regards
Hello Jan, and Zel,
Thank you for your comments.
Jan, LVM volumes are supported, and backed up in sector-by-sector mode. The only disks that might get skipped are independent persistent disks (those that VmWare cannot snapshot). So this might be the case of the problem.
If it's not, please send us the log, and disks screenshot as Zel did.
Zel, I've forwarded your comment, and the information to the Development team. Once they get back to me, I'll update the thread.
Thank you for cooperation, patience, and understanding.
Hello Zel,
I've received some news from the Development team.
They've managed to localize the issue, and are working on implementing the fix into the upcoming hotfix release. Meanwhile they suggest to use VmProtect 6 appliance as a temporary workaround.
To whom it may concern: if you've faced the same issue, please contact any Forum Moderator for the fix via Private message (please don't forget to specify the link to this thread in the subject or a body of the message).
If the problem remains, please send us the diagnostic information.
Thank you.
Hello Yana,
we have the same Problem with a Solaris 10 ZFS Drive. The System Drive is in the Backup the second one with the Oracle DB (ZFS) on it isn´t in the backup.
Is there already an fix.
In March of 2012, it was discovered that vmProtect 7 does not work in several situations, with the error "The Disk is Skipped", as described in this thread and others. I encountered the problem with the Windows agent(mandatory for my environment) backing up Linux VMs. In April 2012 we were assured that the problem was "important" and that Acronis would resolve it soon.
It is now June 2012 and there is still no fix. Update 2 is supposedly coming out "real soon" but, the list of fixes in Update 2 does not currently mention this problem. Therefore, it seems unlikely that the fix will be included.
Is this problem ever going to be fixed? Will Acronis vmProtect 7 ever be able to backup Linux VMs with LVM disks? Should I just forget about Acronis vmProtect and buy the competitors product? I can't/won't wait another three months!
Any chance that Solaris VM could be supported already? 