+1, for my new machine. Haven't set that up yet though. Will use the 'idiot' backup method, initially anyway. Guess we'll soon see @Arvy on the Terabyte thread also .
Already there. The real question now is how long before the total strategy shift away from Reflect as the primary backup/recovery tool is complete. Not very long I think based on results so far. Froggy's recent findings (added to Macrium's history of patching sloppiness) have provided considerable incentive to this user.
I will continue using MR as I am familiar with it, but start using Terabyte (simple mode) also, and maybe Drive Snapshot ... I currently run ATI, and AOMEI Backupper, in addition to MR.
Price wise this Image for Windows is pretty interesting, as a home user is apparently allowed to use the program on 5 PC’s with a single license. Does it have something similar to Reflect’s Intelligent Sector Copy cloning? (Cloning, not imaging)
@XIII - the latest versions do, you just have to get used to syntax (it can be "strange" <sorry Brian >). It also has a CBT-like feature if you do a lot of images while in a single Windows session.
I thought MR had a lot of "template" options. IFW has 'em beat by a long shot, right down to the finest operational detail. But it's not really so bad once you get "terminologically acclimatized", and a lot of them are one time only "set-and-forget" items with reasonable defaults that should work for almost anyone except some "special needs" cases as Brian said above. Don't like the PE/RE builder quite as much as Macrium's, but don't really need it anyhow for my WinBuilder setups. At least the IFW WinPE launch doesn't impose its own "drive letter" reassignments like the Reflect launch does with no "off switch", and the IFL ISO build also boots and works fine using Grub. So far, so good. I'm favorably impressed. __ P.S.: I see that Mr Sills is being very helpful to you -- NOT! Now he want to know if you've "ruled out Anti-Virus software interfering" and wants you to try a backup with no compression. They just don't seem to get the point. Why would resizing an ext partition cause Reflect to report such a drastic change in its used/free space numbers compared with all other tools? You're not talking about a few bytes here. And does Reflect use "intelligent copy" properly for ext partitions as advertised or does it not? Saying that "Linux partitions are copied in exactly the same way as any other partition on your system" just ducks that question. Unfortunately that's not unusual recently.
Hi Arvy Yes, we totally 'get it', apologies if you perceive my answers to be unhelpful. Macrium Reflect uses the ext Super Block to report used and free space for the partition. When backing up an ext partition each block group bitmap is read and only blocks in use are included in the image. It's accurate and uses the ext meta data to ensure that only used blocks are imaged. We have also tried to replicate this issue without success. Imaged blocks are read and written sequentially, so the only way a 'slow down' can occur mid image is if something is consuming more clock cycles during the image process or if there's interference with either the read or the write. The suggestion to create an uncompressed image has two benefits,. 1. It eliminates excess CPU cycles if compressing uncompressible data. It would be interesting to see if this has an effect on the imaging speed. 2. Will show the actual size of the data blocks being backed up. This will indicate whether the super block is in inline with the ext block group bitmaps. If not, then running fsck on the file system will repair the super block header and enable Reflect to show the correct reported used size. We set up a similar system to @TheRollbackFrog over the weekend, dual boot Linux MINT / Win 10 and expanded the /home partition from 50 GB to 100 GB with approx 30 GB of data using GParted. This was repeated several times and was always successful. We noticed an anomaly with the way some Linux utilities report incorrect usage and documented it here: https://knowledgebase.macrium.com/d... file systems used and total space in Reflect However, this doesn't explain why @TheRollbackFrog is seeing a difference in the GParted UI. It's just a for info. Cheers Nick - Macrium
@Macrium -- That's all very interesting, Mr Sills, but the key question is whether Macrium Reflect uses "intelligent copy" properly for ext partitions as advertised or not. You and/or your team are aware, I suppose, of known competitor issues with incorrect detection of "block bitmap corruption" resulting in forced sector-by-sector mode backups of ext partitions. (Example below.) Can Macrium state categorically that Reflect's ext "Super Block" usage is entirely free of any such errors with consequential forced mode operational impacts on ext partition backups. By way of example, this extract is from my own ATI logs for a mixed NTFS/EXT4 SSD (clean fsck.ext4 results) backup operation omitting the lengthy "SnapAPI" lead-up:
Nick- Although your product is indeed excellent, it is disappointing that protection against backups being encrypted is not found in the Free version. I understand that Paid versions of products should be superior to the Free, but this lack of protection is rather unseemly. Meghan
Hi @Arvy 'Intelligent copy' uses the block group bitmaps to determine the blocks in use at the time of the image. For incrementals, each block is then hashed and compared with the same block in the previous image and copied if different. This is the same process that's used for all file systems. Unlike NTFS and FAT32 Reflect doesn't run 'fsck' (or equivalent) on ext file systems prior to imaging. This means that corrupt ext file systems will be backed up without warning, however, the backup will accurately contain the same files, data and usage that is shown when the same file system is mounted in Linux. We are keen to help @TheRollbackFrog with his problem and also to try and understand the cause. Except for the block and group sizes, the super block data is merely used to report the used space and size for the partition, it's informational. It isn't, and can't be used, for the actual backup as it only contains summary information. BTW Arvy my name is Nick. I'd appreciate it if you would refer to me by that name. Cheers Nick - Macrium
I don't see an issue with that. Users should just be glad there is a free version to use. If they want it protected like the paid version, they can buy the paid version.
That's not what I asked. I know what "intelligent" sector copy is, or at least what the documentation says it is. The very straightforward KB description says simply that it copies "only file system sectors/clusters that are in use." But if, as you say, Reflect doesn't check whether or not ext file systems are corrupt, how can it possibly determine whether that strategy is being applied appropriately? Personally, in those circumstances, I'd prefer some kind of pre-check even if it does force sector-by-sector mode when corruption is detected. My references here to the CEO and Founder of Macrium Software have been and shall remain polite but distant as befits our current impersonal relationship. You may respond or not as you see fit. Your appreciation doesn't enter into it.
Hi Arvy I'm sorry I must have missed your politeness and recognition of help. @TheRollbackFrog Please can you continue with this conversation on the Macrium forum. Kind regards Nick -Macrium
@Macrium As a tech-savvy while still not professional Linux user (meaning I'm not working in IT related field, but still in a different scientific area), myself, Arvy, TheRollbackFrog and possibly others are all observing the same discrepancy of the amount of in-use space on EXT4 partitions between MR and many other disk imaging and Linux OS distros. I'm pretty sure we have different hardware setups. I'm pretty sure logically which side has a higher chance of faulty used-sector-recognition issue in this MR vs all other EXT4 capable tools/OS scenario. It's so obvious that I don't even need to go into all these tech details before I (and any ppl with normal reasoning capacities) can clearly see that MR has a problem properly recognizing used sectors in EXT4 partitions. It's just simple logic. Users do not care what method/mechanism a disk imaging software uses to get the job done. They only care about getting the job done properly without hassle. All these friendly critics in this thread are trying to help MR understand the problem and hoping that MR will solve this problem, to make MR great again. From your attitude I sense a very familiar feeling when I dealt with Acronis customer support with their faulty support on EXT4 under UEFI/Secure boot - very defensive about obvious bugs in their product. People here still care about MR because they don't want to see a once great MR product goes belly up like the once promising AX64 Time Machine. When no one cares reporting a bug to you, you know what it means. Consider MR lucky you still get these feedback from loyal fans like Arvy and TheRollbackFrog.
And during your absence Reflect's "intelligent" sector copy has decided that you should actually get a 89.3 GB uncompressed image of an ext partition that Linux (and everything else) says has 39.1 GB used. That quite an interesting algorithm, but I think they'd prefer your not discussing it here any further -- at least not with me.
... and I won't For anyone actually interested in following the technical aspects of the issue. the thread may be seen HERE... there's no reason to duplicate everything here in the Wilders Forum. I will comment if I feel there is anything to pontificate...
Somehow I'm doubtful that open discussion will proceed much further there either ... but I've been wrong once or twice before.
Help ! I know you have to be careful with installing new software ... but ... System: Windows 7 64 bit pro desktop, external harddrive NTFS system. I used an older version of Macrium Reflect Free , version 5.x Some time ago I had to create a restore DVD that would allow for starting the computer with the current hardware/external harddrive. For this version 7 was required. Or possibly 6. Since I was updating a number of programs on this computer, I decided to install the latest version of Macrium Reflect free, probably over the the old version. After that, I created an image using Macrium Reclect. But to my horror, it seems to have deleted a number of older images that I wanted to keep ! I had not specifically selected any option to remove older files, the free space was far more than 5 GB. When I start Macrium Reflect and check the log, I see Free space threshold: Delete oldest backup sets when free space is less than 5.00 GB .... Retention Rules Rules will be applied to all matching backup sets in the destination folder Full: Retain 12 full images Linked incremental and differential images will also be deleted Backup Sets: 28 sets found delete file XXX Retention Rules Rules will be applied to all matching backup sets in the destination folder Full: Retain 12 full images Linked incremental and differential images will also be deleted Backup Sets: 28 sets found Differential: Retain 4 differential images Linked incremental images will also be deleted Differential Backups: 0 found Nothing to delete ENDLOG Apparently, 16 files have been deleted. Including a very old and clean image from 2011. What happened ? Can I restore these files ? I don´t see the files when I check out the external harddrive. Obviously something like this has never happened to me. EDIT: I CHECKED, AND IT SEEMS I HAVE MOST OF THESE FILES ON ANOTHER, OLDER DRIVE. BUT OBVIOUSLY I DON'T WANT HISTORY TO REPEAT ITSELF
I had similar problem after going from v.5 to v.6 and didn't check default retention rules settings. Luckily I got images stored safely on other external drive, so I could restore it from there. If you didn't write additional data to your external HDD, you can try using Recuva or something similar to restore deleted files. If you get them back, I suggest you to try restoring it to VHD and then booting from them in VM to see if they are not corrupted or damaged.
