Failed to read data from the disk
My backup with arconis pretty much never completes without errors. With my old disk and now after replacing it with a new disk as well. Errors usually look like the one below:
The message allows to retry, ignore or cancel. Retry sometimes will continue or not. But if it continues it will usually produce the same error some time (minutes to hours) later with a different sector number. I have found some older posts about similar errors here (can't link them due to forum limits) who suggest it might be an issue with the disk and to run CHKDSK or similar tools. CHKDSK didn't help with my old disk and for the new disk arconis produces now the same errors.
Thanks for the suggestion. It doesn't look like it's related to the hidden partitions. After starting in recovery mode CHKDSK did find issues but won't fix them. Just complains about having not enough space to replace bad clusters. I tried also the software for the disk which finds errors as well but doesn't fix them producing "unknown error" when trying.
Whats strange s.m.a.r.t. says 97% of the spare area is still available and no issue listed there for the disk.
At least I know now it's a problem with the disk and not the software.
Mike, not sure if your comment about trying software for the disk includes the below, but offered in case it doesn't.
Dedicated diagnostic utilities from the disk manufacturers take the longest time to complete the checks, but provide the most accurate methods of checking whether the disk is good or needs replacement:
- Western Digital drives: Data Lifeguard Diagnostic for Windows
- Seagate disks: SeaTools for Windows
- HGST disks: HGST Windows Drive Fitness Test (WinDFT)
- Intel SSDs: Intel Memory and Storage Tool (GUI)
- Samsung drives: Samsung Magician
- ADATA drives: ADATA SSD ToolBox
- Kingston SSDs: Kingston SSD Toolbox, Kingston SSD Manager
- Transcend SSDs: Transcend SSD Scope
- Silicon Power disks: SP ToolBox
- Toshiba disks: Toshiba PC Diagnostic Tool Utility
- Crucial disks: Crucial Storage Executive
- SanDisk disks: SanDisk SSD Dashboard tool / SanDisk SSD Dashboard user guide
I have been using the Samsung Magican as it's a Samsung ssd. A scan with it shows a few defects but trying to repair them with Magican results in the message "unknown error". Doing another scan afterwards shows the same defects again.
Mike, in one of your posts you state:
After starting in recovery mode CHKDSK did find issues but won't fix them. Just complains about having not enough space to replace bad clusters.
This suggests that there is nowhere to relocate the content of the bad sectors; not sure why this would be the case. Two possibilities occur to me, the first is that there is no space on the drive, or that the drive firmware sets aside a limited amount of storage to be used when the content of bad clusters is moved. My recollection is that the latter is the case.
The partitions occupy all of the disk. I tried to shrink a partition but the partitioning software won't allow that before fixing the errors on the disk. From what I read with SSDs there is a spare area of sectors who are used for replacing bad ones. According to SMART 3% of that spare area has been used up so far:
Not sure why it won't use any more of the available spare. The errors did happen in a very big file but even if there would be to many bad sectors I would still assume chkdsk would replace as many sectors as it can before complaining. Whats also strange acronis shows for all of the bad sectors negative numbers. I have only selected the visible partition for the backup so hidden ones shouldn't be related with the number.
Mike, the only real solution here may be to have to do a full drive reformat but that will be a problem if you don't already have a good full disk backup to put back later, as otherwise you will need to do a clean install of Windows etc.
One possible alternative would be to try doing a full sector-by-sector backup with the option to ignore bad sectors. I haven't used this on SSD drives but have used it on failing HDD's in order to recover to another larger drive then run the necessary repair tools to deal with any bad data brought across. This typically was a very long running process taking 12+ hours top both backup and again to recover, plus an extended repair time.
If you have lots of user data on the drive, then consider copying or moving this to another drive while it is still operational. I try to limit my main OS drive to only having Windows plus all my installed applications, with all user data on either a second partition or second drive.
Ok, thanks.
I have copied the most important files to a backup location and run now a sector-by-sector backup with ignoring bad sectors like you suggested in case I missed some files.