Fix Disk help please

peterworks

Ye Olde Bowler
I fitted a new 1tb disk last week and all working perfectly.
Today I decided to run a fixdisk -l. It was running for over four hours and I stopped it. The log file was over 500 mb. I have attached a small bit of the log file's output which shows what it was doing. Should I run it again and just let it run or is there something else I can do ?
I could not open the log in web if but have downloaded it via FTP should the full log be useful.
 

Attachments

  • fix-disk log example.txt
    949 bytes · Views: 7
Though I had better add Attributes and Disk Information...
 

Attachments

  • Attributes.jpg
    Attributes.jpg
    181.9 KB · Views: 8
  • Disk Information.jpg
    Disk Information.jpg
    75.4 KB · Views: 8
Just a bit from a newbie here so don’t know if it will be any use, I upgraded to a 2tb drive and had the same it was taking forever, so as I had a new box and didn’t mind losing anything I just did a format from the box and then redid the fixdisk and it completed in a couple of hours or less, and since then it all works in a few hours if I try it again, as I said don’t know if this helps till someone more knowledgeable comes along
 
Do you routinely visit your doctor regardless of being unwell? Oh well, if you have no better way to spend your time...
 
Do you routinely visit your doctor regardless of being unwell? Oh well, if you have no better way to spend your time...
No but I am invited by my doctor to a Wellman clinic every couple of years.
Also, in this case, running Fixdisk highlighted what 'may' be a problem...
 
Wouldn't a flag in the SMART stats (via WebIf) give notice of possible problems. I thought that was a main purpose of them.
 
SMART is the physical health of the drive, fixdisk looks at the integrity of the file system data structures. There is a very slight overlap in that the physical system can mark sectors as suspect and these need to be accounted for in the file system, so AFAIK there is no need to run fixdisk unless the SMART stats show reallocation errors, or there are apparent faults in the file system (such as going read-only, or continuous delete, or other weirdnesses to numerous to list).

That's my view anyway, take it or leave it (until an expert comments).
 
or there are apparent faults in the file system
Sometimes there are faults and they are not apparent. It is better to fix them while they are fixable and before they become catastrophic.
This is why any decent Linux system runs fsck after a certain number of days or filesystem mounts.
Of course Humax skipped the "decent" bit and don't bother with any of it until it becomes catastrophic - at which point you lose all your recordings as you have to format to recover.
The CF obviously fixes their shortcomings.

On big disks it's probably worth doing a custom format, as it reduces the number of inodes, which reduces the time fixdisk takes to run significantly.
See af123's 2TB blog for details.
 
Fair enough, but this is a 1TB disk which should complete fixdisk fairly quickly but isn't despite being relatively fresh. Does that suggest a defect in the initial formatting?
 
Does that suggest a defect in the initial formatting?
The first time I ran fixdisk, not that long after re-formatting, I was surprised by the number of errors. I have never done any proper analysis though as it's always such a nuisance unless you have a blank spare machine.
 
I am trying to format the hard disk but get the error "Cannot format the hard disk. The capacity is too large". I believe this has come up before but have been unable to find any reference when doing a search.
Can anyone shed some light on this please.
 
I have RMA'd, reformatted, reinstalled WebIf and run fixdisk twice and am having the same problems. I am up to Inode 14302819 (not sure what the total may be).
I will now put everything in WebIf back to where it was and run a fixdisk for about an hour a day until everything is okay. A bit annoying as it is a 'new' disk (which the SMART stats seem to confirm)
Thank you for all your help.
 
Back
Top