fixdisk - inode has illegal block(s)

Hugo

New Member
Hello,

I've searched the forum but not found anything similar, so sorry if this has been covered before.

I just bought a second-hand drive to upgrade my HDR-FOX T2 to 2TB. I formatted it, ran a fixdisk with no errors. I started copying over programmes from the old 500GB drive (maybe 80GBs worth). Now when I run fixdisk again I get a seemingly endless list of errors very similar to the following:

Code:
Inode 1676 has compression flag set on filesystem without compression support.  Clear? yes

Inode 1676 has a bad extended attribute block 133926307.  Clear? yes

Inode 1676 has INDEX_FL flag set but is not a directory.
Clear HTree index? yes

Inode 1676, i_size is 8257387992123064391, should be 0.  Fix? yes

Inode 1676, i_blocks is 2442655043, should be 0.  Fix? yes

Inode 1665 has compression flag set on filesystem without compression support.  Clear? yes

Inode 1665 has illegal block(s).  Clear? yes

Illegal block #0 (4072023470) in inode 1665.  CLEARED.
Illegal block #1 (1801876663) in inode 1665.  CLEARED.
Illegal block #2 (988287222) in inode 1665.  CLEARED.
Illegal block #3 (2367994143) in inode 1665.  CLEARED.
Illegal block #4 (1765709085) in inode 1665.  CLEARED.
Illegal block #5 (4055406825) in inode 1665.  CLEARED.
Illegal block #6 (1516513861) in inode 1665.  CLEARED.
Illegal block #7 (1343610416) in inode 1665.  CLEARED.
Illegal block #8 (3162125683) in inode 1665.  CLEARED.
Illegal block #9 (3307167322) in inode 1665.  CLEARED.
Illegal block #10 (1425700832) in inode 1665.  CLEARED.
Too many illegal blocks in inode 1665.
Clear inode? yes

Inode 1644 has a bad extended attribute block 126749409.  Clear? yes

Inode 1644 has INDEX_FL flag set but is not a directory.
Clear HTree index? yes

Inode 1644, i_size is 1346302851538933158, should be 0.  Fix? yes

Inode 1644, i_blocks is 280342766, should be 0.  Fix? yes

Inode 1478 has compression flag set on filesystem without compression support.  Clear? yes

Inode 1478 has a bad extended attribute block 181180381.  Clear? yes

Inode 1478 has illegal block(s).  Clear? yes

Illegal block #0 (1718277091) in inode 1478.  CLEARED.
Inode 1478 is too big.  Truncate? yes

Block #1 (3191435) causes symlink to be too big.  CLEARED.
Illegal block #2 (673328604) in inode 1478.  CLEARED.
Illegal block #3 (2542786138) in inode 1478.  CLEARED.
Illegal block #4 (1758549571) in inode 1478.  CLEARED.
Illegal block #5 (2300664629) in inode 1478.  CLEARED.
Illegal block #6 (3139957111) in inode 1478.  CLEARED.
Illegal block #7 (1286413470) in inode 1478.  CLEARED.
Illegal block #8 (2178276076) in inode 1478.  CLEARED.
Illegal block #9 (978088827) in inode 1478.  CLEARED.
Illegal block #10 (691935022) in inode 1478.  CLEARED.
Too many illegal blocks in inode 1478.
Clear inode? yes

You get the idea. The buffer of my telnet client only goes back 2000 lines and it's a lot more than that. Almost every inode seems to have errors. This doesn't happen with I run fixdisk on the original drive.

It's a Hitachi CinemaStar 5K2000 (HCS5C2020ALA632) and has 57,137 on-hours and 17,588 power cycles which both sound like quite a lot. No reallocated sectors though.

Are those errors actually fine? Or have I just bought a dodgy hard drive? I thought I'd be sustainable, but I guess there's a reason for second hand drives. The box itself seems to be behaving normally. Maybe I should have gone for a trusty Segate Pipeline..

Thanks for your time.
 
I just bought a second-hand drive to upgrade my HDR-FOX T2 to 2TB.
My rule of thumb is that second hand drives cause far more trouble than the cost saving would justify.
Code:
Illegal block #0 (4072023470) in inode 1665.  CLEARED.
Illegal block #1 (1801876663) in inode 1665.  CLEARED.
Illegal block #2 (988287222) in inode 1665.  CLEARED.
Illegal block #3 (2367994143) in inode 1665.  CLEARED.
Illegal block #4 (1765709085) in inode 1665.  CLEARED.
Illegal block #5 (4055406825) in inode 1665.  CLEARED.
Illegal block #6 (1516513861) in inode 1665.  CLEARED.
Illegal block #7 (1343610416) in inode 1665.  CLEARED.
Illegal block #8 (3162125683) in inode 1665.  CLEARED.
Illegal block #9 (3307167322) in inode 1665.  CLEARED.
Illegal block #10 (1425700832) in inode 1665.  CLEARED.
Too many illegal blocks in inode 1665.
Clear inode? yes

Inode 1644 has a bad extended attribute block 126749409.  Clear? yes

Inode 1644 has INDEX_FL flag set but is not a directory.
Clear HTree index? yes

Inode 1644, i_size is 1346302851538933158, should be 0.  Fix? yes

Inode 1644, i_blocks is 280342766, should be 0.  Fix? yes

Inode 1478 has compression flag set on filesystem without compression support.  Clear? yes

Inode 1478 has a bad extended attribute block 181180381.  Clear? yes

Inode 1478 has illegal block(s).  Clear? yes

Illegal block #0 (1718277091) in inode 1478.  CLEARED.
Inode 1478 is too big.  Truncate? yes

Block #1 (3191435) causes symlink to be too big.  CLEARED.
Illegal block #2 (673328604) in inode 1478.  CLEARED.
Illegal block #3 (2542786138) in inode 1478.  CLEARED.
Illegal block #4 (1758549571) in inode 1478.  CLEARED.
Illegal block #5 (2300664629) in inode 1478.  CLEARED.
Illegal block #6 (3139957111) in inode 1478.  CLEARED.
Illegal block #7 (1286413470) in inode 1478.  CLEARED.
Illegal block #8 (2178276076) in inode 1478.  CLEARED.
Illegal block #9 (978088827) in inode 1478.  CLEARED.
Illegal block #10 (691935022) in inode 1478.  CLEARED.
Too many illegal blocks in inode 1478.
Clear inode? yes

You get the idea. The buffer of my telnet client only goes back 2000 lines and it's a lot more than that. Almost every inode seems to have errors. This doesn't happen with I run fixdisk on the original drive.
Those errors don't sound good to me. Are you running the latest version of the custom firmware and packages? If not update everything and try again.
 
Hitachi CinemaStar 5K2000
I don't think we've heard of these being used before. Why wouldn't you go with a drive we've found satisfactory?
 
I don't think we've heard of these being used before. Why wouldn't you go with a drive we've found satisfactory?
I seem to recall similar problems before but since it didn't affect me I didn't pay too much attention.
Of course I cant find the thread now, possibly https://hummy.tv/forum/threads/fixdisk-stuck-in-loop-showing-illegal-block.9116/

@Hugo
Did you let the Humax format the drive or was it formatted on another machine?
What level of CF do you have installed? It should be 3.13
Assuming you still have the files on your original disk I would reformat the disk on the humax and try again copying 1 programme to start with and then rerun fix-disk.
What method are you using to do the copying?
 
Hitachi CinemaStar 5K2000
The disks were advertised as being designed for PVR use but the data sheets include the phrase
* Actual storage may vary depending on the compression rate applied. Capacities represented may not be combined.
This suggests to me that the disk supports compression but the Humax software doesn't.
You need to find a way to turn this feechur off
 
I formatted it
How? Via the Humax on-screen menus? I saw same on my boxes.
Suggest you blat it all and start again, using mkfs.ext3 at the command line. Details are on the 2TB Disk Installation Blog (ignore the partitioning bit):
Code:
humax# mkfs.ext3 -m 0 -O sparse_super /dev/sda1
humax# mkfs.ext3 -m 0 -O sparse_super -T largefile /dev/sda2
humax# mkfs.ext3 -m 0 -O sparse_super /dev/sda3
making sure you only have the hard disk connected (no USBs).
Maybe I should have gone for a trusty Segate Pipeline..
It's nothing to do with the physical drive. It's a filesystem problem apparently created by the crappy default Humax formatting software.
 
It's a Hitachi CinemaStar 5K2000 (HCS5C2020ALA632) and has 57,137 on-hours and 17,588 power cycles which both sound like quite a lot.
Yes. How much did you pay for that? A proper new one is only a few 10s of quid.
 
Thanks so much for your responses!

Intially, I checked the drive on my linux machine, then installed it in the Humax and formatted it using the on-screen menu. Then re-installed webif and started copying using the on-screen menus. Interestingly, when I tried to format the drive again through the on-screen menu, it came up with the message 'Cannot format the hard disk. The capacity is too large.'

But the solution was to reformat the partitions manually following @prpr's instructions:
Suggest you blat it all and start again, using mkfs.ext3 at the command line. Details are on the 2TB Disk Installation Blog (ignore the partitioning bit):
Code:
humax# mkfs.ext3 -m 0 -O sparse_super /dev/sda1
humax# mkfs.ext3 -m 0 -O sparse_super -T largefile /dev/sda2
humax# mkfs.ext3 -m 0 -O sparse_super /dev/sda3
making sure you only have the hard disk connected (no USBs).

It's nothing to do with the physical drive. It's a filesystem problem apparently created by the crappy default Humax formatting software.

I've copied back a few programmes and recorded some things and now fixdisk completes without error :). So looks like it's solved 🤞. Thanks so much!

My rule of thumb is that second hand drives cause far more trouble than the cost saving would justify.
Next time I'll buy a new hard drive. I was being sustainable, but I should have learnt that you shouldn't buy second-hand hard drives.

I don't think we've heard of these being used before. Why wouldn't you go with a drive we've found satisfactory?
That's a useful post. Thanks for that.

I seem to recall similar problems before but since it didn't affect me I didn't pay too much attention.
Of course I cant find the thread now, possibly https://hummy.tv/forum/threads/fixdisk-stuck-in-loop-showing-illegal-block.9116/
Thanks. Not sure why that didn't show up in my search.
 
it came up with the message 'Cannot format the hard disk. The capacity is too large.'
That happens if CF is running - the standard firmware can't get exclusive access to the HDD so can't format it, but wasn't programmed to expect such an eventuality therefore put up a misleading error message.

The solution is to is to enter Safe Mode first: WebIF >> Diagnostics >> Safe Mode (or on the Telnet menu).
 
Back
Top