pbatard / rufus

The Reliable USB Formatting Utility
https://rufus.ie
GNU General Public License v3.0
28.36k stars 2.54k forks source link

Check device takes weeks, and can't abort or shut down #372

Closed HenryS137 closed 9 years ago

HenryS137 commented 10 years ago

I had a new 32GB SDHC card which I planned on using a lot, and I decided to experiment with your format option of "check device for bad blocks". I had already done a quick check under Windows.

Progress was extremely slow, but when I tried to abort the check it said that it would make the card unusable. So I continued the check for about four days, at which point it was about 40% of the way through the first pass, when I had to abort because Windows 8.1 was forcing a shutdown for a forced update.

When I tried the card after rebooting, it would not format or accept any operations.

To be forced to tie up a PC for weeks or ruin a card is NOT a good forced choice for your users without any warning.

Other than this, I have been a very happy user, and have preferred your utility to any of the other alternatives, many of which I have tried. I am using 1.4.10.514 on a very fast PC.

pbatard commented 10 years ago

Progress was extremely slow

The only way I can see this happening is if your device was connected at USB 1.0 low speed rate (~180 KB/s). At an expected USB 2.0 write speed of 5 MB/s, the default test should take a few hours at worst for a 32 GB card, or faster if you have a card with a decent write speed (5 MB/s is low, even for USB 2.0).

Rufus does write as fast as Windows lets it (we're just calling on the standard Windows API to write data during the bad blocks check, the same as the ones that are called to copy a file), so if the write operation was slow, it was either a configuration issue (plugged on a USB 1.0 port, such as the one you could find on an USB keyboard that doubles as a hub), a Windows issue (might have some heavy handed security application filtering access to USB devices), or something wrong with your card.

4 days for 40% of the first pass on a 32 GB card means that 32000 * .8 = 25600 MB or data was written in 4 days, which gives an approximate write speed of circa 75 KB/s, which is not something you would ever expect from an USB 2.0 or USB 3.0 device. I also never heard of anybody else seeing that slow a speed with the bad blocks check with Rufus, so I can only conclude that there is something wrong with either your PC or your card.

I have also never heard of a non-fake card or USB device actually becoming unusable after interrupting a bad blocks check. I have interrupted the Rufus bad blocks check many times, on many different devices, and I've had many reports of users also doing so after finding that the test took longer than they expected, but these devices were all usable afterwards. The warning notice is mostly there to prevent novice users from believing that their flash device is destroyed, when it is simply not formatted, but there again, I have never heard from anyone seeing a device rendered inoperable from running a bad blocks check (apart from people using fake USB flash drives, which are not devices you can use to their advertised capacity in the first place anyway).

I would be very curious to get the full log from Rufus when you plug your card. I would also encourage you to find out what the Windows device manager shows when you plug it in.

But really, the fact that your write speed was abnormally slow to start with seems to indicate that your card had an issue in the first place. The speed you saw is not common at all!

HenryS137 commented 10 years ago

The speed was excellent on everything but the check blocks. I am using a USB3.0 card reader on a fast computer. Formatting and other functions were quick and fine on that same card. It was just when I did the block check that it went into super slow mode.

The same reader is functioning fine with identical cards, I am just going to avoid the check blocks function. I can't afford to kill more cards or tie up the computer.

The card was new, and tested fine and quickly under Windows. It was ONLY when I used Rufus to check blocks that I ran into the problems.

Henry

I will try plugging the card into Rufus and get some log info. If I can, I'll send it along.

pbatard commented 10 years ago

The speed was excellent on everything but the check blocks. I am using a USB3.0 card reader on a fast computer. Formatting and other functions were quick and fine on that same card.

Unless you changed the options, formatting will use "quick format", which is very fast even on super slow USB speeds, because it only writes a few blocks (hence the name). I'd be curious to find out what happens if you use Rufus or the Windows formatting tool with the "quick format" option unchecked.

This may also be an issue with the card reader controller trying to optimize access when partitions are being accessed, but not when raw blocks are. I have also been using an USB 3.0 card reader with an UHS-I micro SD card and found that while I could get speeds of 30 MB/s or greater when writing files onto a partition (UHS-I speeds), writing raw blocks, which is what the bad blocks check does, was a lot slower (but still in the tens of MB/s).

Are your card and card reader UHS-I by any chance?

HenryS137 commented 10 years ago

I did do a non-quick format before. And imaged it with an ISO.

Here is the log when I attempt to do a format now:

(The references to the Passport are another USB 3 device on the same USB card.)

Partition 1:
  Type: FAT32 LBA (0x0c)
  Size: 31.2 GB (33553383424 bytes)
  Start Sector: 2048, Boot: No, Recognized: Yes

Format operation started
Requesting disk access...
Opened drive \\.\PHYSICALDRIVE9 for write access
Will use 'M:' as volume mountpoint
Analyzing existing boot records...
Drive has an unknown Master Boot Record
Volume has an unknown Partition Boot Record
Deleting partitions...
Clearing MBR/PBR/GPT structures...
Erasing 128 sectors
Partitioning (MBR)...
Closing existing volume...
Waiting for logical drive to reappear...
Formatting (FAT32)...
Using cluster size: 16384 bytes
Quick format was selected
Creating file system: Task 1/5 completed
Creating file system: Task 2/5 completed
Creating file system: Task 3/5 completed
Creating file system: Task 4/5 completed
Creating file system: Task 5/5 completed
Error while formatting
Format error: Undetermined error while formatting.
Re-mounted volume as 'M:' after error

Could not get node connection information (V2) for device 'USB\VID_1058&PID_0748\575837314335323932343530': [0x00000001] Incorrect function.
Found USB 3.0 device 'WD My Passport 0748 USB Device' (1058:0748)
Device eliminated because it was detected as an USB Hard Drive (score 23 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- SD/MMC USB Device' (????:????)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Generic- Compact Flash USB Device' (0BDA:0111)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Apple iPod USB Device' (05AC:1209)
Device eliminated because it was detected as an USB Hard Drive (score 10 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- MS/MS-Pro USB Device' (????:????)
Ignoring autorun.inf label for drive L: No media
Found USB device 'Generic- SM/xD-Picture USB Device' (????:????)
Device eliminated because it appears to contain no media
Could not get node connection information (V2) for device 'USB\VID_05E3&PID_0743\0000006072': [0x00000001] Incorrect function.
Found USB 3.0 device 'Generic STORAGE DEVICE USB Device' (05E3:0743)
2 devices found
No volume information for drive 0x89
Disk type: Removable, Sector Size: 512 bytes
Cylinders: 4079, TracksPerCylinder: 255, SectorsPerTrack: 63
Partition type: MBR, NB Partitions: 1
Disk ID: 0x00000001
Drive has an unknown Master Boot Record
Partition 1:
  Type: Small FAT16 (0x04)
  Size: 31.3 GB (33554432000 bytes)
  Start Sector: 0, Boot: No, Recognized: Yes
Could not get node connection information (V2) for device 'USB\VID_1058&PID_0748\575837314335323932343530': [0x00000001] Incorrect function.
Found USB 3.0 device 'WD My Passport 0748 USB Device' (1058:0748)
Device eliminated because it was detected as an USB Hard Drive (score 23 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- SD/MMC USB Device' (????:????)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Generic- Compact Flash USB Device' (0BDA:0111)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Apple iPod USB Device' (05AC:1209)
Device eliminated because it was detected as an USB Hard Drive (score 10 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- MS/MS-Pro USB Device' (????:????)
Ignoring autorun.inf label for drive L: No media
Found USB device 'Generic- SM/xD-Picture USB Device' (????:????)
Device eliminated because it appears to contain no media
Could not get node connection information (V2) for device 'USB\VID_05E3&PID_0743\0000006072': [0x00000001] Incorrect function.
Found USB 3.0 device 'Generic STORAGE DEVICE USB Device' (05E3:0743)
2 devices found
No volume information for drive 0x89
Disk type: Removable, Sector Size: 512 bytes
Cylinders: 4079, TracksPerCylinder: 255, SectorsPerTrack: 63
Partition type: MBR, NB Partitions: 1
Disk ID: 0x00000001
Drive has an unknown Master Boot Record
Partition 1:
  Type: Small FAT16 (0x04)
  Size: 31.3 GB (33554432000 bytes)
  Start Sector: 0, Boot: No, Recognized: Yes
Could not get node connection information (V2) for device 'USB\VID_1058&PID_0748\575837314335323932343530': [0x00000001] Incorrect function.
Found USB 3.0 device 'WD My Passport 0748 USB Device' (1058:0748)
Device eliminated because it was detected as an USB Hard Drive (score 23 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- SD/MMC USB Device' (????:????)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Generic- Compact Flash USB Device' (0BDA:0111)
Device eliminated because it appears to contain no media
Found USB 2.0 device 'Apple iPod USB Device' (05AC:1209)
Device eliminated because it was detected as an USB Hard Drive (score 10 > 0)
If this device is not an USB Hard Drive, please e-mail the author of this application
NOTE: You can enable the listing of USB Hard Drives in 'Advanced Options' (after clicking the white triangle)
Found USB device 'Generic- MS/MS-Pro USB Device' (????:????)
Ignoring autorun.inf label for drive L: No media
Found USB device 'Generic- SM/xD-Picture USB Device' (????:????)
Device eliminated because it appears to contain no media
Could not get node connection information (V2) for device 'USB\VID_05E3&PID_0743\0000006072': [0x00000001] Incorrect function.
Found USB 3.0 device 'Generic STORAGE DEVICE USB Device' (05E3:0743)
2 devices found
Disk type: Removable, Sector Size: 512 bytes
Cylinders: 4079, TracksPerCylinder: 255, SectorsPerTrack: 63
Partition type: MBR, NB Partitions: 4
Disk ID: 0x1B5F3E20
Drive has an unknown Master Boot Record
Partition 1:
  Type: FAT32 LBA (0x0c)
  Size: 31.2 GB (33553383424 bytes)
  Start Sector: 2048, Boot: No, Recognized: Yes
Partition 2:
  Type: Minix (0x80)
  Size: 6.2 MB (6488064 bytes)
  Start Sector: 6293568, Boot: No, Recognized: Yes
Partition 3:
  Type: Windows Hibernation (0x84)
  Size: 0 bytes (0 bytes)
  Start Sector: 0, Boot: No, Recognized: Yes
Partition 4:
  Type: Venix 80286 (0x40)
  Size: 0 bytes (0 bytes)
  Start Sector: 0, Boot: No, Recognized: Yes
pbatard commented 10 years ago

I did do a non-quick format before. And imaged it with an ISO.

What speed were you getting then. If you did a "slow" format, and even if you have a fast car, even at USB 3.0 speed, it probably took quite a bit of time, didn't it? Do you remember how long it took?

Here is the log when I attempt to do a format now

So your device is still detected...

Erasing 128 sectors

... and it can be written to in raw mode (if writing any of these 128 blocks didn't work, you would get an error here). So my guess is that it probably works fine, but it's just the Windows formatting API (which is what Rufus will call) that has some issue with it.

Here's what I would try:

HenryS137 commented 10 years ago

Just a bit of an update:

I tried doing a slow format again. It progresses very rapidly to 25%, then slowed down to extremely slow as before, and I aborted after an overnight where it did not progress by 1%. I tried to cancel at 29%. It hung "Cancelling - Please Wait", and never finished cancelling. Had to use Task Manager to force an exit from Rufus after an hour or so. Cancelling (from main app) was last entry in log before I forced the quit. After I used Task Manager to end Rufus, when I tried to run Rufus again it reported: Another instance detected. Another Rufus application is running. Please close the first application before running another one. Killing that process in Task Manager and repeating just continued the loop, so I rebooted the PC. Rufus did not show as a running process after ending, until I ran it again. So there was nothing else I could find to do in Task Manager to put the PC in a state where Rufus would run again without encountering the "Another Instance detected". At least that sequence was better than immediately aborting in the format attempt, from the point of view of progress towards a complete format.

Here is the log up to the point I cancelled:

Format operation started
Requesting disk access...
Opened drive \\.\PHYSICALDRIVE9 for write access
Will use 'M:' as volume mountpoint
Analyzing existing boot records...
Drive has a Zeroed Master Boot Record
Volume does not have an x86 Partition Boot Record
Deleting partitions...
Clearing MBR/PBR/GPT structures...
Erasing 128 sectors
Partitioning (MBR)...
Closing existing volume...
Waiting for logical drive to reappear...
Formatting (NTFS)...
Using cluster size: 4096 bytes
Slow format was selected
Cancelling (from main app)

I tried doing the check drive and aborting again, it still was in the same state.

After all the time it has been under operations, even if the card was good to start with it may be damaged now, so I will try some of the other suggestions like the Alt-L, but assume the card is responsible for the results, not Rufus.. I just offer the above for any interest you might have.

Thanks again!

Henry

pbatard commented 10 years ago

Thanks for the update. The fact that the non-quick format seems to slow down without reason after 1/4th of the operation seems to indicate an issue with the card indeed. At this stage of the formatting operation, Rufus has handed control to the Windows native formatting functions (the same as the one being used by the regular Windows formatting utility), so I expect the exact same result even when not using Rufus.

I'd encourage you to try Alt-L if you can, keeping in mind that this alternate formatting method, which only applies for FAT32, will always use quick format, so even if your card has an issue past the 25% mark and higher, you may only find out about it after you already have copied more than 8 GB of data onto it.

Please let me know how things work out with the other cards (either using Rufus or using the Windows native format with the "quick format" option unchecked). If you also see failures past the 25% mark, it probably means that you may have gotten a bunch of "fake" cards, which are memory devices that are set to inaccurately report the actual amount of flash memory they hold (e.g. the firmware on it may report 32GB, when there is only 8 GB of flash), to trick users into believing they have a good card, until they start writing more data than the actual size.

Unfortunately, this is not as uncommon as should be when buying from unscrupulous cheap retailers, and the reason why the bad blocks check in Rufus also has a feature to detect fakes (provided the flash device does not corrupt itself during such a test).

GlassedSilver commented 6 years ago

I just ran into this myself... No matter the hardware involved, the potential that this might leave the computer busy for a VERY long time should certainly be noted beforehand and CERTAINLY that cancelling - for whatever reason - is by design "uninterruptible".

This ticket was opened over 3 years ago... whew!

lock[bot] commented 5 years ago

This thread has been automatically locked since there has not been any recent activity after it was closed. Please open a new issue if you think you have a related problem or query.