Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- From: "Nick Tucker" <nick.tucker@xxxxxxxxxxxxxxxxx>
- Date: Thu, 11 Sep 2008 13:31:01 +0100
Hi Alexander
Thanks for repling.
The code looks like this
HANDLE h =CreateFile("testfile.txt",GENERIC_READ|GENERIC_WRITE,0,NULL,CREATE_NEW,FILE_FLAG_WRITE_THROUGH,NULL);
BYTE b[1024]={0};
DWORD w;
WriteFile(h,b,sizeof(b),&w,NULL);
POWER OFF!!!
On restart, the file testfile.txt exists but is zero bytes long because although
the file data has been written to disk, the metadata has not.
So our so-called safe write to disk, is in this situation completely useless!
I think I've now grasped why this is so - you MUST call FileFlushBuffers() if
the data you're writing has not previously been written out and fully flushed by
FileFlushBuffers() (or by the file being closed).
So if we add the steps:
FlushFileBuffers();
LONG l;
SetFilePointer(h,0,&l,SEEK_SET);
memset(b,1,sizeof(b));
We can then happily cut the power, confident that the data will have been
overwritten with 1's, even though we haven't flushed the file.
I hope you're of my opinion that Microsoft could do better in spelling out
this behaviour!
Thanks
Nick
"Alexander Grigoriev" <alegr@xxxxxxxxxxxxx> wrote in message news:u9N%23Fx7EJHA.4104@xxxxxxxxxxxxxxxxxxxxxxx
What does your CreateFile call look like?
"Nick Tucker" <nick.tucker@xxxxxxxxxxxxxxxxx> wrote in message news:A1959885-3F61-4C46-A742-4AA2FB55FEB2@xxxxxxxxxxxxxxxxHi everyone
If I create a file using CreateFile(), specifying the
FILE_FLAG_WRITE_THROUGH flag, then write to that file, the file still shows
as zero bytes long in the directory and in Windows Explorer, and if I then
turn the power off, that file is still zero bytes long on restart, so the
data I wrote is unavailable. I have tried this test on a variety of
operating systems including Vista, XP and Windows Server 2003/2008 using a
variety of hardware platforms with both SATA and SCSI disk systems, and they
all behave the same way.
Surely the whole point of specifying that flag is that I want the data to be
written to disk, together with it's corresponding metadata, so if a
catastrophic failure occurs, the data is available to me on restart.
My researches of newsgroup articles on this subject suggest that if I want
this, then I also have to use FlushFileBuffers(), but if that is so then
surely I don't need to specify the write through flag since all file and
folder data will be committed to disk when I call that function.
So, what is the point of specifying the write-through flag in this scenario?
I think the documentation of CreateFile() could do with a little
clarification!
Thanks in advance for responses.
Nick
.
- Follow-Ups:
- Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- From: Alexander Grigoriev
- Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- References:
- CreateFile() and FILE_FLAG_WRITE_THROUGH
- From: Nick Tucker
- Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- From: Alexander Grigoriev
- CreateFile() and FILE_FLAG_WRITE_THROUGH
- Prev by Date: Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- Next by Date: GINA: token becomes invalid all of a sudden
- Previous by thread: Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- Next by thread: Re: CreateFile() and FILE_FLAG_WRITE_THROUGH
- Index(es):
Relevant Pages
|