David Koski writes:
> <snip log>
> Unable to read 621621 bytes from /home/backuppc/pc/mail/new//f%2f/RStmp
> got=208181, seekPosn=23909067 (1536,253,10003,24117248,24576026)Unable to read 621621 bytes from /home/backuppc/pc/mail/new//f%2f/RStmp
> <snip log>
> The second line is repeated about 125 times. I thought it might be a disk read
> error so I did "dd if=/dev/hda of=/dev/null bs=1M" and the same for hdc. (It
> is using software raid with /dev/hda and dev/hdc). It produced no errors and
> the system logs reveal nothing. Any clues as to what it might be?
RStmp is a temporary file created to rebuild the incoming file
deltas. For smaller files this is done in memory. It is created
by uncompressing the original file from the last full backup.
The file is really 24117248 bytes long. But the attributes
say that it is 24576026 bytes long. It attempts to read
621621 bytes from position 23909067, which only returns
208181 bytes since the file is only 24117248 bytes.
So for some reason the actual uncompressed file size doesn't
match the attributes in the attrib file form the previous full
backup. I can't explain why.
Are all 125 error messages similar; ie: is this the same file
in each case?
Is rsync checksum caching turned on?
Get latest updates about Open Source Projects, Conferences and News.