> Perhaps the file was truncated to zero bytes on your local filesystem > before you ran "tahoe backup" or "tahoe cp" or used the WUI to upload > that file?
Im using mainly FTP interface, not commandline. I found two files with zero filesize (so with LIT cap), which should not be zero - backup of my thesis (size in megabytes) and one photo ( ~200kB). Both files were 100% closed on my filesystem during transfers. As I currently know what "LIT" is now, I can tell these two files were not corrupted in meaning of storage layer - they had correct cap, path, I saw them in directory listing etc. I think about bug in layer _before_ file is stored (something in FTP, transfers to helper, ...). Idea: As my computer is sometimes unstable and it powered off without any significant reason (solving at these days), Im pretty sure that some transfers to helper were cancelled. Is there any specific phase, when can helper decide it is empty file? Is this scenario possible? I think metadata about file (and it's link to directory) should be stored as final step, after uploading shares. So the worst scenario in case of transfer cancellation should be no file on grid (and not zero one). Marek P.S. Sorry for my english, Im too tired to fight with language at this time. > Please investigate! If Tahoe-LAFS is at fault then this is a critical > issue for us. Even if it is not at fault I'd like to know how to make > it easier for users to detect and correct this problem in the future. > > Regards, > > Zooko > _______________________________________________ > tahoe-dev mailing list > [email protected] > http://allmydata.org/cgi-bin/mailman/listinfo/tahoe-dev > _______________________________________________ tahoe-dev mailing list [email protected] http://allmydata.org/cgi-bin/mailman/listinfo/tahoe-dev
