> From: Brandon Allbery [mailto:[email protected]] > > On Wed, Oct 30, 2013 at 1:54 PM, Edward Ned Harvey (lopser) > <[email protected]> wrote: > I agree, except, that "usable != clean" is irrelevant. The whole point of the > journal is that your filesystem doesn't *need* to be clean. Any part of the > FS > that isn't clean, by definition, you are not using. As soon as you use it, it > becomes clean. > > This sounds like a severe misunderstanding of how journaling works to me. It > is not, in fact, a magic wand that fixes filesystems on the fly; it is a tool > to try > to avoid corruption and to speed up bringing a filesystem up to date. And, > poorly implemented and applied with the mentality you're showing, it is > quite good at covering up serious corruption until it is far too late to fix > it.
No. What you said sounds like a misunderstanding. The journal is used to systematically log the intent to make all changes which could possibly result in filesystem corruption, if the change were to fail before completion. Therefore, at a later time, all corruption that could possibly have been previously introduced is immediately detectable and undoable on the fly. Correct. It's not a magic wand. It's a logical deterministic wand. "poorly implemented" is the main thing that makes your paragraph sound like a misunderstanding. The system administrator doesn't need to do anything to implement journaling. You just install the OS, including a journaling filesystem (ext3, ext4, and others) and now your filesystem is journaling. You have to trust that the kernel and filesystem code developers released a robust and reliable product. They are the implementers. Also, "covering up corruption until it's too late" seems to demonstrate a misunderstanding. You're talking about headlight fluid. _______________________________________________ Tech mailing list [email protected] https://lists.lopsa.org/cgi-bin/mailman/listinfo/tech This list provided by the League of Professional System Administrators http://lopsa.org/
