Re: BACKUP/FAST (was Re: Backup Command)

On May 29, 2:05 pm, "Richard Brodie" <R.Bro...@xxxxxxxx> wrote:
"AEF" <spamsink2...@xxxxxxxxx> wrote in message

news:e90bf9d2-8593-434f-9cfc-85c881f2a257@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx

I don't know what it means to synchronize with the file system and why
write access is needed if no update is to be made.

I would guess it acquires write locks to force the filesystem to flush
its cache.

That's your guess. My guess is that the write lock is to prevent other
people/apps doing writes while BACKUP is doing its best to save a
consistent copy of the data.

Maybe rather than guessing we should hope for an informed answer from
someone with a clue, e.g. maybe once they've recovered from Thursday's
party?
.

Relevant Pages

  • Re: [PATCH 1/7] block: Add block_flush_device()
    ... some random filesystem shouldn't disable it. ... this whole insane belief that "write cache" has anything to do with "unable to flush" is just bogus. ... The way the barriers work does absolutely give you full ordering. ...
    (Linux-Kernel)
  • Re: vfs: Add MS_FLUSHONFSYNC mount flag
    ... cache should be flush or not on fsync/fdatasync. ... That way the filesystem can always do "the right thing" and call the ... it's built with a nonvolatile write cache or if it has no write cache. ... can tune it once, instead of needing to mount all 8 filesystems with the ...
    (Linux-Kernel)
  • Re: FUSE merging?
    ... > plays nicely with the page cache. ... on the underlying file server. ... > talking about a virtual filesystem where all calls should be passed ... We have exclusive open semantics but not locks in the Posix ...
    (Linux-Kernel)
  • Re: [PATCH 1/7] block: Add block_flush_device()
    ... [Although it may be related in that maybe the admin can _force_ a EOPNOTSUPP thing for when he wants to disable any "write barrier implies flush" thing. ... we may end up with an _implementation_ detail where we overload a potential QUEUE_FLUSH_EOPNOTSUPP flag with two meanings - either "the driver told me a barrier isn't supported" or "the admin set that same flag by hand to disable barrier-related flush commands". ... We have _CPU's_ that corrupt memory on cache writeback, and those things are a lot more tested than most driver firmware is. ... People _are_ supposed to be able to move their data around, and the filesystem shouldn't make judgement calls. ...
    (Linux-Kernel)
  • Re: BACKUP/FAST (was Re: Backup Command)
    ... I would guess it acquires write locks to force the filesystem to flush ...
    (comp.os.vms)