On 25/09/2019 15:36, Rowan Worth wrote:
On Wed, 25 Sep 2019 at 12:58, Simon Slavin <slav...@bigfraud.org> wrote:
When I first learned the SQLite had problems with Network File Systems I
read a ton of stuff to learn why there doesn't seem to be a Network File
Systems that implements locking properly. <snip>
Still, I wonder why someone working on a Linux network file system, or
APFS, or ZFS, hasn't done it.
I'm not sure what your definition of "locking properly" is or when your
research was done, but POSIX advisory locks¹ work just fine on linux over
nfs (since at least v3) and lustre.
¹ That's the F_SETLK/F_GETLK/F_SETLKW commands via the fcntl() syscall,
which is also sqlite's default locking mechanism under UNIX.
I don't see it as that much of a problem, I've been locking
database-type files over NFS/RFS/DECNET since the 1980s, and SMB since
the 1990s.
Now, there have been a *lot* of crappy implementations of NFS out there,
probably the crappiest currently in use is the Linux version, but it is
better than it used to be (I wonder if sharing a file system still
causes the entire NFS server to re-start), and let's not mention the
reasoning behind, "Why should we drop back to NFSv3 if the NFSv4
initiation fails?"
Although I have had to convince a few people of the right /way/ to take
out a lock...
Cheers,
Gary B-)
_______________________________________________
sqlite-users mailing list
sqlite-users@mailinglists.sqlite.org
http://mailinglists.sqlite.org/cgi-bin/mailman/listinfo/sqlite-users
