In the last episode (Jul 26), Brandon D. Valentine said:
On Fri, 19 Jul 2002, David Miller wrote:
A year ago there was a problem with backing up files larger than
either 2GB or 4GB, I forget which. A beta version of star would
handle it, but all the native versions of tar and gtar failed.
On Fri, 19 Jul 2002, David Miller wrote:
A year ago there was a problem with backing up files larger than either
2GB or 4GB, I forget which. A beta version of star would handle it, but
all the native versions of tar and gtar failed.
That's often not a problem, but if you're backing up db
In the last episode (Jul 21), [EMAIL PROTECTED] said:
On Thu, 18 Jul 2002, Mark W. Krentel wrote:
Dump on a live FS is always risky. FreeBSD in 4.x and earlier
will have up to about a 30 second delay before a write() makes it
to physical disk.
Is this regardless of the sync(8)
On Thu, 18 Jul 2002, Mark W. Krentel wrote:
Dump on a live FS is always risky. FreeBSD in 4.x and earlier will have
up to about a 30 second delay before a write() makes it to physical disk.
Is this regardless of the sync(8) command used ? And if so - what does
sync(8) actually sync - and
Do you check your backups, or does Amanda do it for you? I think dump
is on the way out in Linux.
I've managed to restore them when needed (despite Redhat's best
efforts over the years, including shipping a version of restore
that couldn't restore symlinks). In general we don't store users
Dump on a live FS is always risky. FreeBSD in 4.x and earlier will have
up to about a 30 second delay before a write() makes it to physical disk.
Ok, assume all the writes have finished, we wait 1-2 minutes before
starting dump, no new writes happen during the dump, then are we
assured that
After upgrading some Redhat machines to 1GB of ram it became nearly
impossible to dump any filesystem without dump going crazy trying
to read nonexistant blocks (previously it had worked fine). Upgrading
the version of the linux dump program which we use helped significantly
and now we can
Mark W. Krentel wrote:
Dump still works on a mounted file system in Freebsd, right? That is,
a write that completes before dump is started will be in the dump,
even if the data is in memory? I don't mean writing to a file during
the dump, that's a separate problem.
I only recently
On Wed, Jul 17, 2002 at 01:29:53AM -0400, Mark W. Krentel wrote:
I only recently learned that this doesn't work in Linux and I wanted
to check that it's (still?) ok in Freebsd. Apparently, in the 2.4
Linux kernels, the buffer and page caches make it impossible for dump
to always get the
Dump still works on a mounted file system in Freebsd, right? That is,
a write that completes before dump is started will be in the dump,
even if the data is in memory? I don't mean writing to a file during
the dump, that's a separate problem.
I only recently learned that this doesn't work in
10 matches
Mail list logo
