This introduces a new modifier to the output of the diff command. It
appears whenever two files being compared only differ in their content
but not in their metadata. As far as we know, under normal
circumstances, this should only ever happen if some kind of bitrot has
happened in the source file. The prerequisite for this detection to work
is that the right-side snapshot of the comparison has been created with
"backup --force".
Bug #1681 suggests that restic should not be nice to user and should
refrain from creating a mountpoint if it does not exist. Nevertheless,
it currently opens the repository before checking for the mountpoint's
existence. In the case of large or remote repositories, this process
can be time-consuming, delaying the inevitable outcome.
/restic mount --repo=REMOTE --verbose /tmp/backup
repository 33f14e42 opened (version 2, compression level max)
[0:38] 100.00% 162 / 162 index files loaded
Mountpoint /tmp/backup doesn't exist
stat /tmp/backup: no such file or directory
real 0m39.534s
user 1m53.961s
sys 0m3.044s
In this scenario, 40 seconds could have been saved if the nonexistence
of the path had been verified beforehand.
This patch relocates the mountpoint check to the beginning of the
runMount function, preceding the opening of the repository.
/restic mount --repo=REMOTE --verbose /tmp/backup
Mountpoint /tmp/backup doesn't exist
stat /tmp/backup: no such file or directory
real 0m0.136s
user 0m0.018s
sys 0m0.027s
Signed-off-by: Sébastien Gross <seb•ɑƬ•chezwam•ɖɵʈ•org>
inodes are only unique within a device. Use the HardlinkIndex from the
restorer instead of the custom (broken) hashmap to correctly account for
both inode and deviceID.
Go 1.21 has switched the default from GOARM=5 to GOARM=7. So far there
have been complaints from Raspberry Pi 1 users, as the first raspberry
pi version only supports ARMv6. Exclude older ARM versions as these are
likely not relevant (rest-server also only supports ARMv6/7) and enforce
the use of software floating point emulation.
Allow setting custom arguments for the `sftp` backend, by using the
`sftp.args` option. This is similar to the approach already implemented
in the `rclone` backend, to support new arguments without requiring
future code changes for each different SSH argument.
Closes #4241
This introduces the inode attribute to the JSON output emitted for nodes
in `ls` and matches in `find`. There doesn't seem to be any discernible
reason to omit the inode and it can be useful in scripting scenarios.
Linux allows the use of non-`user.` extended attributes on symlinks. One
of the main users of this functionality is SELinux's `security.selinux`
xattr for storing a path's label. By storing symlink xattrs, restic is
now suitable for backing up the root filesystem on Linux distributions
that use SELinux.
This commit adds support for symlink xattrs when backing up data,
restoring data, and mounting snapshots via a fuse mount. All calls to
the xattr library have been updated to the use `L` variants of the
various functions, which always operate on the path given, without
following symlinks.
Fixes: #4375
Signed-off-by: Andrew Gunnerson <accounts+github@chiller3.com>
The ETA restic displays was based on a rate computed across the entire
backup operation. Often restic can progress at uneven rates. In the worst
case, restic progresses over most of the backup at a very high rate and
then finds new data to back up. The displayed ETA is then unrealistic and
never adapts.
Restic now estimates the transfer rate based on a sliding window, with the
goal of adapting to observed changes in rate. To avoid wild changes in the
estimate, several heuristics are used to keep the sliding window wide
enough to be relatively stable.
Modifies format module to add options for human readable storage size formatting, using size parsing already in ui/format.
Cmd flag --human-readable added to ls and find commands.
Additional option added to formatNode to support printing size in regular or new human readable format
restic must be able to refresh lock files in time. However, large
uploads over slow connections can cause the lock refresh to be stuck
behind the large uploads and thus time out.
Since 0.15 (#4020), inodes are generated as hashes of names, xor'd with
the parent inode. That means that the inode of a/b/b is
h(a/b/b) = h(a) ^ h(b) ^ h(b) = h(a).
I.e., the grandchild has the same inode as the grandparent. GNU find
trips over this because it thinks it has encountered a loop in the
filesystem, and fails to search a/b/b. This happens more generally when
the same name occurs an even number of times.
Fix this by multiplying the parent by a large prime, so the combining
operation is not longer symmetric in its arguments. This is what the FNV
hash does, which we used prior to 0.15. The hash is now
h(a/b/b) = h(b) ^ p*(h(b) ^ p*h(a))
Note that we already ensure that h(x) is never zero.
Collisions can still occur, but they should be much less likely to occur
within a single path.
Fixes #4253.
