percentage of verified groups, interpolating based on snapshot count
within the group. in most cases, this will also be closer to 'real'
progress since added snapshots (those which will be verified) in active
backup groups will be roughly evenly distributed, while number of total
snapshots per group will be heavily skewed towards those groups which
have existed the longest, even though most of those old snapshots will
only be re-verified very infrequently.
WalkDir does not follow symlinks by default anyway, and this behaviour
is not documented anywhere. e.g., if a sysadmin mounts 'extra storage'
for some backup group or type (not knowing that only metadata is stored
in those directories), GC will ignore all the indices contained within
and happily garbage collect their chunks..
gc: log index files found outside of expected scheme
for safety reason, GC finds and marks all index files below the
datastore base path. as a result of regular operations, only index files
within the expected scheme of <TYPE>/<ID>/<TIMESTAMP> should exist.
add a small check + warning if the index list contains index files out
side of this expected scheme, so that an admin with shell access can
investigate.
Dominik Csapak [Thu, 26 Nov 2020 08:27:31 +0000 (09:27 +0100)]
ui: fix usage of findRecord
findRecord does not match exactly, but only at the beginning and
case insensitive, by default. Change all calls to be case sensitive
and an exactmatch (we never want the default behaviour afaics).
else users have to manually search through a potentially very long task
log to find the entries that are different.. this is the same summary
printed at the end of a manual verify task.
if the manifest is signed/the contained archives/blobs are encrypted.
stored in 'unprotected' area, since there is already a strong binding
between key and manifest via the signature, and this avoids breaking
backwards compatibility for a simple usability improvement.
Dominik Csapak [Thu, 19 Nov 2020 09:02:52 +0000 (10:02 +0100)]
fix systemd-encoded upid strings in http client
since we systemd-encode parts of the upid string, and those can contain
characters that are invalid in urls (e.g. '\'), we have to percent encode
those
add a 'percent_encode_component' helper, so that we can maybe change
the AsciiSet for all uses at the same time
Fabian Ebner [Wed, 11 Nov 2020 09:07:03 +0000 (10:07 +0100)]
prune sim: make numberfields more similar to PBS's
by creating a new class that adds a clear trigger and also uses the
clear-trigger image. Code was taken from the one in PBS's prune window,
but we have default values here, so a bit of adapting was necessary. For
example, we don't want to reset to the original value (which might have
been one of the defaults) when clearing, but always to 'null'.
Stefan Reiter [Thu, 12 Nov 2020 14:50:08 +0000 (15:50 +0100)]
gc: treat .bad files like regular chunks
Simplify the phase 2 code by treating .bad files just like regular
chunks, with the exception of stat logging.
To facilitate, we need to touch .bad files in phase 1. We only do this
under the condition that 1) the original chunk is missing (as before),
and 2) the original chunk is still referenced somewhere (since the code
lives in the error handler for a failed chunk touch, it only gets called
for chunks we expect to be there, i.e. ones that are referenced).
Untouched they will then be cleaned up after 24 hours (or after the last
longer-running task finishes).
Reason 2) is also a fix for .bad files not being cleaned up at all if
the original is no longer referenced anywhere (e.g. a user deleting all
snapshots after seeing some corrupt chunks appear).
cond_touch_path is introduced to touch arbitrary paths in the chunk
store with the same logic as touching chunks.
Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
unprivileged users should only see the counts related to their part of
the datastore.
while we're at it, switch to a list groups, filter groups, count
snapshots approach (like list_snapshots) to speedup calls to this
endpoint when many unprivileged users share a datastore.
Fabian Ebner [Mon, 16 Nov 2020 09:22:47 +0000 (10:22 +0100)]
client: expose all-file-systems option
Useful to avoid the need for a long (and possibly changing) list of include-dev
options in certain situations, e.g. nested ZFS file systems. The option is
already implemented and seems to work as expected. The checks for virtual
filesystems are not affected by this option.
Fabian Ebner [Thu, 12 Nov 2020 11:38:06 +0000 (12:38 +0100)]
pxar: avoid including archive root's exclude patterns in .pxarexclude-cli
The patterns from the archive root's .pxarexclude file are already present in
self.patterns when encode_pxarexclude_cli is called. Pass along the number of
CLI patterns and slice accordingly.
Suggested-By: Wolfgang Bumiller <w.bumiller@proxmox.com> Signed-off-by: Fabian Ebner <f.ebner@proxmox.com> Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Fabian Ebner [Thu, 12 Nov 2020 09:03:52 +0000 (10:03 +0100)]
pxar: only generate .pxarexclude-cli if there were CLI parameters
previously a .pxarexclude entry in the root of the archive caused the file to
be generated as well, because the patterns are read before calling
generate_directory_file_list and within the function it wasn't possible to
distinguish between a pattern coming from the CLI and a pattern coming from
archive/root/.pxarexclude
Signed-off-by: Fabian Ebner <f.ebner@proxmox.com> Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Fabian Ebner [Thu, 12 Nov 2020 09:03:50 +0000 (10:03 +0100)]
pxar: fix anchored exclusion at archive root
There is no leading slash in an entry's full_path, causing an anchored
exclude at the root level to fail, e.g. having "/name" as the content of the
file archive/root/.pxarexclude didn't match the file archive/root/name
Fix this by prepending a leading slash before matching.
Signed-off-by: Fabian Ebner <f.ebner@proxmox.com> Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
Thomas Lamprecht [Thu, 12 Nov 2020 08:50:48 +0000 (09:50 +0100)]
ui: tell ESLint to be strict in check target
the .lint-incremental target, which is implicitly used by the install
target, is still more forgiving to allow faster "change, build, test"
iteration when developing.
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Mira Limbeck [Tue, 10 Nov 2020 13:22:18 +0000 (14:22 +0100)]
add versions command to proxmox-backup-manager
Add the versions command to proxmox-backup-manager with a similar output
to pveversion [-v]. It prints the packages line by line with only the
package name, followed by the version and, for proxmox-backup and
proxmox-backup-server, some additional information (running kernel,
running version).
In addition it supports the optional output-format parameter which can
be used to print the complete data in either json, json-pretty or text
format. If output-format is specified, the --verbose parameter is
ignored and the detailed list of packages is printed.
With the addition of the versions command, the report is extended as
well.
Signed-off-by: Mira Limbeck <m.limbeck@proxmox.com>
Mira Limbeck [Tue, 10 Nov 2020 13:22:17 +0000 (14:22 +0100)]
add extra_info field to APTUpdateInfo
Add an optional string field to APTUpdateInfo which can be used for
extra information.
This is used for passing running kernel and running version information
in the versions API call together with proxmox-backup and
proxmox-backup-server.
Signed-off-by: Mira Limbeck <m.limbeck@proxmox.com>
Dominik Csapak [Wed, 11 Nov 2020 12:41:07 +0000 (13:41 +0100)]
api2/node/syslog: use 'real_service_name' here also
for now this only does the 'postfix' -> 'postfix@-' conversion,
fixes the issue that we only showed the 'postfix' service syslog
(which is rather empty in a default setup) instead of the instance one
projects / proxmox-backup.git / log