]> git.proxmox.com Git - pve-manager.git/log
pve-manager.git
6 years agoceph dashboard: add missing summary/message access level
Dominik Csapak [Fri, 25 Aug 2017 08:23:28 +0000 (10:23 +0200)]
ceph dashboard: add missing summary/message access level

in luminous, the output of the status/health has changed (again),
so we have to access the correct properties

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoqemu/create: use 'virtio-scsi-single' if 'iothread' is set
Thomas Lamprecht [Mon, 21 Aug 2017 07:20:54 +0000 (09:20 +0200)]
qemu/create: use 'virtio-scsi-single' if 'iothread' is set

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
Reported-by: Nils <gismo3303@gmail.com>
6 years agoqemu/HDEdit: whitespace cleanup
Thomas Lamprecht [Mon, 21 Aug 2017 07:20:53 +0000 (09:20 +0200)]
qemu/HDEdit: whitespace cleanup

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agofix nodeselector when no element in the tree is selected
Dominik Csapak [Fri, 18 Aug 2017 07:04:53 +0000 (09:04 +0200)]
fix nodeselector when no element in the tree is selected

it can happen that nothing in the tree is selected,
then PVE.curSelectedNode is undefined, generating an exception there.
so we check if this is available

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agomake the ceph warning window modal
Dominik Csapak [Wed, 9 Aug 2017 09:42:55 +0000 (11:42 +0200)]
make the ceph warning window modal

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix ceph gui icons
Dominik Csapak [Wed, 9 Aug 2017 09:42:54 +0000 (11:42 +0200)]
fix ceph gui icons

this adds a new css class for fontawesome which we need under certain
circumstances (it is copied from fontawesome minus a font-size:inherited)

and it changes the icons in the osd tree from x-fa to fa
(which is the correct one there)

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoceph: add application parameter to createpool
Dominik Csapak [Wed, 9 Aug 2017 09:42:53 +0000 (11:42 +0200)]
ceph: add application parameter to createpool

and use 'rbd' by default, since we cannot create a cephfs or
rgw from the gui currently

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoceph: make size 3/min 2 default with createpool
Dominik Csapak [Wed, 9 Aug 2017 09:42:52 +0000 (11:42 +0200)]
ceph: make size 3/min 2 default with createpool

in the gui this is already the default, so make it also the default
in the backend (also 2/1 is really bad as a default)

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agobuild: ensure REPOID has a length of 8
Fabian Grünbichler [Wed, 9 Aug 2017 09:42:32 +0000 (11:42 +0200)]
build: ensure REPOID has a length of 8

because unlike the git-rev-parse man page says, the default
length is actually calculated based on the number of packed
objects in the clone of the repository.

we don't want this to depend on the build environment and
how the repository was cloned there, so fix a length of 8
for now.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
6 years agobump version to 5.0-30
Fabian Grünbichler [Tue, 8 Aug 2017 09:08:24 +0000 (11:08 +0200)]
bump version to 5.0-30

6 years agoadd information about bluestore in the gui disklist
Dominik Csapak [Fri, 4 Aug 2017 14:32:49 +0000 (16:32 +0200)]
add information about bluestore in the gui disklist

this adds if an osd uses bluestore in the disklist,
and shows if a disk is used for osd journal/db/wal

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix jslint error
Dominik Csapak [Fri, 4 Aug 2017 14:28:51 +0000 (16:28 +0200)]
fix jslint error

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoshow bluestore info in the osd tree
Dominik Csapak [Fri, 4 Aug 2017 14:28:50 +0000 (16:28 +0200)]
show bluestore info in the osd tree

namely the osd type (filestore/bluestore), bluestore device,
db device and wal device

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadd bluestore information to osd api call
Dominik Csapak [Fri, 4 Aug 2017 14:28:49 +0000 (16:28 +0200)]
add bluestore information to osd api call

this adds information about bluestore (which devices and if
bluestore/filestore) to show in the gui

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agobump version to 5.0-29
Wolfgang Bumiller [Mon, 31 Jul 2017 13:47:17 +0000 (15:47 +0200)]
bump version to 5.0-29

6 years agoceph: ensure mgr starts after pve-cluster
Fabian Grünbichler [Mon, 31 Jul 2017 09:33:16 +0000 (11:33 +0200)]
ceph: ensure mgr starts after pve-cluster

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
6 years agobump version to 5.0-28
Fabian Grünbichler [Mon, 31 Jul 2017 13:31:33 +0000 (15:31 +0200)]
bump version to 5.0-28

6 years agoalso mention manager in create monitor title
Dominik Csapak [Mon, 31 Jul 2017 13:15:33 +0000 (15:15 +0200)]
also mention manager in create monitor title

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoindentation and jslint fix
Dominik Csapak [Mon, 31 Jul 2017 13:15:32 +0000 (15:15 +0200)]
indentation and jslint fix

indentation was wrong on those lines, and js lint complains about
alias not being an array later, so make thoses lines not an array

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadd a bluestore checkbox to the create osd window
Dominik Csapak [Mon, 31 Jul 2017 13:15:31 +0000 (15:15 +0200)]
add a bluestore checkbox to the create osd window

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agouse better icons in the ceph osd tree view
Dominik Csapak [Mon, 31 Jul 2017 13:15:30 +0000 (15:15 +0200)]
use better icons in the ceph osd tree view

this uses the same icons for hosts/osds as in the resource tree,
and also uses the same arrow style

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoreload the pool grid correctly
Dominik Csapak [Mon, 31 Jul 2017 13:15:29 +0000 (15:15 +0200)]
reload the pool grid correctly

whenever a window is closed (creation, deletion) we want to reload the
pool grid, for not having to wait on the next refresh

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoalso get the crush rule name in the ceph pool list
Dominik Csapak [Mon, 31 Jul 2017 13:15:28 +0000 (15:15 +0200)]
also get the crush rule name in the ceph pool list

we get the names in the backend, and give them as an additional field
in the api call, and use it in the grid

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix #1465: use a combobox for the crush rule instead of the id
Dominik Csapak [Mon, 31 Jul 2017 13:15:27 +0000 (15:15 +0200)]
fix #1465: use a combobox for the crush rule instead of the id

this patch does a few things

1. we introduce a new api call /nodes/nodename/ceph/rules
   which gets us a list of crush rules

2. we introduce a new CephRuleSelector which is a simple combobox
   with the data from the api call ceph/rules

3. we use this in the create pool window

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agomake crush rules working again
Dominik Csapak [Mon, 31 Jul 2017 13:15:26 +0000 (15:15 +0200)]
make crush rules working again

since ceph 12.1.1 the (deprecated) parameter 'crush_ruleset' is removed
and replaced with 'crush_rule' while changing this, change from
integer to string so that we can later use the names of the rules
instead of the id

(for now there seems to be a bug that you can only use the name and
not the id)

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadd osd device class to osd tree
Dominik Csapak [Mon, 31 Jul 2017 13:15:25 +0000 (15:15 +0200)]
add osd device class to osd tree

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agocorrectly remove partitions for ceph bluestore osds
Dominik Csapak [Mon, 31 Jul 2017 13:15:24 +0000 (15:15 +0200)]
correctly remove partitions for ceph bluestore osds

we now have to remove 5 types of partitions:

data/metadata
journal
block
block.db
block.wal

this patch fixes the detection of block/block.db/block.wal
generalizes it

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoenable using a device for bluestore block db/wal
Dominik Csapak [Mon, 31 Jul 2017 13:15:23 +0000 (15:15 +0200)]
enable using a device for bluestore block db/wal

we reuse the 'journal_dev' parameter for bluestores block.db
and add a new parameter 'wal_dev' for bluestores write ahead log

if only journal_dev is given, use it for both db and wal

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix creating the first ceph monitor
Dominik Csapak [Thu, 27 Jul 2017 13:02:41 +0000 (15:02 +0200)]
fix creating the first ceph monitor

we cannot use a rados connection before having at least one monitor,
so we have to move it down

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agobump version to 5.0-27
Wolfgang Bumiller [Wed, 26 Jul 2017 14:52:18 +0000 (16:52 +0200)]
bump version to 5.0-27

6 years agoadd missing '-' to service name pattern
Dominik Csapak [Wed, 26 Jul 2017 14:46:16 +0000 (16:46 +0200)]
add missing '-' to service name pattern

since ids can contain '-', we have to include them in the pattern
for the service names

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadd missing changelog entry
Wolfgang Bumiller [Wed, 26 Jul 2017 14:51:06 +0000 (16:51 +0200)]
add missing changelog entry

6 years agobump version to 5.0-26
Wolfgang Bumiller [Wed, 26 Jul 2017 13:40:29 +0000 (15:40 +0200)]
bump version to 5.0-26

6 years agodc/Backup: do not focus node selector on window focus
Thomas Lamprecht [Tue, 25 Jul 2017 14:27:00 +0000 (16:27 +0200)]
dc/Backup: do not focus node selector on window focus

Do not focus anything here in the case he backup window gains focus
again. Else, if we click outside of the window the nodefield gets
always focused again - even if another field was focused previously.
This is a bit confusing, rather have no field focused at all -
especially as a blank value will be often used here.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agoComboGrid: open picker also when focusing an editable field
Thomas Lamprecht [Tue, 25 Jul 2017 14:26:59 +0000 (16:26 +0200)]
ComboGrid: open picker also when focusing an editable field

ComboBoxes provided by UI frameworks normally show their selection
possibilities once focused, ExtJS does this too but not on editable
(and thus filterable) input fields. Add a logic which allows this too
and streamlines the behavior of ComboGrid/Box like components.

To avoid a glitch, where clicking the "picker trigger" (the down
arrow which normally toggles the selection picker) would first focus
the element - thus expanding it - and only then causing a toggle
which collapses it again, we overwrite the 'Ext.form.trigger.Trigger'
onMouseDown so that we can skip a expansion on this event.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agoFileStore: make editable to allow filtering
Thomas Lamprecht [Tue, 25 Jul 2017 14:26:58 +0000 (16:26 +0200)]
FileStore: make editable to allow filtering

Making this ComboGrid based component editable allows implicit
filtering the backing store.
This is quite helpful on file selection, as for example the ISO file
selector from the create wizard is.
Clear the filter (and its displayed value) on store selection change,
as there the filter becomes void - at least this is a simple and easy
solution without glitches.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agoComboGrid: improve setting 'editable' default value
Thomas Lamprecht [Tue, 25 Jul 2017 14:26:57 +0000 (16:26 +0200)]
ComboGrid: improve setting 'editable' default value

This allows child classes to use another default, the initialConfig
didn't allowed that.
This is also the ExtJS like way to set defaults.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agoFileSelector: adapt combogrid total and column width
Thomas Lamprecht [Tue, 25 Jul 2017 14:26:56 +0000 (16:26 +0200)]
FileSelector: adapt combogrid total and column width

By default the "Size" column ellipsed a lot entries, which is just
annoying. Adapt it so that all sizes can be viewed comfortably.

Further increase the total width of this combogrid, a lot less could
be seen there since theme change from the 4.X era.
Use an new empirical found out good  working value.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 years agowhitespace cleanup
Dominik Csapak [Wed, 26 Jul 2017 09:11:53 +0000 (11:11 +0200)]
whitespace cleanup

and intendation fix

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agopveceph: add createmgr/destroymgr commands
Dominik Csapak [Wed, 26 Jul 2017 09:11:52 +0000 (11:11 +0200)]
pveceph: add createmgr/destroymgr commands

this patch adds the create-/destroymgr commands to the api and pveceph,
so that advanced users can split monitor and manager daemons

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agopveceph: create mgr with mon, use nodename for id
Dominik Csapak [Wed, 26 Jul 2017 09:11:51 +0000 (11:11 +0200)]
pveceph: create mgr with mon, use nodename for id

we now want to add a ceph-mgr daemon to every node where a ceph-mon
daemon runs, as per ceph documentation recommendation, because in
luminous the mgr daemons will not be automatically created/started
with a monitor anymore

we also give the createmon an optional id parameter, so that one
can set a custom id, and make the creation/removal of the manager
optional but the default

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoImprove a bit the esthetics of the User Settings window
Emmanuel Kasper [Thu, 6 Jul 2017 12:01:31 +0000 (14:01 +0200)]
Improve a bit the esthetics of the User Settings window

 * group the action buttons on the left, to distinguish them from the modal
 'close' button
 * display the action buttons as grey instead of blue, to be consistent
 with the rest of the GUI

6 years agoRename the pve-help-button to pve-inline-button
Emmanuel Kasper [Thu, 6 Jul 2017 12:01:30 +0000 (14:01 +0200)]
Rename the pve-help-button to pve-inline-button

This will allow reuse of the class for other inlined buttons

6 years agoClose #1268: Show migration start server and migration end server
Emmanuel Kasper [Thu, 6 Jul 2017 12:01:29 +0000 (14:01 +0200)]
Close #1268: Show migration start server and migration end server

6 years agoDo not display the abbreviated commit SHA-1 in the workspace version string
Emmanuel Kasper [Thu, 6 Jul 2017 12:01:28 +0000 (14:01 +0200)]
Do not display the abbreviated commit SHA-1 in the workspace version string

This string is ugly and unfit for human consumption for 99% of our users.
The same information is available in a number of place in the GUI via Node
Summary, Package Versions, Subscription (Report)

6 years agobump version to 5.0-25
Wolfgang Bumiller [Mon, 24 Jul 2017 14:17:16 +0000 (16:17 +0200)]
bump version to 5.0-25

6 years agoadapt the health message list to luminous
Dominik Csapak [Thu, 20 Jul 2017 14:16:14 +0000 (16:16 +0200)]
adapt the health message list to luminous

they restructured the json structure, so we have to adapt to it.
additionally, we now have a 'detail' property for the messages
which we show on a click of the 'info' button in the new actioncolumn

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadapt ceph status detail to luminous and make it more robust
Dominik Csapak [Thu, 20 Jul 2017 14:16:13 +0000 (16:16 +0200)]
adapt ceph status detail to luminous and make it more robust

they restructured the json interface
(e.g. they removed the timechecks from ceph status)

so we have to generate those differently

also make the whole thing more robust to changes,
as in do not bail out if one thing is missing, but check the needed
data only when we need it and omit that part

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agorename 'monitors' to 'states' in pg list
Dominik Csapak [Thu, 20 Jul 2017 14:16:12 +0000 (16:16 +0200)]
rename 'monitors' to 'states' in pg list

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadapt render_ceph_health for luminous
Dominik Csapak [Thu, 20 Jul 2017 14:16:11 +0000 (16:16 +0200)]
adapt render_ceph_health for luminous

it is using 'status' instead of 'overall_health' now

also only give the health object now

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoadd health detail to ceph status api call
Dominik Csapak [Thu, 20 Jul 2017 14:16:10 +0000 (16:16 +0200)]
add health detail to ceph status api call

with ceph 12.1.1 luminous, ceph has reorganized its json
interface of the ceph status and ceph health call

so to get everything we need, we have to also get
the ceph health detail information into our
status call

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoceph: end ceph.list with a newline
Wolfgang Bumiller [Mon, 3 Jul 2017 09:10:01 +0000 (11:10 +0200)]
ceph: end ceph.list with a newline

6 years agobump version to 5.0-24
Wolfgang Bumiller [Wed, 19 Jul 2017 10:49:31 +0000 (12:49 +0200)]
bump version to 5.0-24

6 years agoUse emptyText to inform the user of the value that will set be when the field is...
Emmanuel Kasper [Mon, 26 Jun 2017 12:18:05 +0000 (14:18 +0200)]
Use emptyText to inform the user of the value that will set be when the field is empty

emptyText is not submitted, because it does not count as value when getSubmitValue() is
called on it in PVE.panel.InputPanel.getValues()

6 years agoFix pveceph createosd
Wolfgang Link [Tue, 18 Jul 2017 09:07:12 +0000 (11:07 +0200)]
Fix pveceph createosd

The --filestore flag is now required see doc of ceph.

If the --bluestore argument is given, a bluestore objectstore will be
created.  If --filestore is provided, a legacy FileStore objectstore
will be created.  If neither is specified, we default to BlueStore.

6 years agoRevert "trust StartCom_Certification_Authority_2.crt"
Wolfgang Bumiller [Wed, 19 Jul 2017 07:27:30 +0000 (09:27 +0200)]
Revert "trust StartCom_Certification_Authority_2.crt"

This reverts commit 12fe9183cb82c3ea148ac31990c67b518c50aabf.

Revert "add missing file"

This reverts commit c11885e0a0da0ad0c944a48e645d829553c6705d.

We've switched to Let's Encrypt.

postinst configure: run update-ca-certificates if the
  previous version is <= 5.0-23.

6 years agobuildsys: use CC, CPPFLAGS and CFLAGS
Wolfgang Bumiller [Wed, 19 Jul 2017 06:49:58 +0000 (08:49 +0200)]
buildsys: use CC, CPPFLAGS and CFLAGS

pvemailforward is a tiny oneliner, but for the sake of best
practices, let's use the build tools from the environment.
For example, with dpkg-buildpackage this will make us use
-D_FORTIFY_SOURCE=2 etc.

6 years agoFix #1450 : restore setgid bit on pvemailforward binary
Emmanuel Kasper [Tue, 18 Jul 2017 15:37:27 +0000 (17:37 +0200)]
Fix #1450 : restore setgid bit on pvemailforward binary

When calling chown on a setuid/setgid executable, the
setuid/setgid bits are reset to 0. So ordering matters
here. See chown(2).

6 years agocheck if storage is in the selector
Dominik Csapak [Tue, 11 Jul 2017 12:39:20 +0000 (14:39 +0200)]
check if storage is in the selector

when having an lxc mountpoint/rootfs on a storage which does not
allow the 'rootfs' content (e.g. somebody disabled it after creating
a container there), trying to edit the mp/rootfs broke the gui

now we check if the record exists

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix breaking gui when trying to edit a bindmount rootfs
Dominik Csapak [Tue, 11 Jul 2017 12:39:19 +0000 (14:39 +0200)]
fix breaking gui when trying to edit a bindmount rootfs

when having a simple directory as rootfs,
trying to edit it in the gui broke it, because
we tried to disable the backup checkbox which did not exists

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agofix wrong osd calculation in ceph dashboard
Dominik Csapak [Tue, 11 Jul 2017 12:39:18 +0000 (14:39 +0200)]
fix wrong osd calculation in ceph dashboard

in luminous, the error message is not
'x/y in osds are down' anymore, but
'x osds down'

so we need to adapt the parsing, and it means we cannot check
the number of in osds there anymore (was never really needed, so
we can simply omit it)

when an osd is down but marked as out, those errors disappear

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 years agoSelect OS Type after selecting the Installation Media
Emmanuel Kasper [Mon, 10 Jul 2017 14:35:04 +0000 (16:35 +0200)]
Select OS Type after selecting the Installation Media

This prepares for the future, in case we want to infer the
OS Type from the installation media name or content.

6 years agoUse a View Controller for making components react to each other
Emmanuel Kasper [Mon, 10 Jul 2017 14:35:03 +0000 (16:35 +0200)]
Use a View Controller for making components react to each other

This allow to declare all the components inside the same items
array, and separate behaviour from view better.

No functional changes.

6 years agoUse camel case for InputPanels aliases
Emmanuel Kasper [Mon, 10 Jul 2017 14:35:02 +0000 (16:35 +0200)]
Use camel case for InputPanels aliases

Camel Case names for component alias follow Ext recommended practises
and are used otherwise everyelse in the code base.

No functionnal changes, aliases for these components were not used
anyway.

6 years agoadd missing file
Dietmar Maurer [Tue, 4 Jul 2017 09:57:07 +0000 (11:57 +0200)]
add missing file

6 years agobump version to 5.0-23
Dietmar Maurer [Tue, 4 Jul 2017 09:51:00 +0000 (11:51 +0200)]
bump version to 5.0-23

6 years agotrust StartCom_Certification_Authority_2.crt
Dietmar Maurer [Tue, 4 Jul 2017 09:48:52 +0000 (11:48 +0200)]
trust StartCom_Certification_Authority_2.crt

6 years agoReplication.js: translate 'Error'
Dietmar Maurer [Tue, 4 Jul 2017 07:06:25 +0000 (09:06 +0200)]
Replication.js: translate 'Error'

6 years agoavoid extra gettext string
Dietmar Maurer [Tue, 4 Jul 2017 07:02:24 +0000 (09:02 +0200)]
avoid extra gettext string

6 years agobump version to 5.0-22
Dietmar Maurer [Tue, 4 Jul 2017 06:18:55 +0000 (08:18 +0200)]
bump version to 5.0-22

6 years agoupdate aplinfo.dat
Dietmar Maurer [Tue, 4 Jul 2017 06:17:45 +0000 (08:17 +0200)]
update aplinfo.dat

6 years agobump version to 5.0-21
Dietmar Maurer [Mon, 3 Jul 2017 14:09:03 +0000 (16:09 +0200)]
bump version to 5.0-21

6 years agoupdate aplinfo.dat
Dietmar Maurer [Mon, 3 Jul 2017 14:06:49 +0000 (16:06 +0200)]
update aplinfo.dat

7 years agobump version to 5.0-20
Dietmar Maurer [Fri, 30 Jun 2017 09:19:10 +0000 (11:19 +0200)]
bump version to 5.0-20

7 years agoreplication: reuse fitting translations
Thomas Lamprecht [Thu, 29 Jun 2017 14:37:32 +0000 (16:37 +0200)]
replication: reuse fitting translations

while it does not makes sense do over-reuse translations for the sake
of translating less, imo, here we can safely reuse already existing
ones and pull out the unit 'MB/s' from the gettext.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agopvesr: allow schedule now from CLI
Thomas Lamprecht [Thu, 29 Jun 2017 10:08:59 +0000 (12:08 +0200)]
pvesr: allow schedule now from CLI

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agobump version to 5.0-19
Dietmar Maurer [Thu, 29 Jun 2017 10:02:14 +0000 (12:02 +0200)]
bump version to 5.0-19

7 years agopvesr status: handle disabled jobs
Thomas Lamprecht [Thu, 29 Jun 2017 09:35:02 +0000 (11:35 +0200)]
pvesr status: handle disabled jobs

As the nodes replication status call also returns disabled jobs now,
we need to handle them here too.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agoreplication; merge status and status text and add to dc mode
Thomas Lamprecht [Thu, 29 Jun 2017 09:22:53 +0000 (11:22 +0200)]
replication; merge status and status text and add to dc mode

It was a bit strange to have two separate status columns, which do
effectively the same thing. So merge them together to save a bit
space and have less columns.

We do not need to translate 'Status Text' as a nice side effect.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agoreplication: show disabled jobs and allow enabling/disabling
Thomas Lamprecht [Thu, 29 Jun 2017 08:08:29 +0000 (10:08 +0200)]
replication: show disabled jobs and allow enabling/disabling

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agoPVE/APLInfo.pm - use correct version for user agent
Dietmar Maurer [Thu, 29 Jun 2017 06:07:11 +0000 (08:07 +0200)]
PVE/APLInfo.pm - use correct version for user agent

7 years agoRevert "add beta text with link to bugtracker"
Dietmar Maurer [Thu, 29 Jun 2017 05:58:37 +0000 (07:58 +0200)]
Revert "add beta text with link to bugtracker"

This reverts commit 6531861e690bbe2c077dd0388689e565cf28df5a.

Prepare for final 5.0 release.

7 years agoreplication: fix regression tests
Dietmar Maurer [Thu, 29 Jun 2017 05:46:56 +0000 (07:46 +0200)]
replication: fix regression tests

7 years agoPVE/API2/Replication.pm: cleanup - avoid modifying jobcfg
Dietmar Maurer [Thu, 29 Jun 2017 05:45:57 +0000 (07:45 +0200)]
PVE/API2/Replication.pm: cleanup - avoid modifying jobcfg

This is simply no longer necessary.

7 years agoreplication: adapt help buttons reference
Thomas Lamprecht [Wed, 28 Jun 2017 16:11:49 +0000 (18:11 +0200)]
replication: adapt help buttons reference

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agoAdd Help Button for Backup Modal Windows
Emmanuel Kasper [Wed, 28 Jun 2017 13:26:03 +0000 (15:26 +0200)]
Add Help Button for Backup Modal Windows

Modal Windows disable the Workspace behind them, so it was not possible
to get onlineHelp when starting a oneshot backup or configuring a backup job.

7 years agobump version to 5.0-18
Wolfgang Bumiller [Wed, 28 Jun 2017 12:24:19 +0000 (14:24 +0200)]
bump version to 5.0-18

7 years agoFix create replication job
Wolfgang Link [Wed, 28 Jun 2017 11:50:29 +0000 (13:50 +0200)]
Fix create replication job

Add node to load config. this is needed if you are on an other node then the guest.

7 years agobump version to 5.0-17
Dietmar Maurer [Wed, 28 Jun 2017 11:34:42 +0000 (13:34 +0200)]
bump version to 5.0-17

7 years agoreplication GUI: add schedule now Button
Dietmar Maurer [Wed, 28 Jun 2017 06:35:02 +0000 (08:35 +0200)]
replication GUI: add schedule now Button

7 years agoPVE/API2/Replication.pm: add schedule_now API
Dietmar Maurer [Wed, 28 Jun 2017 04:55:16 +0000 (06:55 +0200)]
PVE/API2/Replication.pm: add schedule_now API

7 years agogui: drop cirrus from the GUI
Wolfgang Bumiller [Wed, 28 Jun 2017 09:39:33 +0000 (11:39 +0200)]
gui: drop cirrus from the GUI

now only available from the CLI

7 years agoreplica: improve schedule format example
Thomas Lamprecht [Wed, 28 Jun 2017 08:24:39 +0000 (10:24 +0200)]
replica: improve schedule format example

Add one more to show that also weekdays + intervals are possible

Further improve wording + reduce translation needs for others entries.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agogrid/replication: show textual description of default value
Thomas Lamprecht [Wed, 28 Jun 2017 07:19:49 +0000 (09:19 +0200)]
grid/replication: show textual description of default value

When a user sees this the first time and wants to add a job he
shouldn't be confused what the default value means, so display this
through the emptyText property, which does not get submitted to the
backend.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 years agoMake sure checkboxes always appear under the ACL combobox for a bit of esthetic
Emmanuel Kasper [Tue, 27 Jun 2017 15:07:09 +0000 (17:07 +0200)]
Make sure checkboxes always appear under the ACL combobox for a bit of esthetic

7 years agoAdd a checkbox 'Skip replication' to filter out container mount points to replicate
Emmanuel Kasper [Tue, 27 Jun 2017 15:07:08 +0000 (17:07 +0200)]
Add a checkbox 'Skip replication' to filter out container mount points to replicate

7 years agotests: mock PVE::Cluster::get_members and adapt fail test
Wolfgang Bumiller [Tue, 27 Jun 2017 13:47:24 +0000 (15:47 +0200)]
tests: mock PVE::Cluster::get_members and adapt fail test

Commit 3385399339c94 ("replication: keep retrying every 30 minutes in
error state") changed the retry behavior to not stop after the 3rd error
and then stick to half-hour intervals. This needs to be reflected in the
tests. The numbers here match. (1900 + 30*60 = 3700).

Commit fd844180a7efa ("replication: don't sync to offline targets on
error states) changed the retry behavior to check whether the target
node is online. If this is not the case we fail right away. This
introduced a dependency on PVE::Cluster::get_members which we now need
to mock. Tests currently use node names "node{1,2,3}", so I just mock
those 3.

7 years agoceph: keys: add mgr allow * to client.admin
Fabian Grünbichler [Tue, 27 Jun 2017 14:42:16 +0000 (16:42 +0200)]
ceph: keys: add mgr allow * to client.admin

needed for data that now gets queried via the mgr instead of
via the mon.

7 years agopveceph: createosd: allow to create bluestore OSDs
Fabian Grünbichler [Tue, 27 Jun 2017 14:12:13 +0000 (16:12 +0200)]
pveceph: createosd: allow to create bluestore OSDs

7 years agopveceph: init: add min_size, change default to 3/2
Fabian Grünbichler [Mon, 22 May 2017 07:54:30 +0000 (09:54 +0200)]
pveceph: init: add min_size, change default to 3/2

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>