pve-manager.git
22 hours agoui: qemu/HWView: cloudinit: use the font-awesome provided cloud symbol master
Thomas Lamprecht [Wed, 17 Jul 2019 11:28:20 +0000 (13:28 +0200)]
ui: qemu/HWView: cloudinit: use the font-awesome provided cloud symbol

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
22 hours agoui: qemu/HWView: small code cleanup for remove dialog
Thomas Lamprecht [Wed, 17 Jul 2019 11:27:35 +0000 (13:27 +0200)]
ui: qemu/HWView: small code cleanup for remove dialog

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
22 hours agoui: qemu/HWView: only render icon when in grid
Thomas Lamprecht [Wed, 17 Jul 2019 11:26:59 +0000 (13:26 +0200)]
ui: qemu/HWView: only render icon when in grid

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
23 hours agogui: fix debug-mode error
Dominik Csapak [Mon, 15 Jul 2019 12:03:45 +0000 (14:03 +0200)]
gui: fix debug-mode error

setData on an 'empty' store does not work in extjs debug mode
setting store to '{}' here does create a store on component creation
so the store is not 'empty'

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
23 hours agoceph destroymon: actually die on errors
Fabian Grünbichler [Tue, 16 Jul 2019 13:51:04 +0000 (15:51 +0200)]
ceph destroymon: actually die on errors

instead of silently ignoring them. since we are in a task worker here
this is especially important - otherwise the task status/result is also
wrong!

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
23 hours agoceph init: actually die on errors
Fabian Grünbichler [Tue, 16 Jul 2019 13:51:03 +0000 (15:51 +0200)]
ceph init: actually die on errors

instead of silently ignoring them

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 days agoui: window/migrate: use space separator between type and VMID
Thomas Lamprecht [Mon, 15 Jul 2019 18:14:37 +0000 (20:14 +0200)]
ui: window/migrate: use space separator between type and VMID

Like previous to commit 29fd539de02089196c9e61257aaff362d3c85c41
which broke this, we have some assumptions in our gui-test/screenshot
automation which lays assumptions on certain window titles, so try to
keep them stable.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 days agobump version to 6.0-4
Thomas Lamprecht [Mon, 15 Jul 2019 16:48:57 +0000 (18:48 +0200)]
bump version to 6.0-4

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 days agopve5to6: check for ceph-volume osds
Dominik Csapak [Mon, 15 Jul 2019 08:20:27 +0000 (10:20 +0200)]
pve5to6: check for ceph-volume osds

do not warn in case there exist nautilus osds
since the upgrade has to be done by the time a user should add new
osds, it does not make sense to warn in that case

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
3 days agogui: ceph installer: fix typo
Tim Marx [Mon, 15 Jul 2019 09:06:13 +0000 (11:06 +0200)]
gui: ceph installer: fix typo

Signed-off-by: Tim Marx <t.marx@proxmox.com>
3 days agofollowup code cleanup
Thomas Lamprecht [Mon, 15 Jul 2019 08:56:01 +0000 (10:56 +0200)]
followup code cleanup

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
3 days agoceph: mon list: show only as running when monitor is quorate
Dominik Csapak [Mon, 15 Jul 2019 08:20:28 +0000 (10:20 +0200)]
ceph: mon list: show only as running when monitor is quorate

nautilus puts non running monitors also in the monmap, so only show
as running when it has quorum

this is also not 100% correct, but the only 'correct' alternative is
to try and get/parse the systemd status of the units and broadcast it
to the pmxcfs

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
3 days agoui: cephfs: also add new tooltip like pool create
Thomas Lamprecht [Mon, 15 Jul 2019 05:47:31 +0000 (07:47 +0200)]
ui: cephfs: also add new tooltip like pool create

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
3 days agoui: dc/Cluster: rename ring columns to link
Thomas Lamprecht [Mon, 15 Jul 2019 05:43:04 +0000 (07:43 +0200)]
ui: dc/Cluster: rename ring columns to link

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
5 days agofollowup coide cleanup
Thomas Lamprecht [Fri, 12 Jul 2019 12:07:25 +0000 (14:07 +0200)]
followup coide cleanup

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
5 days agoui: improve console fallback logic if default is not available
Dominik Csapak [Fri, 12 Jul 2019 11:33:07 +0000 (13:33 +0200)]
ui: improve console fallback logic if default is not available

instead of instantly defaulting to novnc when the chosen
default is not available, try the other (better variant)
before falling back to novnc

e.g. having no default set is equivalent to having it set
to xtermjs, and having a vm with spice then previously
defaulted to novnc, now to spice (the reverse is also true)

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
5 days agoui: fix wrong defaultConsole fallback in tree
Dominik Csapak [Fri, 12 Jul 2019 11:33:06 +0000 (13:33 +0200)]
ui: fix wrong defaultConsole fallback in tree

having spice does not mean it can also have xtermjs, and
we get the information in the status call already
so give the correct parameters to openDefaultConsoleWindow

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
5 days agobump version to 6.0-3
Thomas Lamprecht [Fri, 12 Jul 2019 10:31:04 +0000 (12:31 +0200)]
bump version to 6.0-3

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
5 days agofilter unusable lxc mountoptions for rootfs
Oguz Bektas [Thu, 11 Jul 2019 13:08:30 +0000 (15:08 +0200)]
filter unusable lxc mountoptions for rootfs

disable nodev and noexec per thomas' suggestion[0].

[0]: https://pve.proxmox.com/pipermail/pve-devel/2019-July/037994.html

Signed-off-by: Oguz Bektas <o.bektas@proxmox.com>
6 days agoRevert "add beta text with link to bugtracker"
Thomas Lamprecht [Thu, 11 Jul 2019 19:59:05 +0000 (21:59 +0200)]
Revert "add beta text with link to bugtracker"

This reverts commit 6f4fa3d9419ae3708eb7244f4fcb6e7711b026b9.

6 days agoui: node/disks: retune column width and hide vendor by defauöt
Thomas Lamprecht [Thu, 11 Jul 2019 19:37:10 +0000 (21:37 +0200)]
ui: node/disks: retune column width and hide vendor by defauöt

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: node/disks: reuse Utils.format_boolean for GPT column
Thomas Lamprecht [Thu, 11 Jul 2019 19:33:37 +0000 (21:33 +0200)]
ui: node/disks: reuse Utils.format_boolean for GPT column

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agopveam: add mail to sections
Thomas Lamprecht [Thu, 11 Jul 2019 19:22:55 +0000 (21:22 +0200)]
pveam: add mail to sections

it includes the PMG

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agobump version to 6.0-2
Thomas Lamprecht [Thu, 11 Jul 2019 18:42:02 +0000 (20:42 +0200)]
bump version to 6.0-2

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: dc/cluster: enforce second link if in join info
Thomas Lamprecht [Thu, 11 Jul 2019 18:41:09 +0000 (20:41 +0200)]
ui: dc/cluster: enforce second link if in join info

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: dc/cluster: join: s/ring/link/
Thomas Lamprecht [Thu, 11 Jul 2019 18:40:51 +0000 (20:40 +0200)]
ui: dc/cluster: join: s/ring/link/

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: dc/cluster: allow to create cluster with second link
Thomas Lamprecht [Thu, 11 Jul 2019 18:16:20 +0000 (20:16 +0200)]
ui: dc/cluster: allow to create cluster with second link

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: dc/cluster: add onlineHelp reference to create call
Thomas Lamprecht [Thu, 11 Jul 2019 18:12:00 +0000 (20:12 +0200)]
ui: dc/cluster: add onlineHelp reference to create call

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agogui: ceph configdb: fix store load
Dominik Csapak [Thu, 11 Jul 2019 12:52:38 +0000 (14:52 +0200)]
gui: ceph configdb: fix store load

instead of using API2Request, use the proper grid store load

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 days agoui: ceph/pool: enable add_storages by default and add tooltip
Thomas Lamprecht [Thu, 11 Jul 2019 12:55:26 +0000 (14:55 +0200)]
ui: ceph/pool: enable add_storages by default and add tooltip

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/service list: re-do column widths with flex
Thomas Lamprecht [Thu, 11 Jul 2019 12:47:01 +0000 (14:47 +0200)]
ui: ceph/service list: re-do column widths with flex

this way we see thew full status of a MDS, and for our minimal
resolution of 720p and targeted resolution of 1080p it works good.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/pool: rework status grid a bit
Thomas Lamprecht [Thu, 11 Jul 2019 12:10:54 +0000 (14:10 +0200)]
ui: ceph/pool: rework status grid a bit

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: remove notify calls on viewmodel
Thomas Lamprecht [Thu, 11 Jul 2019 10:50:03 +0000 (12:50 +0200)]
ui: remove notify calls on viewmodel

They should not be required and result in exceptions thrown if this
component is used in combination with ExtJS debug sources

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: merge apply + commit latency columns
Thomas Lamprecht [Thu, 11 Jul 2019 09:16:22 +0000 (11:16 +0200)]
ui: ceph/osd: merge apply + commit latency columns

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: split out used% and total avail column
Thomas Lamprecht [Thu, 11 Jul 2019 09:15:45 +0000 (11:15 +0200)]
ui: ceph/osd: split out used% and total avail column

as "used -> total" is just confusing

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: hide Type column by default
Thomas Lamprecht [Thu, 11 Jul 2019 08:48:13 +0000 (10:48 +0200)]
ui: ceph/osd: hide Type column by default

this column shows what type the current row is (host, osd, root),
nowadays that's quite visible through the tree view + icons, so just
hide it by default to make this interface a bit less noisier.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: only show at max 5 digits after decimal point for weight
Thomas Lamprecht [Thu, 11 Jul 2019 08:37:53 +0000 (10:37 +0200)]
ui: ceph/osd: only show at max 5 digits after decimal point for weight

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: rename Action to More
Thomas Lamprecht [Thu, 11 Jul 2019 07:40:09 +0000 (09:40 +0200)]
ui: ceph/osd: rename Action to More

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph OSD: tweak default column width a bit
Thomas Lamprecht [Wed, 10 Jul 2019 20:44:17 +0000 (22:44 +0200)]
ui: ceph OSD: tweak default column width a bit

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoui: ceph/osd: reintroduce Used column
Thomas Lamprecht [Thu, 11 Jul 2019 07:45:26 +0000 (09:45 +0200)]
ui: ceph/osd: reintroduce Used column

previous commit 76235ae132c69898048c1f7579e7da40adbd7a6e accidentally
removed this column during refactoring, let's reintroduce it again.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agogui: ceph osd: rewrite osd panel and add scrub button
Dominik Csapak [Wed, 10 Jul 2019 13:13:52 +0000 (15:13 +0200)]
gui: ceph osd: rewrite osd panel and add scrub button

this rewrite of the osd panel converts it to a static declaration of
the layout with a controller/viewmodel logic

while doing this, add icons to buttons, and add a scrub/deep scrub entry

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
6 days agoapi/ceph: code cleanup
Thomas Lamprecht [Thu, 11 Jul 2019 12:14:27 +0000 (14:14 +0200)]
api/ceph: code cleanup

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
6 days agoapi/ceph: add osd scrub api call
Dominik Csapak [Wed, 10 Jul 2019 13:13:51 +0000 (15:13 +0200)]
api/ceph: add osd scrub api call

can be called to (deep) scrub a specific osd

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
7 days agokeyring: use ckeyring_path variable in chown cmd
Alwin Antreich [Wed, 10 Jul 2019 13:35:09 +0000 (15:35 +0200)]
keyring: use ckeyring_path variable in chown cmd

A fixed path for the ceph.client.admin.keyring was used in the chown
command. This patch uses the ckeyring_path variable instead, to minimize
changes should the path change.

Signed-off-by: Alwin Antreich <a.antreich@proxmox.com>
7 days agoFix: typo in ckeyring_path
Alwin Antreich [Wed, 10 Jul 2019 13:35:08 +0000 (15:35 +0200)]
Fix: typo in ckeyring_path

pveceph init failed, as the copy used a wrong path with typo as
target and then the chown tried to operate on the correct keyring
path, which was then non-existing.

Signed-off-by: Alwin Antreich <a.antreich@proxmox.com>
7 days agoui: dc/options: minor followup cleanups
Thomas Lamprecht [Wed, 10 Jul 2019 12:58:19 +0000 (14:58 +0200)]
ui: dc/options: minor followup cleanups

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 days agoui/Parser: printPropertyString: sort non-defaultKey parts for stability
Thomas Lamprecht [Wed, 10 Jul 2019 12:57:38 +0000 (14:57 +0200)]
ui/Parser: printPropertyString: sort non-defaultKey parts for stability

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
7 days agoui: dc/options: use proxmoxNetworkSelector for network
Thomas Lamprecht [Wed, 10 Jul 2019 12:38:49 +0000 (14:38 +0200)]
ui: dc/options: use proxmoxNetworkSelector for network

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
8 days agopveproxy: allow to request debug sources directly over GET parameter
Thomas Lamprecht [Tue, 9 Jul 2019 13:06:34 +0000 (15:06 +0200)]
pveproxy: allow to request debug sources directly over GET parameter

Avoids the need to stop the current pveproxy and restart it in debug
mode, if one wants to get the debug un-minified sources for ExtJS,
for example.

For a local PVE instance the following URL would ensure you load the
debug sources:
https://localhost:8006/?debug#v1:0:18:4:25:19::8:20:28

For convenience allow to omit the parameters value, and default to
true, iow, ?debug and ?debug=1 are the same, one could force it off -
even if the server was started in debug mode - by using ?debug=0

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
9 days agofollowup: code cleanup to arrow functions
Thomas Lamprecht [Mon, 8 Jul 2019 17:26:13 +0000 (19:26 +0200)]
followup: code cleanup to arrow functions

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
9 days agogui: show 0 for max_relocate/restart correctly
Dominik Csapak [Mon, 8 Jul 2019 12:12:37 +0000 (14:12 +0200)]
gui: show 0 for max_relocate/restart correctly

0 || '1' will always return '1'

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
9 days agofix #2267: delete address(6) and netmas(6) with cidr(6)
Dominik Csapak [Mon, 8 Jul 2019 14:29:18 +0000 (16:29 +0200)]
fix #2267: delete address(6) and netmas(6) with cidr(6)

otherwise a user cannot delete an ip from an interface

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
10 days agobuild: bump versioned dependency on pve-cluster
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:10 +0000 (14:44 +0200)]
build: bump versioned dependency on pve-cluster

needed for PVE::Corosync::resolve_hostname_like_corosync

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days ago5to6: make corosync totem checks more verbose
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:09 +0000 (14:44 +0200)]
5to6: make corosync totem checks more verbose

to avoid just printing the subheader with no results

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days ago5to6: add more corosync subheaders
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:08 +0000 (14:44 +0200)]
5to6: add more corosync subheaders

to improve readability

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days ago5to6: fail if a corosync node has neither ring0 nor ring1 defined
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:07 +0000 (14:44 +0200)]
5to6: fail if a corosync node has neither ring0 nor ring1 defined

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days ago5to6: reword/-structure corosync message
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:06 +0000 (14:44 +0200)]
5to6: reword/-structure corosync message

and fix a typo as well

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days ago5to6: attempt to resolve corosync rings
Fabian Grünbichler [Fri, 5 Jul 2019 12:44:05 +0000 (14:44 +0200)]
5to6: attempt to resolve corosync rings

and only fail if unable to

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
10 days agoupdate shipped aplinfo index
Thomas Lamprecht [Mon, 8 Jul 2019 05:52:34 +0000 (07:52 +0200)]
update shipped aplinfo index

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
12 days agofollowup: balance columns
Thomas Lamprecht [Fri, 5 Jul 2019 16:55:05 +0000 (18:55 +0200)]
followup: balance columns

having this new selector on the left column makes it a better fit for
both possible edit situations, rootfs is now fully balanced and both
mp/rootfs kip the "skip replication" setting on the same place - in
sync with VMs

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
12 days agofix #1451: allow some extra mount options for container
Oguz Bektas [Fri, 5 Jul 2019 11:27:06 +0000 (13:27 +0200)]
fix #1451: allow some extra mount options for container

This allows the following non-problematic mount options for lxc
container mountpoints:
* noexec - Do not permit execution of binaries on the mounted FS
* noatime - Do not update inode access times on this filesystem
* nosuid - Do not allow suid or sgid bits to take effect
* nodev - Do not interpret character or block devices on the FS

Signed-off-by: Oguz Bektas <o.bektas@proxmox.com>
12 days agofollowup: hide mask, level and RO by default
Thomas Lamprecht [Fri, 5 Jul 2019 10:30:28 +0000 (12:30 +0200)]
followup: hide mask, level and RO by default

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days agogui: ceph: show config database under the config
Dominik Csapak [Fri, 5 Jul 2019 07:16:34 +0000 (09:16 +0200)]
gui: ceph: show config database under the config

the column titles are the same as the 'ceph config dump' cli

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
13 days agoceph: add api call for config database
Dominik Csapak [Fri, 5 Jul 2019 07:16:33 +0000 (09:16 +0200)]
ceph: add api call for config database

simply returns the database content

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
13 days agouse new url
Thomas Lamprecht [Thu, 4 Jul 2019 15:57:27 +0000 (17:57 +0200)]
use new url

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days agobump version to 6.0-1
Thomas Lamprecht [Thu, 4 Jul 2019 14:54:49 +0000 (16:54 +0200)]
bump version to 6.0-1

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days agoceph: remove obsolete ceph.service and do not distribute it anymore
Dominik Csapak [Thu, 27 Jun 2019 12:51:36 +0000 (14:51 +0200)]
ceph: remove obsolete ceph.service and do not distribute it anymore

in nautilus there is no ceph-disk anymore and osd activation
does not use udev anymore so this service is not needed anymore

remove it and do not copy it when installing a new ceph cluster

in pve-storage.target we replace ceph.service with ceph.target

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
13 days agopveceph install: set repo URL to download.proxmox.com nautilus repo
Thomas Lamprecht [Thu, 4 Jul 2019 13:52:51 +0000 (15:52 +0200)]
pveceph install: set repo URL to download.proxmox.com nautilus repo

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days agogui: ceph: parse the whole message for monitor ids
Dominik Csapak [Thu, 4 Jul 2019 13:36:13 +0000 (15:36 +0200)]
gui: ceph: parse the whole message for monitor ids

sometimes a monitor id is only mentioned in a later part of the message
not always the first

for example, if more than one monitor is missing the msgr2 protocol
each monitor gets its own line

before this patch the whole error would only be attributed to the first
monitor mentioned, now it is attributed to all

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
13 days ago5to6: only tell to solve problems if fails are present
Thomas Lamprecht [Thu, 4 Jul 2019 13:32:35 +0000 (15:32 +0200)]
5to6: only tell to solve problems if fails are present

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: improve final note on errors and/or warning a bit
Thomas Lamprecht [Thu, 4 Jul 2019 13:30:54 +0000 (15:30 +0200)]
5to6: improve final note on errors and/or warning a bit

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: fiy typo: s/detailled/detailed/
Thomas Lamprecht [Thu, 4 Jul 2019 13:23:19 +0000 (15:23 +0200)]
5to6: fiy typo: s/detailled/detailed/

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: check common services pveproxy pvedaemon pvestatd
Thomas Lamprecht [Thu, 4 Jul 2019 13:20:12 +0000 (15:20 +0200)]
5to6: check common services pveproxy pvedaemon pvestatd

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: add log_systemd_unit_state and use for pve-cluster and corosync
Thomas Lamprecht [Thu, 4 Jul 2019 13:10:16 +0000 (15:10 +0200)]
5to6: add log_systemd_unit_state and use for pve-cluster and corosync

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: import run_command
Thomas Lamprecht [Thu, 4 Jul 2019 12:56:50 +0000 (14:56 +0200)]
5to6: import run_command

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: clarify "cluster quorate" message a bit
Thomas Lamprecht [Thu, 4 Jul 2019 12:21:08 +0000 (14:21 +0200)]
5to6: clarify "cluster quorate" message a bit

As this is the state from the pmxcfs

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: fixup: really set $total_votes to 0 if not defined
Thomas Lamprecht [Thu, 4 Jul 2019 12:20:41 +0000 (14:20 +0200)]
5to6: fixup: really set $total_votes to 0 if not defined

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
13 days ago5to6: drop detailed quorumtool output
Fabian Grünbichler [Thu, 4 Jul 2019 11:52:24 +0000 (13:52 +0200)]
5to6: drop detailed quorumtool output

since we already print most of that anyway, and it is rather long.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
13 days ago5to6: more quorum / vote checking
Fabian Grünbichler [Thu, 4 Jul 2019 11:52:23 +0000 (13:52 +0200)]
5to6: more quorum / vote checking

handle expected votes set to non-standard value, and try to adjust
calculations for qdevice setups.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: quote some Ceph flags/options
Fabian Grünbichler [Thu, 4 Jul 2019 08:55:49 +0000 (10:55 +0200)]
5to6: quote some Ceph flags/options

to improve readability

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: check for ceph global keyring config
Fabian Grünbichler [Thu, 4 Jul 2019 08:55:48 +0000 (10:55 +0200)]
5to6: check for ceph global keyring config

which causes issues after upgrading to Nautilus.

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: followup: also detect ceph conf keys separated with -
Thomas Lamprecht [Thu, 4 Jul 2019 08:37:35 +0000 (10:37 +0200)]
5to6: followup: also detect ceph conf keys separated with -

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks ago5to6: ceph mon host check: switch config check from defined to hash values exist
Thomas Lamprecht [Thu, 4 Jul 2019 08:31:06 +0000 (10:31 +0200)]
5to6: ceph mon host check: switch config check from defined to hash values exist

we can get here if a empty (or one with just comments) ceph.conf
exists

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks ago5to6: ceph global mon host check: also detect mon-host
Thomas Lamprecht [Thu, 4 Jul 2019 08:29:58 +0000 (10:29 +0200)]
5to6: ceph global mon host check: also detect mon-host

and adapt message a bit

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks agopve5to6: check ipv6/ipv4 settings in ceph.conf
Dominik Csapak [Thu, 4 Jul 2019 07:35:13 +0000 (09:35 +0200)]
pve5to6: check ipv6/ipv4 settings in ceph.conf

having ipv6 enabled while not disabling ipv4 prevents
nautilus osds to start if no ipv4 network is given (because they
are trying to bind to both ip families and die if one of them
is not found)

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks agopve5to6: check ceph config for mon_host line
Dominik Csapak [Thu, 4 Jul 2019 07:35:12 +0000 (09:35 +0200)]
pve5to6: check ceph config for mon_host line

this already works on luminous, so it does not harm to add it already,
and is recommended when both msgr1 and msgr2 is activated in nautilus

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks agoceph: mon create: add known monitor ips to mon_host if it is empty
Dominik Csapak [Wed, 3 Jul 2019 13:53:43 +0000 (15:53 +0200)]
ceph: mon create: add known monitor ips to mon_host if it is empty

this fixes an issue where only one monitor is in mon_host, which is
offline, prevents a client connection

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks agoceph: services: improve addr selection
Dominik Csapak [Wed, 3 Jul 2019 13:53:42 +0000 (15:53 +0200)]
ceph: services: improve addr selection

we map '$type addr' to '$type_addr' anyway in the ceph.conf parser,
so this is not necessary

also use 'public_addr' if it is set

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks agoceph: osd: use get-or-create to create a bootstrap-osd key on demand
Dominik Csapak [Wed, 3 Jul 2019 13:53:41 +0000 (15:53 +0200)]
ceph: osd: use get-or-create to create a bootstrap-osd key on demand

if for some reason the cluster does not have this key, generate it

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks agoceph: osd create: check for auth before getting bootstrap key
Dominik Csapak [Wed, 3 Jul 2019 13:53:40 +0000 (15:53 +0200)]
ceph: osd create: check for auth before getting bootstrap key

we do not need it if auth is 'none'

Signed-off-by: Dominik Csapak <d.csapak@proxmox.com>
2 weeks ago5to6: invert check for noout for nautilus
Fabian Grünbichler [Wed, 3 Jul 2019 13:28:38 +0000 (15:28 +0200)]
5to6: invert check for noout for nautilus

mainly because it looks strange to get a warning after the upgrade is
finished and noout has been removed again

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: followup: still include nodename in IP check
Thomas Lamprecht [Thu, 4 Jul 2019 07:33:43 +0000 (09:33 +0200)]
5to6: followup: still include nodename in IP check

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks ago5to6: reuse $nodename
Fabian Grünbichler [Wed, 3 Jul 2019 13:28:37 +0000 (15:28 +0200)]
5to6: reuse $nodename

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: improve some log messages
Fabian Grünbichler [Wed, 3 Jul 2019 13:28:36 +0000 (15:28 +0200)]
5to6: improve some log messages

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: followup sheepdog message
Thomas Lamprecht [Thu, 4 Jul 2019 07:35:59 +0000 (09:35 +0200)]
5to6: followup sheepdog message

do not suggest that we ever supported it as first class storage, it
was always just a experimental support and upstream is
EOL/unmaintained now.

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks ago5to6: add check for configured Sheepdog storages
Fabian Grünbichler [Wed, 3 Jul 2019 13:28:35 +0000 (15:28 +0200)]
5to6: add check for configured Sheepdog storages

Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks agofollowup: do not use string comparision for integers
Thomas Lamprecht [Wed, 3 Jul 2019 13:32:02 +0000 (15:32 +0200)]
followup: do not use string comparision for integers

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
2 weeks agoapi: ceph: automatically create manager after the first monitor
Tim Marx [Wed, 3 Jul 2019 12:40:10 +0000 (14:40 +0200)]
api: ceph: automatically create manager after the first monitor

Signed-off-by: Tim Marx <t.marx@proxmox.com>
2 weeks agoceph: init: fix undef var access
Fabian Grünbichler [Wed, 3 Jul 2019 13:10:02 +0000 (15:10 +0200)]
ceph: init: fix undef var access

Fixes: 64f3e4a522f2c1b2732122fd1c2a69959900fb0c
Signed-off-by: Fabian Grünbichler <f.gruenbichler@proxmox.com>
2 weeks ago5to6: improve ceph warning on non OK state
Thomas Lamprecht [Wed, 3 Jul 2019 11:14:05 +0000 (13:14 +0200)]
5to6: improve ceph warning on non OK state

Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>