From: Thomas Lamprecht
The whole thing is already prepared for this, the systemd timer was
just a fixed periodic timer with a frequency of one minute. And we
just introduced it as the assumption was made that less memory usage
would be generated with this approach, AFAIK.
But logging 4+ lines ju
Signed-off-by: Dominik Csapak
---
data/PVE/Cluster.pm | 1 +
data/src/status.c | 1 +
2 files changed, 2 insertions(+)
diff --git a/data/PVE/Cluster.pm b/data/PVE/Cluster.pm
index 4d09c60..ab8f713 100644
--- a/data/PVE/Cluster.pm
+++ b/data/PVE/Cluster.pm
@@ -45,6 +45,7 @@ my $dbbackupdir = "/
the only difference is that reload-or-try-restart does not
do anything if the service is not started already.
on upgrade, we explicitely check if the service is enabled, and only
then do this action. so it would now start daemons that were
stopped but not disabled (which is not really valid state
with this series, we implement a new daemon (pvescheduler) that takes
over from pvesrs' systemd timer (original patch from thomas[0]) and
extends it with a generic job handling mechanism
then i convert the vzdump cron jobs to these jobs, the immediate
gain is that users can use calendarevent sched
PVE/Jobs is responsible to decide if the job must run (e.g. with a
schedule)
Signed-off-by: Dominik Csapak
---
PVE/Service/pvescheduler.pm | 19 +--
1 file changed, 17 insertions(+), 2 deletions(-)
diff --git a/PVE/Service/pvescheduler.pm b/PVE/Service/pvescheduler.pm
index ce55
this adds a SectionConfig handling for jobs (only 'vzdump' for now) that
represents a job that will be handled by pvescheduler and a basic
'job-state' handling for reading/writing state json files
this has some intersections with pvesrs state handling, but does not
use a single state file for all
in addition to listing the vzdump.cron jobs, also list from the
jobs.cfg file.
updates/creations go into the new jobs.cfg only now
and on update, starttime+dow get converted to a schedule
this transformation is straight forward, since 'dow'
is already in a compatible format (e.g. 'mon,tue') and we
Signed-off-by: Dominik Csapak
---
PVE/API2/Backup.pm | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/PVE/API2/Backup.pm b/PVE/API2/Backup.pm
index 3b343636..9dc3b48e 100644
--- a/PVE/API2/Backup.pm
+++ b/PVE/API2/Backup.pm
@@ -17,6 +17,8 @@ use PVE::VZDump::Common;
us
we can now show the id (since its not autogenerated anymore),
and we can always show/edit the schedule instead of the dow+starttime
also add an 'ID' field to the edit/create window and update the
backupinfo window as well
Signed-off-by: Dominik Csapak
---
www/manager6/dc/Backup.js | 47
we normally use `d/` for changes to the debian/ folder, iow, packaging
changes.
On 07.10.21 10:27, Dominik Csapak wrote:
> the only difference is that reload-or-try-restart does not
> do anything if the service is not started already.
>
> on upgrade, we explicitely check if the service is enable
saves a list of pciid <-> vmid mappings in /var/run
that we can check when we start a vm
if we're not given a pid but a timeout, we save the time when the
reservation will run out (current time + timeout + 5s) since each
vm start (until we can save the pid) varies from config to config
Signed-off
by having a vmid <-> pciid mapping in /var/run
i did not check if the vm has the pci device really in the config,
but we should not need that, since we remove the reservation again
in the cleanup step and check the running pid anyway.
changes from v1:
* use time-based reservation before starting (
on vm start, we reserve all pciids that we use, and
remove the reservation again in vm_stop_cleanup
first with only a time-based reservation but after the vm is started,
we reserve again but with the pid.
for this, we have to move the start_timeout calculation above the
hostpci handling.
this wa
Some distributions like CentOS 8 and Gentoo don't have the file, so
only update if it already existed.
A slight change in behavior in set_timezone is that the warning will
now trigger if /etc/localtime is a link to $tz_path, but $tz_path does
not exist. Previously, it would return early if the lin
To avoid an error with 'pct create ... --timezone host'.
Reported in the community forum:
https://forum.proxmox.com/threads/pct-create-command-with-timezone-host-option-fails-to-create-a-container.97538/
Signed-off-by: Fabian Ebner
---
src/PVE/LXC/Setup.pm | 4 +++-
1 file changed, 3 insertions
On 07.10.21 11:37, Dominik Csapak wrote:
> saves a list of pciid <-> vmid mappings in /var/run
> that we can check when we start a vm
>
> if we're not given a pid but a timeout, we save the time when the
> reservation will run out (current time + timeout + 5s) since each
> vm start (until we can s
On 07.10.21 12:48, Fabian Ebner wrote:
> To avoid an error with 'pct create ... --timezone host'.
>
> Reported in the community forum:
> https://forum.proxmox.com/threads/pct-create-command-with-timezone-host-option-fails-to-create-a-container.97538/
>
> Signed-off-by: Fabian Ebner
> ---
> src/
On 07.10.21 12:48, Fabian Ebner wrote:
> Some distributions like CentOS 8 and Gentoo don't have the file, so
> only update if it already existed.
>
> A slight change in behavior in set_timezone is that the warning will
> now trigger if /etc/localtime is a link to $tz_path, but $tz_path does
> not
On 06.10.21 17:52, Stefan Reiter wrote:
> Signed-off-by: Stefan Reiter
> ---
> qm.adoc | 30 --
> 1 file changed, 24 insertions(+), 6 deletions(-)
>
>
applied, thanks!
___
pve-devel mailing list
pve-devel@lists.proxmox.co
On 06.10.21 17:52, Stefan Reiter wrote:
> Signed-off-by: Stefan Reiter
> ---
> qm.adoc | 30 ++
> 1 file changed, 30 insertions(+)
>
>
applied, thanks!
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.
Signed-off-by: Lorenz Stechauner
---
www/manager6/storage/PBSEdit.js | 7 +++
1 file changed, 7 insertions(+)
diff --git a/www/manager6/storage/PBSEdit.js b/www/manager6/storage/PBSEdit.js
index fcbc9548..0d7b8033 100644
--- a/www/manager6/storage/PBSEdit.js
+++ b/www/manager6/storage/PBSEdi
Am 28.09.21 um 15:07 schrieb Lorenz Stechauner:
the plugins for file based storages
* BTRFS
* CIFS
* Dir
* Glusterfs
* NFS
now allow the option 'preallocation'.
'preallocation' can have four values:
* default
* off
* metadata
* falloc
* full
see man pages for `qemu-img` for w
Am 28.09.21 um 15:07 schrieb Lorenz Stechauner:
Signed-off-by: Lorenz Stechauner
---
www/manager6/controller/StorageEdit.js | 6 ++
www/manager6/storage/Base.js | 18 ++
www/manager6/storage/NFSEdit.js| 2 +-
3 files changed, 25 insertions(+), 1 delet
Series looks good to me, except for a small nit.
Reviewed-by: Fabian Ebner
Am 28.09.21 um 15:07 schrieb Lorenz Stechauner:
changes to v1:
* adjusted preallocation api description
* moved sub preallocation_cmd_option above `# Storage implementation`
* updated PreallocationSelector to work with
there are a few ways to implement that and the commit message gives zero
reasoning
about the chosen one, I do not really like that...
why not allowing one to enter it as suffix in the host field like we do in PBS
for
the remotes.
An extra field is always distracting for newer users as it's yet
On 07.10.21 14:55, Thomas Lamprecht wrote:
> there are a few ways to implement that and the commit message gives zero
> reasoning
> about the chosen one, I do not really like that...
>
> why not allowing one to enter it as suffix in the host field like we do in
> PBS for
> the remotes.
>
> An e
the current default chains end with an expired root certificate for
maximum compatibility with old Android versions. this breaks some other
older clients (openssl, gnutls) which don't expect chains to contain any
expired certificates, even if they are 'above' the trust anchor.
setting $root will d
On 06.10.21 11:18, Fabian Ebner wrote:
> pve-storage:
>
> Fabian Ebner (6):
> diskmanage: add change_parttype and is_partition helpers
> diskmanage: wipe blockdev: also change partition type
> diskmanage: don't set usage for unused partitions
> api: disks: initgpt: explicitly abort for par
makes the vm start a bit less crowded
Signed-off-by: Dominik Csapak
---
PVE/QemuServer.pm | 14 +-
PVE/QemuServer/PCI.pm | 20
2 files changed, 21 insertions(+), 13 deletions(-)
diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm
index e5175b3..f78b2cc 100644
saves a list of pciid <-> vmid mappings in /var/run
that we can check when we start a vm
if we're not given a pid but a timeout, we save the time when the
reservation will run out (current time + timeout + 5s) since each
vm start (until we can save the pid) varies from config to config
reserve_pc
on vm start, we reserve all pciids that we use, and
remove the reservation again in vm_stop_cleanup
first with only a time-based reservation but after the vm is started,
we reserve again but with the pid.
for this, we have to move the start_timeout calculation above the
hostpci handling.
also mo
by having a vmid <-> pciid mapping in /var/run
i did not check if the vm has the pci device really in the config,
but we should not need that, since we remove the reservation again
in the cleanup step and check the running pid anyway.
notes:
1/3 is more or less independent, but made the code less
Huge Thanks for addressing this so fast!
quickly tested it on a (publicly accessible) system of mine
against Let's Encrypt official API (boulder)
It works as advertised and fetches the (shorter) certificate chain without
the signed ISRG X1 as intermediate (signed by DST Root CA X3)
So this fixes
33 matches
Mail list logo
