On Thu, Sep 22, 2022 at 01:37:57PM +0200, Dominik Csapak wrote:
> On 9/22/22 12:14, Matthias Heiserer wrote:
> > On 21.09.2022 14:49, Dominik Csapak wrote:
> > > instead of always sending a SIGKILL to the target pid.
> > > It was not that much of a problem since the timeout previously was 5
> > > s
On Thu, Sep 22, 2022 at 04:19:33PM +0200, Dominik Csapak wrote:
> currently, the 'forced_cleanup' (sending SIGKILL to the qemu process),
> is intended to be triggered 5 seconds after sending the initial shutdown
> signal (SIGTERM) which is sadly not enough for some setups.
>
> Accidentally, it cou
Only print it when there is a snapshot that would've been removed
without the safeguard. Mostly relevant when a new volume is added to
an already replicated guest.
Fixes replication tests in pve-manager.
Fixes: c0b2948 ("replication: prepare: safeguard against removal if expected
snapshot is mis
this is functionally the same, but sending SIGTERM has the ugly side
effect of printing the following to the log:
> QEMU[]: kvm: terminating on signal 15 from pid (/usr/sbin/qmeventd)
while sending a QMP quit command does not.
Signed-off-by: Dominik Csapak
---
qmeventd/qmeventd.c | 14 +++
instead of always sending a SIGKILL to the target pid.
It was not that much of a problem since the timeout previously was 5
seconds and we used pifds where possible, thus the chance of killing the
wrong process was rather slim.
Now we increased the timeout to 60s which makes the race a bit more li
currently, the 'forced_cleanup' (sending SIGKILL to the qemu process),
is intended to be triggered 5 seconds after sending the initial shutdown
signal (SIGTERM) which is sadly not enough for some setups.
Accidentally, it could be triggered earlier than 5 seconds, if a
SIGALRM triggers in the times
includes the following improvements:
* increases 'force cleanup' timeout to 60s (from 5)
* saves individual timeout for each vm
* don't force cleanup for vms where normal cleanup worked
* sending QMP quit instead of SIGTERM (less log noise)
changes from v2:
* change from cast of the function to ca
On Thu, Sep 22, 2022 at 04:19:34PM +0200, Dominik Csapak wrote:
> instead of always sending a SIGKILL to the target pid.
> It was not that much of a problem since the timeout previously was 5
> seconds and we used pifds where possible, thus the chance of killing the
> wrong process was rather slim.
sorry disregard, was too fast with this version and did not see that
wolfgang wrote something about 2/3 too
___
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
On 9/23/22 10:31, Wolfgang Bumiller wrote:
On Thu, Sep 22, 2022 at 04:19:34PM +0200, Dominik Csapak wrote:
instead of always sending a SIGKILL to the target pid.
It was not that much of a problem since the timeout previously was 5
seconds and we used pifds where possible, thus the chance of kill
this is functionally the same, but sending SIGTERM has the ugly side
effect of printing the following to the log:
> QEMU[]: kvm: terminating on signal 15 from pid (/usr/sbin/qmeventd)
while sending a QMP quit command does not.
Signed-off-by: Dominik Csapak
---
qmeventd/qmeventd.c | 14 +++
includes the following improvements:
* increases 'force cleanup' timeout to 60s (from 5)
* saves individual timeout for each vm
* don't force cleanup for vms where normal cleanup worked
* sending QMP quit instead of SIGTERM (less log noise)
changes from v3:
* merge CleanupData into Client (pidfd,t
currently, the 'forced_cleanup' (sending SIGKILL to the qemu process),
is intended to be triggered 5 seconds after sending the initial shutdown
signal (SIGTERM) which is sadly not enough for some setups.
Accidentally, it could be triggered earlier than 5 seconds, if a
SIGALRM triggers in the times
Signed-off-by: Stefan Hanreich
---
examples/guest-example-hookscript.pl | 12
1 file changed, 12 insertions(+)
diff --git a/examples/guest-example-hookscript.pl
b/examples/guest-example-hookscript.pl
index adeed59..345b5d9 100755
--- a/examples/guest-example-hookscript.pl
+++ b/exa
Signed-off-by: Stefan Hanreich
---
PVE/API2/Qemu.pm | 4
1 file changed, 4 insertions(+)
diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm
index 3ec31c2..23a7658 100644
--- a/PVE/API2/Qemu.pm
+++ b/PVE/API2/Qemu.pm
@@ -3417,6 +3417,8 @@ __PACKAGE__->register_method({
my ($conffil
This patch adds pre/post-clone hooks when the when the user clones a CT/VM
from the Web UI / CLI. I have tested this with both VMs/CTs via Web UI and CLI.
Are there any other places where the hook should get triggered that I missed?
Clone is a bit special since it can either target the same node o
Signed-off-by: Stefan Hanreich
---
src/PVE/API2/LXC.pm | 4
1 file changed, 4 insertions(+)
diff --git a/src/PVE/API2/LXC.pm b/src/PVE/API2/LXC.pm
index 589f96f..d6ebc08 100644
--- a/src/PVE/API2/LXC.pm
+++ b/src/PVE/API2/LXC.pm
@@ -1609,6 +1609,8 @@ __PACKAGE__->register_method({
On 9/23/22 11:55, Stefan Hanreich wrote:
Signed-off-by: Stefan Hanreich
---
examples/guest-example-hookscript.pl | 12
1 file changed, 12 insertions(+)
diff --git a/examples/guest-example-hookscript.pl
b/examples/guest-example-hookscript.pl
index adeed59..345b5d9 100755
--- a
applied both patches, thanks
On Fri, Sep 23, 2022 at 11:51:13AM +0200, Dominik Csapak wrote:
> includes the following improvements:
> * increases 'force cleanup' timeout to 60s (from 5)
> * saves individual timeout for each vm
> * don't force cleanup for vms where normal cleanup worked
> * sending
Am 21/07/2022 um 12:45 schrieb Matthias Heiserer:
> This makes it consistent with the naming scheme in PVE/GUI.
> Keep value for API stability reasons, and remove it in next major version.
>
> Signed-off-by: Matthias Heiserer
> ---
> src/tools/disks/smart.rs | 9 +++--
> 1 file changed, 7 in
Am 21/07/2022 um 12:45 schrieb Matthias Heiserer:
> This makes it consistent with the naming scheme in PBS/GUI.
> Keep value for API stability reasons and remove it in the next major version.
>
> Signed-off-by: Matthias Heiserer
> ---
> PVE/Diskmanage.pm | 2 ++
> ..
Am 22/09/2022 um 12:35 schrieb Dominik Csapak:
>
> the only (minor) thing is that the wt patch could have handled the current
> situation
> (pve raw+value, pbs normalized+value), e.g by using the field 'convert' or
> 'calculate'
> methods of extjs (we could have had a 'real_raw' and 'real_normal
the output can get quite long and warnings can easily be missed
otherwise.
Signed-off-by: Fabian Grünbichler
---
src/mirror.rs | 15 +--
1 file changed, 13 insertions(+), 2 deletions(-)
diff --git a/src/mirror.rs b/src/mirror.rs
index e655847..f8afd2b 100644
--- a/src/mirror.rs
+++
to make fetching errors from broken repositories non-fatal.
Signed-off-by: Fabian Grünbichler
---
based on top of "extend/add commands" series from
20220921081242.1139249-1-f.gruenbich...@proxmox.com
src/bin/proxmox-offline-mirror.rs | 2 ++
src/bin/proxmox_offline_mirror_cmds/conf
Hi,
does somebody had time to review this patch series ?
Does I need to rework it ? Any commment ?
Regards,
Alexandre
Le mercredi 24 août 2022 à 13:34 +0200, Alexandre Derumier a écrit :
> This patch add virtio-mem support, through a new maxmemory option.
>
> a 4GB static memory is needed fo
applied both patches and cleaned up 2 error handlers (one from this
patch and one older one)
On Fri, Sep 23, 2022 at 12:33:51PM +0200, Fabian Grünbichler wrote:
> to make fetching errors from broken repositories non-fatal.
>
> Signed-off-by: Fabian Grünbichler
> ---
> based on top of "extend/add
V1 -> V2:
# pve-storage
* removed because patch is not needed
# pve-manager (1/3)
* remove storage controller from V1
* added custom ComboBox with API URL & setNodeName function
* added scan node selection for iSCSI
* scan node selection field no longer send to server
## (optional) pve-manager (
Signed-off-by: Stefan Hrdlicka
---
www/manager6/storage/Base.js| 10 +-
www/manager6/storage/IScsiEdit.js | 6 +++---
www/manager6/storage/LVMEdit.js | 14 +++---
www/manager6/storage/LvmThinEdit.js | 18 +-
www/manager6/storage/ZFSPoolEdit.js | 23 +
This adds a dropdown box for iSCSI, LVM, LVMThin & ZFS storage options where a
cluster node needs to be chosen. As default the current node is
selected. It restricts the the storage to be only availabe on the
selected node.
Signed-off-by: Stefan Hrdlicka
---
www/manager6/Makefile
Hi Dominik,
I have finished my tests with pci passthrough && mdev,
I didn't have any problem this time, all is working fine for me !
Le 20/09/22 à 14:50, Dominik Csapak a écrit :
> this series aims to add a cluster-wide device mapping for pci and usb devices.
> so that an admin can configure a
30 matches
Mail list logo
