Hi,
On 6/12/19 1:02 AM, Alexandre DERUMIER wrote:
> can you already apply this patch series ? (it don't touch
> pve-manager/pve-common)
>
yes, sure, looks good and applied, thanks!
___
pve-devel mailing list
pve-devel@pve.proxmox.com
Hi,
can you already apply this patch series ? (it don't touch
pve-manager/pve-common)
Thanks
Alexandre
- Mail original -
De: "aderumier"
À: "pve-devel"
Cc: "aderumier"
Envoyé: Vendredi 7 Juin 2019 06:52:11
Objet: [PATCH pve-network 0/2] prepare for merging other network code
-
use a format string this time, so we can drop the parameter count a
bit. With address as default_key we can still use it conveniently
from CLI.
There will be the link priority flag added in a future patch.
Signed-off-by: Thomas Lamprecht
---
changes v1 (RFC) -> v2:
* drop bindnet property
For now in passive mode, a link with a higher value has a lower
priority. If the current active link fails the one with the next
higher priority will take over. Use 255 as maximum, as internally
kronosnet uses an uin8_t variable for this, and while there can be
"only" 8 links currently it may be
v2 with a few more patches and some changes regarding Fabian's feedback
(thanks) from the initial RFC version[0].
Last two patches, both new, aren't really tested by me to much, just FYI,
the others seen mostly reasonable adaptions and got smoke-tested for them, as I
looked more close last time
Preparation for enhanced compatibility with new corosync 3/knet
transport. Pretty straight forward switch from ringX_addr to links,
*but*, for configuration backward compatibility corosync still uses
"ringX_addr" as "link address", this will surely add confusion"link
address", this will surely add
Corosync has moved its rings a layer up, i.e., abstracted away from
the network layer below. That what early were called rings are now
links, knet can have up to 8 all others 1, for now.
Let our parser understand this change in the totem section of the
config, but keep backwards compatibility and
Similar to the change to cluster creation use now also the
corosync-link definition for the rest of the cluster join/add calls.
As link0, former ring0, is not special anymore allow that it's not
passed and only default back to nodename if it's configured in the
totem section of the configuration.
no users of this anymore, so drop it.
Signed-off-by: Thomas Lamprecht
---
NO changes v1 (RFC) -> v2
data/PVE/API2/ClusterConfig.pm | 16
1 file changed, 16 deletions(-)
diff --git a/data/PVE/API2/ClusterConfig.pm b/data/PVE/API2/ClusterConfig.pm
index 9c91880..dabdeb4 100644
allows to join clusters with more than the two links/rings we and
corosync supported earlier, further it reduced the special treatment
of link0 a lot.
Signed-off-by: Thomas Lamprecht
---
new in v2
data/PVE/API2/ClusterConfig.pm | 37 --
data/PVE/Cluster.pm
This adds basic infrastructure for link[0..7] parameters and allows
to use it for the create cluster call. Try to be as link independent
as possible, i.e., no real link0 special handling, it'll only get set
if there's no single link passed.
Signed-off-by: Thomas Lamprecht
---
new in v2
On 6/11/19 5:55 PM, Oguz Bektas wrote:
> make the message a bit more informative (with help from thomas), namely
> mentioning the ability to change/increase the limit.
>
> Signed-off-by: Oguz Bektas
> ---
>
> v1 -> v2:
> * change wording with suggestion from thomas
>
> PVE/VZDump.pm | 4 +++-
On 6/11/19 3:31 PM, Christian Ebner wrote:
> On some occasions e.g. license checking, the manufacturer string in the
> SMBIOS settings edit has to allow characters such as whitespaces.
> https://forum.proxmox.com/threads/proxmox-and-windows-rok-license-for-dell.53236/
> In principle SMBIOS allows
On 6/11/19 12:13 PM, Christian Ebner wrote:
> On some occasions e.g. license checking, the manufacturer string in the
> SMBIOS settings edit has to allow characters such as whitespaces.
> https://forum.proxmox.com/threads/proxmox-and-windows-rok-license-for-dell.53236/
> In principle SMBIOS allows
make the message a bit more informative (with help from thomas), namely
mentioning the ability to change/increase the limit.
Signed-off-by: Oguz Bektas
---
v1 -> v2:
* change wording with suggestion from thomas
PVE/VZDump.pm | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git
On 6/11/19 5:41 PM, Oguz Bektas wrote:
> On Tue, Jun 11, 2019 at 05:34:16PM +0200, Thomas Lamprecht wrote:
>> not really true.
>> There's a _storage_ (not drive) wide possible maxfiles limit which can
>> get overwritten through the CLI or API with the --maxfile parameter of
>> vzdump, so the
hi
On Tue, Jun 11, 2019 at 05:34:16PM +0200, Thomas Lamprecht wrote:
> On 6/11/19 5:18 PM, Oguz Bektas wrote:
> > make the message to be a little more informative, namely mentioning the
> > ability to change this option.
> >
> > Signed-off-by: Oguz Bektas
> > ---
> > PVE/VZDump.pm | 3 ++-
> >
On 6/11/19 5:18 PM, Oguz Bektas wrote:
> make the message to be a little more informative, namely mentioning the
> ability to change this option.
>
> Signed-off-by: Oguz Bektas
> ---
> PVE/VZDump.pm | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/PVE/VZDump.pm
make the message to be a little more informative, namely mentioning the
ability to change this option.
Signed-off-by: Oguz Bektas
---
PVE/VZDump.pm | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/PVE/VZDump.pm b/PVE/VZDump.pm
index 0a35eea7..0f8572fc 100644
---
high level looks ok, some comments inline
On 6/11/19 2:04 PM, Tim Marx wrote:
Signed-off-by: Tim Marx
---
Makefile| 1 +
form/NetworkSelector.js | 121
2 files changed, 122 insertions(+)
create mode 100644
On 6/11/19 2:28 PM, Alwin Antreich wrote:
> Note: fuse3 (3.4.1) in buster supports _netdev
>
It got "supported" (read, ignored) by FUSE's default mount helper
since quite a bit longer:
https://github.com/libfuse/libfuse/commit/b32c79b97079d2de63213bbeb12d3bf57e7e6643
(included since 2.5.0,
On 6/11/19 2:22 PM, Dominik Csapak wrote:
> it does not make sense to have it in the pci passthrough chapter
> (it was there by accident)
>
> Signed-off-by: Dominik Csapak
> ---
> qm.adoc | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
applied, thanks!
On 6/11/19 2:15 PM, Tim Marx wrote:
> Signed-off-by: Tim Marx
> ---
> www/manager6/ceph/Status.js | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/www/manager6/ceph/Status.js b/www/manager6/ceph/Status.js
> index 9364de14..48967279 100644
> ---
On 6/11/19 2:47 PM, Tim Marx wrote:
> Signed-off-by: Tim Marx
> ---
> www/manager6/ceph/CephInstallWizard.js | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/www/manager6/ceph/CephInstallWizard.js
> b/www/manager6/ceph/CephInstallWizard.js
> index 65d71006..5013066c
Signed-off-by: Tim Marx
---
www/manager6/ceph/CephInstallWizard.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/ceph/CephInstallWizard.js
b/www/manager6/ceph/CephInstallWizard.js
index 65d71006..5013066c 100644
--- a/www/manager6/ceph/CephInstallWizard.js
+++
fuse 2.9.x does not support the _netdev option and a cephfs storage
mount fails. This patch ignores the _netdev option if 'fuse 1' is set on
the cephfs storage.
Signed-off-by: Alwin Antreich
---
Note: fuse3 (3.4.1) in buster supports _netdev
PVE/Storage/CephFSPlugin.pm | 2 +-
1 file changed,
it does not make sense to have it in the pci passthrough chapter
(it was there by accident)
Signed-off-by: Dominik Csapak
---
qm.adoc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/qm.adoc b/qm.adoc
index 2a26673..81fd39e 100644
--- a/qm.adoc
+++ b/qm.adoc
@@ -1081,7
Signed-off-by: Tim Marx
---
www/manager6/ceph/Status.js | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/www/manager6/ceph/Status.js b/www/manager6/ceph/Status.js
index 9364de14..48967279 100644
--- a/www/manager6/ceph/Status.js
+++ b/www/manager6/ceph/Status.js
@@ -306,7
replace ceph-disk with ceph-volume
add note about db/wal size
and note about filestore
Signed-off-by: Dominik Csapak
---
pveceph.adoc | 72 +---
1 file changed, 25 insertions(+), 47 deletions(-)
diff --git a/pveceph.adoc b/pveceph.adoc
On 5/29/19 10:50 AM, Fabian Grünbichler wrote:
> I wonder whether we should support clusters without link0 as well? might
> happen easily:
>
> cluster only has link0 configured
> network topology change, add new link1
> deconfigure link0
>
> cluster joining no longer possible without specifying
Signed-off-by: Tim Marx
---
Makefile| 1 +
form/NetworkSelector.js | 121
2 files changed, 122 insertions(+)
create mode 100644 form/NetworkSelector.js
diff --git a/Makefile b/Makefile
index b9dc8b9..d12a4da 100644
---
Signed-off-by: Tim Marx
---
www/manager6/ceph/CephInstallWizard.js | 7 +++
1 file changed, 3 insertions(+), 4 deletions(-)
diff --git a/www/manager6/ceph/CephInstallWizard.js
b/www/manager6/ceph/CephInstallWizard.js
index 65d71006..bb7868f2 100644
---
On 6/11/19 1:43 PM, Rhonda D'Vine wrote:
> It is useful to have a way to see the translation statistics. The
> output could potentially get used to have an overview page about the
> translations and invite people to help out through that.
>
> Signed-off-by: Rhonda D'Vine
> ---
> Makefile | 3
It is useful to have a way to see the translation statistics. The
output could potentially get used to have an overview page about the
translations and invite people to help out through that.
Signed-off-by: Rhonda D'Vine
---
Makefile | 3 +++
1 file changed, 3 insertions(+)
diff --git
this parameter changed sometime between luminous and nautilus
note that with this change, it is not possible to delete pools in
luminous anymore
Signed-off-by: Dominik Csapak
---
PVE/Ceph/Tools.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/PVE/Ceph/Tools.pm
applied all three patches, thanks!
___
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
On some occasions e.g. license checking, the manufacturer string in the
SMBIOS settings edit has to allow characters such as whitespaces.
https://forum.proxmox.com/threads/proxmox-and-windows-rok-license-for-dell.53236/
In principle SMBIOS allows to pass any zero terminated string to the
On some occasions e.g. license checking, the manufacturer string in the
SMBIOS settings edit has to allow characters such as whitespaces.
https://forum.proxmox.com/threads/proxmox-and-windows-rok-license-for-dell.53236/
In principle SMBIOS allows to pass any zero terminated string to the
On some occasions e.g. license checking, the manufacturer string in the
SMBIOS settings edit has to allow characters such as whitespaces.
https://forum.proxmox.com/threads/proxmox-and-windows-rok-license-for-dell.53236/
In principle SMBIOS allows to pass any zero terminated string to the
>>We could either:
>>* see if we keep INotify in pve-common, if we really need we could overwrite,
>>or extend it with a INotify "sub-class" in pve-network, did not really
>>looked into it, so not sure how reasonable and feasible that is..
>>* move INotify and keep a copy of it for the PMG
ceph luminous does not use the 'name' property in the metadata
everywhere, so fall back to 'id'
this makes the ceph dashboard usable while having still luminous
(relevant for upgrading)
Signed-off-by: Dominik Csapak
---
PVE/API2/Cluster.pm | 2 +-
PVE/Ceph/Services.pm | 3 ++-
2 files
uses cpeh-volumes --dmcrypt parameter to encrypt the osd
Signed-off-by: Dominik Csapak
---
PVE/API2/Ceph/OSD.pm | 7 +++
1 file changed, 7 insertions(+)
diff --git a/PVE/API2/Ceph/OSD.pm b/PVE/API2/Ceph/OSD.pm
index b9b0c6cf..42dee361 100644
--- a/PVE/API2/Ceph/OSD.pm
+++
Signed-off-by: Dominik Csapak
---
www/manager6/ceph/OSD.js | 7 +++
1 file changed, 7 insertions(+)
diff --git a/www/manager6/ceph/OSD.js b/www/manager6/ceph/OSD.js
index cadc2b0e..db2e1bd9 100644
--- a/www/manager6/ceph/OSD.js
+++ b/www/manager6/ceph/OSD.js
@@ -73,6 +73,13 @@
On 6/11/19 9:46 AM, Alexandre DERUMIER wrote:
>>> I have send patches, but I don't have look at PMG currently. (I never have
>>> used it to be honest)
>>> (I have only move the network configuration part, and was only used in
>>> Network API)
> I have looked in PMG,
>
> seem that it shouldn't
>>I have send patches, but I don't have look at PMG currently. (I never have
>>used it to be honest)
>>(I have only move the network configuration part, and was only used in
>>Network API)
I have looked in PMG,
seem that it shouldn't break it (maybe simply need to add a depend on
pve-network)
45 matches
Mail list logo
