Re: [pve-devel] LXC Container

On Thu, Feb 23, 2017 at 01:33:04AM +0100, Detlef Bracker wrote: > Every scenerario breaks with other problems - So has somebody a resolution? > > Plesk will not work on LCX container about mount problems! A backup and > restore as an unprivileged server with option ignore restore fails, > brings n

[pve-devel] LXC container with unprivileged mode

Installation of diferent things hangs and long log files with this message comes: Debian 8 Proxmox LCX-Image - updated to systemd mode - rebooted - updated - installation of plesk expl. Feb 23 01:59:34 ct1004 systemd[1]: Looping too fast. Throttling execution a little. Feb 23 01:59:35 ct1004

[pve-devel] LXC Container

Every scenerario breaks with other problems - So has somebody a resolution? Plesk will not work on LCX container about mount problems! A backup and restore as an unprivileged server with option ignore restore fails, brings new other problems why then the system is no more usable - possible parts o

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

> values.enable = values.disable ? 0 : 1; > + // reverse the logic for the checkbox > + values.nowritecache = values.nowritecache == '0'?'1':'0'; It would be much nicer if this gets handled by the form element itself ... But we can apply this

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On Wed, 22 Feb 2017 14:42:28 +0100 Dominik Csapak wrote: > the logic was reversed, checking writecache enabled 'nowritecache', > now the logic is correct, setting 'nowritecache' if 'writecache' is > unchecked > > Signed-off-by: Dominik Csapak > --- > changes from v1: > reverse also the logic w

[pve-devel] [PATCH cluster 3/8] pvecm create: remove rrp_mode parameter

I detected a bug where we overwrote the whole $interfaces variable (and so all interfaces from the corosync config) if the 'rrp_mode' param was set. While this would be easy to with by changing the line to $interfaces .= ... I removed the whole rrp_mode parameter instead. As: a) I've seen no one

[pve-devel] [PATCH cluster 6/8] pvecm addnode: ensure ring1_addr is set if ring 1 is configured

Else the joining node will not be able to work correctly. Also improve the respective error messages. Signed-off-by: Thomas Lamprecht --- data/PVE/CLI/pvecm.pm | 5 - 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/data/PVE/CLI/pvecm.pm b/data/PVE/CLI/pvecm.pm index 1c5d863..e4d

[pve-devel] [PATCH cluster 1/8] pvecm add: fix check if corosync alread runs

`corosync-quorumtool` exit with 1 (CS_OK) if corosync runs and is quorate. Use `corosync-quorumtool -l` (list nodes) instead, this returns 1 if corosync does not run 0 if corosync runs, independent if a cluster is quorate or not. Signed-off-by: Thomas Lamprecht --- corosync and its exit codes ar

[pve-devel] [PATCH cluster 8/8] pvecm add: assert that ringX IPs are available on node add

If 'ringX_addr' parameters are used on adding a node to a cluster check if those addresses are actually configured on the to-be-added node. It makes no sense that the address is not or multiple times configured. This prevents a node in limbo, waiting for quorum (if it was the second node in a clus

[pve-devel] [PATCH cluster 7/8] pvecm add: report all errors found at once

Else only the first error got reported and we had no idea what else was possible wrong. I'll also use the $err method more in next commits Signed-off-by: Thomas Lamprecht --- data/PVE/CLI/pvecm.pm | 23 +++ 1 file changed, 19 insertions(+), 4 deletions(-) diff --git a/data/

[pve-devel] [PATCH cluster 5/8] pvecm addnode: ensure ring address isn't already used by cluster

If someone enters the wrong address by accident when adding a node it may cause havoc in the cluster (meaning a reset of the whole cluster when HA is used, may even happen more often during the recovery tries. Also a whole lot of problems get triggered in gneral, even witouth HA). Further, user ge

[pve-devel] [PATCH cluster 4/8] pvecm addnode: error out on interactive call

addnode is thought to be used by the `add` command only. So check if STDIN or STOUT are connected to a tty and exit with an error message if this is the case. The force flag allows overwriting this check. Fixes bug #294 Signed-off-by: Thomas Lamprecht --- data/PVE/CLI/pvecm.pm | 5 + 1 file

[pve-devel] [PATCH cluster 2/8] pvecm: small cleanup

clvm is was used in 3.4 and earlier, it won't come back anytime soon Signed-off-by: Thomas Lamprecht --- data/PVE/CLI/pvecm.pm | 2 -- 1 file changed, 2 deletions(-) diff --git a/data/PVE/CLI/pvecm.pm b/data/PVE/CLI/pvecm.pm index c95e8c6..f15f467 100755 --- a/data/PVE/CLI/pvecm.pm +++ b/data/P

[pve-devel] [PATCH cluster 0/8] pvecm: fixes and adding additional checks

First 3 patches are fixes Fourth patch addresses a possible misconception of the addnode call The rest of the series adds additional error checks when adding a node to a cluster and improves those checks by outputting all at once, not only the first. I tested some new cluster creation and node add

[pve-devel] [PATCH spiceterm] use color_table to get the correct terminal colors

this was missing, and the colors in the terminal were wrong Signed-off-by: Dominik Csapak --- screen.c | 15 +-- 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/screen.c b/screen.c index 9072dce..1fe28a5 100644 --- a/screen.c +++ b/screen.c @@ -56,6 +56,9 @@ static int

[pve-devel] [PATCH] updated to latest stable upstream (1.0.1)

We still used RC1 from the 1.0 version, a few bugs and documentation fixes happened in the meanwhile so updated to the latest stable version 1.0.1 Signed-off-by: Thomas Lamprecht --- After applying the patch a $ make download needs to be made, the deleting of the old source tarball and the addin

[pve-devel] applied: [PATCH kvm 1/2] fix CVE-2017-2620: display: cirrus: out-of-bounds access issue

Signed-off-by: Fabian Grünbichler --- ...s_unsafe_call_to_cirrus_bitblt_cputovideo.patch | 52 ++ debian/patches/series | 1 + 2 files changed, 53 insertions(+) create mode 100644 debian/patches/extra/CVE-2017-2620_cirrus_add_blit_is_unsafe_call

[pve-devel] applied: [PATCH kvm 2/2] bump version to 2.7.1-4

Signed-off-by: Fabian Grünbichler --- Makefile | 2 +- debian/changelog | 6 ++ 2 files changed, 7 insertions(+), 1 deletion(-) diff --git a/Makefile b/Makefile index fc4a6e8..1d18b92 100644 --- a/Makefile +++ b/Makefile @@ -2,7 +2,7 @@ RELEASE=4.4 # also update debian/changelog

[pve-devel] [PATCH 05/19] phase2 : create a new vm on external node

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 15 +++ 1 file changed, 15 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 71af817..a2fa17a 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -471,6 +471,21 @@ sub phase2 { my $conf =

[pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

the logic was reversed, checking writecache enabled 'nowritecache', now the logic is correct, setting 'nowritecache' if 'writecache' is unchecked Signed-off-by: Dominik Csapak --- changes from v1: reverse also the logic when using in an editwindow have to use strings because we get a string fro

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On 02/22/2017 02:28 PM, datanom.net wrote: On 2017-02-22 14:05, Dominik Csapak wrote: On 02/22/2017 02:03 PM, Fabian Grünbichler wrote: On Wed, Feb 22, 2017 at 11:50:56AM +0100, Emmanuel Kasper wrote: no idea what extJS does here - but it looks like this will always result in nowritecache being

[pve-devel] [PATCH 06/19] phase2 : write vm config to externalcluster new vmid.conf

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 10 ++ 1 file changed, 10 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index a2fa17a..616632c 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -473,6 +473,7 @@ sub phase2 { if ($self->{opts

[pve-devel] [PATCH 13/19] phase3_cleanup : don't move source vmid.conf

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 13 - 1 file changed, 8 insertions(+), 5 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 7bd0d6f..67f7cd1 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -960,12 +960,15 @@ sub phase3_clean

[pve-devel] [PATCH 12/19] phase3 : don't destroy local copy

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 2 ++ 1 file changed, 2 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 13952cb..7bd0d6f 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -924,6 +924,8 @@ sub phase3 { my $volids = $self->{volumes

[pve-devel] [PATCH 19/19] vm_start : targetstorage : create disks also for shared storage if !migratedfrom

Signed-off-by: Alexandre Derumier --- PVE/QemuServer.pm | 7 --- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index e084981..d7f2540 100644 --- a/PVE/QemuServer.pm +++ b/PVE/QemuServer.pm @@ -4498,9 +4498,10 @@ sub vm_start {

[pve-devel] [PATCH 16/19] phase3_cleanup : don't free disk of source vm

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 22 +- 1 file changed, 13 insertions(+), 9 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 6c238b7..64f3565 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -1020,15 +1020,19 @@ sub

[pve-devel] [PATCH 09/19] phase2 : fix vmid for unix socket

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index f1d621a..ab396b3 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -562,7 +562,7 @@ sub phase2 { } e

[pve-devel] [PATCH 08/19] phase2: add targetvmid

default is targetvmid=vmid, until we create a new one in case of externalcluster Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 9 + 1 file changed, 5 insertions(+), 4 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index b711564..f1d621a 100644 --- a/PVE/Q

[pve-devel] [PATCH 18/19] vm_start : remove mandatory migratedfrom for targetstorage option check

Signed-off-by: Alexandre Derumier --- PVE/API2/Qemu.pm | 3 --- 1 file changed, 3 deletions(-) diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm index e6d46ba..49e9816 100644 --- a/PVE/API2/Qemu.pm +++ b/PVE/API2/Qemu.pm @@ -1704,9 +1704,6 @@ __PACKAGE__->register_method({ raise_param_exc

[pve-devel] [PATCH 15/19] phase3_cleanup : clear migration lock on source vm

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 6 ++ 1 file changed, 6 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 35b752e..6c238b7 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -1043,6 +1043,12 @@ sub phase3_cleanup { } # cl

[pve-devel] [PATCH 07/19] phase2_cleanup : destroy remote external vm on error

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 28 ++-- 1 file changed, 26 insertions(+), 2 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 616632c..b711564 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -483,6 +483,7 @@ su

[pve-devel] [PATCH 14/19] phase3_cleanup: add targetvmid

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 9 ++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 67f7cd1..35b752e 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -940,6 +940,9 @@ sub phase3 { sub phase

[pve-devel] [PATCH 17/19] phase3_cleanup : don't write config in current cluster

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 12 1 file changed, 8 insertions(+), 4 deletions(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 64f3565..8779b00 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -955,10 +955,14 @@ sub phase3_cleanu

[pve-devel] [PATCH 11/19] phase2 : write target vm config after disk create

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 19 +++ 1 file changed, 19 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 1ca445d..13952cb 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -594,6 +594,25 @@ sub phase2 { die "u

[pve-devel] [PATCH 10/19] phase2 : don't sent migratedfrom for externalmigration

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 6 +- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index ab396b3..1ca445d 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -514,7 +514,11 @@ sub phase2 { $spic

[pve-devel] [PATCH 03/19] disable target storage avaibility for externalcluster (fixme)

need to implement a method to check that remotly. that later in ssh tunnel ? Signed-off-by: Alexandre Derumier --- PVE/API2/Qemu.pm | 2 +- PVE/QemuMigrate.pm | 4 ++-- 2 files changed, 3 insertions(+), 3 deletions(-) diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm index 01d97bc..e6d46ba 100

[pve-devel] [PATCH 04/19] sync_disk: migrate all disk for externalcluster

Signed-off-by: Alexandre Derumier --- PVE/QemuMigrate.pm | 9 + 1 file changed, 9 insertions(+) diff --git a/PVE/QemuMigrate.pm b/PVE/QemuMigrate.pm index 8d1f9e8..71af817 100644 --- a/PVE/QemuMigrate.pm +++ b/PVE/QemuMigrate.pm @@ -293,6 +293,12 @@ sub sync_disks { my $targe

[pve-devel] [PATCH 02/19] add qm migrateexternal

Signed-off-by: Alexandre Derumier --- PVE/API2/Qemu.pm | 84 PVE/CLI/qm.pm| 2 ++ 2 files changed, 86 insertions(+) diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm index 76bba70..01d97bc 100644 --- a/PVE/API2/Qemu.pm +++ b/PVE/API2/Q

[pve-devel] [PATCH 01/19] qm create : make vmid optionnal

if ommit, we generate it with PVE::Cluster::next_vmid Signed-off-by: Alexandre Derumier --- PVE/API2/Qemu.pm | 14 +++--- 1 file changed, 11 insertions(+), 3 deletions(-) diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm index a077ed7..76bba70 100644 --- a/PVE/API2/Qemu.pm +++ b/PVE/API2

[pve-devel] RFC: vm migration+storage to external/remote proxmox cluster

Hi, This patches serie implement a new method qm migrateexternal -targetstorage to allow to live migrate+storage migrate a vm to a remote proxmox cluster, with a new vmid on the target cluster. Main differences with classic live storage migration: - new vmid on target cluster - src disks on

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On 2017-02-22 14:05, Dominik Csapak wrote: On 02/22/2017 02:03 PM, Fabian Grünbichler wrote: On Wed, Feb 22, 2017 at 11:50:56AM +0100, Emmanuel Kasper wrote: no idea what extJS does here - but it looks like this will always result in nowritecache being set to 0, no matter whether the check box

[pve-devel] [PATCH kernel 1/4] update to Ubuntu 4.4.0-63.84

Signed-off-by: Fabian Grünbichler --- Makefile | 2 +- ubuntu-xenial.tgz | Bin 145945176 -> 146076086 bytes 2 files changed, 1 insertion(+), 1 deletion(-) diff --git a/Makefile b/Makefile index e7edb76..814ec49 100644 --- a/Makefile +++ b/Makefile @@ -125,7 +125,7 @@ ${VIRTUAL_HDR_DE

[pve-devel] [PATCH kernel 3/4] update drbd to 9.0.6-1

Signed-off-by: Fabian Grünbichler --- Makefile| 2 +- drbd-9.0.5-1.tar.gz | Bin 390720 -> 0 bytes drbd-9.0.6-1.tar.gz | Bin 0 -> 394082 bytes 3 files changed, 1 insertion(+), 1 deletion(-) delete mode 100644 drbd-9.0.5-1.tar.gz create mode 100644 drbd-9.0.6-1.tar.gz diff --git

[pve-devel] [PATCH kernel 2/4] drop patches applied upstream

Signed-off-by: Fabian Grünbichler --- Makefile | 3 - ...hrottle-on-IO-only-when-there-are-too-man.patch | 118 -- 0002-Revert-mm-oom-rework-oom-detection.patch | 255 - ...x86-fix-emulation-of-MOV-SS-null-selector.patch

[pve-devel] [PATCH kernel 4/4] bump version to 4.4-81, bump ABI to 4.4.44-1-pve

Signed-off-by: Fabian Grünbichler --- changelog.Debian| 12 proxmox-ve/changelog.Debian | 8 Makefile| 4 ++-- 3 files changed, 22 insertions(+), 2 deletions(-) diff --git a/changelog.Debian b/changelog.Debian index f1da444..6733682 100644

[pve-devel] applied: [PATCH kernel 0/4] kernel updates

new Ubuntu upstream kernel, including a new/alternative fix for the OOM-killer issue from January updated DRBD 9 module to current version Fabian Grünbichler (4): update to Ubuntu 4.4.0-63.84 drop patches applied upstream update drbd to 9.0.6-1 bump version to 4.4-81, bump ABI to 4.4.44-1

[pve-devel] applied: [PATCH manager 0/3] Add a Clone button to the VM toolbar

On Mon, Feb 06, 2017 at 07:03:24PM +0100, Emmanuel Kasper wrote: > Cloning a VM was up to now only accessible via right clicking the > resource tree, a path that many novice PVE users did not notice. > Adding this is as a general button will make the feature easier to find. > > Emmanuel Kasper (3

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On 02/22/2017 02:03 PM, Fabian Grünbichler wrote: On Wed, Feb 22, 2017 at 11:50:56AM +0100, Emmanuel Kasper wrote: On 02/22/2017 11:12 AM, Dominik Csapak wrote: the logic was reversed, checking writecache enabled 'nowritecache', now the logic is correct, setting 'nowritecache' if 'writecache' i

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On Wed, Feb 22, 2017 at 11:50:56AM +0100, Emmanuel Kasper wrote: > On 02/22/2017 11:12 AM, Dominik Csapak wrote: > > the logic was reversed, checking writecache enabled 'nowritecache', > > now the logic is correct, setting 'nowritecache' if 'writecache' is > > unchecked > > > > Signed-off-by: Domi

Re: [pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

On 02/22/2017 11:12 AM, Dominik Csapak wrote: > the logic was reversed, checking writecache enabled 'nowritecache', > now the logic is correct, setting 'nowritecache' if 'writecache' is > unchecked > > Signed-off-by: Dominik Csapak > --- > i think this is a better and more elegant approach to the

Re: [pve-devel] [PATCH manager 0/3] Add a Clone button to the VM toolbar

On 02/06/2017 07:03 PM, Emmanuel Kasper wrote: Cloning a VM was up to now only accessible via right clicking the resource tree, a path that many novice PVE users did not notice. Adding this is as a general button will make the feature easier to find. Emmanuel Kasper (3): Fix handling of failed

[pve-devel] [PATCH manager] correctly set 'nowritecache' on zfsplugin

the logic was reversed, checking writecache enabled 'nowritecache', now the logic is correct, setting 'nowritecache' if 'writecache' is unchecked Signed-off-by: Dominik Csapak --- i think this is a better and more elegant approach to the patch from m...@datanom.net, and still has the same functio

Re: [pve-devel] [RFC PATCH manager] Make explicit the meaning of an empty storage field

On 02/09/2017 04:13 PM, Emmanuel Kasper wrote: + emptyText: gettext('Same as source'), On 02/08/2017 07:05 PM, Dietmar Maurer wrote: I would like to avoid additional gettext entries if possible. Do we know the source storage? If so, can we set that storage as 'emptyText'? Other idea