Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Alexandre DERUMIER
>>But that is true for file systems in general? Even if you're on a local >>"standard" filesystem and have a reader/writer process pair you need to ensure >>some level of atomicity regulation, either a (shared) rw_lock or like in our >>case, where the reader reopens the file on every read loop

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Thomas Lamprecht
On 1/8/19 10:19 PM, Alexandre DERUMIER wrote: >>> or those cases i use something like (pseudocode - i use salt not puppet): >>> >>> - manage copy of file >>> - if file has changed trigger: >>> - mv -v $managedfile $realfile >>> > >>> Greets, >>> Stefan > > Thanks Stefan, works fine indeed. >

Re: [pve-devel] [PATCH qemu-server v2 0/3] Fix #2041 and #413

2019-01-08 Thread Alexandre DERUMIER
Hi, they are also ich9-intel-hda as sound controller. (maybe when q35 is used) could be great to use intel hda for linux too, and not only modern windows. - Mail original - De: "Andreas Steinel" À: "pve-devel" Envoyé: Mardi 8 Janvier 2019 23:57:58 Objet: [pve-devel] [PATCH

[pve-devel] [PATCH qemu-server v2 0/3] Fix #2041 and #413

2019-01-08 Thread Andreas Steinel
Add a new option 'spicedesktop' to enable audio and folder sharing. More precisely, add another serial port in order to get the service spice-webdavd on Linux and Windows working. Afterwards you can use remote-viewer and enable folder sharing therein to get a new virtual inside of your guest for

[pve-devel] [PATCH qemu-server v2 3/3] Adding new config option 'spicedesktop'

2019-01-08 Thread Andreas Steinel
--- PVE/QemuServer.pm | 32 1 file changed, 20 insertions(+), 12 deletions(-) diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index 173ae82..657cfad 100644 --- a/PVE/QemuServer.pm +++ b/PVE/QemuServer.pm @@ -607,6 +607,12 @@ EODESCR default => "1

[pve-devel] [PATCH qemu-server v2 1/3] Fix #2041: add spice webdav / folder sharing

2019-01-08 Thread Andreas Steinel
Adding the device and serial port for the service spice-webdavd on Linux and Windows. Signed-off-by: Andreas Steinel --- PVE/QemuServer.pm | 5 + PVE/QemuServer/PCI.pm | 1 + 2 files changed, 6 insertions(+) diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index 1ccdccf..225f0c0

[pve-devel] [PATCH qemu-server v2 2/3] fix #413: add SPICE audio device

2019-01-08 Thread Andreas Steinel
If you enable SPICE, the audio device will be automatically added. Intel HD for newer Windows and AC97 otherwise. --- PVE/QemuServer.pm | 9 + PVE/QemuServer/PCI.pm | 1 + 2 files changed, 10 insertions(+) diff --git a/PVE/QemuServer.pm b/PVE/QemuServer.pm index 225f0c0..173ae82

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Alexandre DERUMIER
>>or those cases i use something like (pseudocode - i use salt not puppet): >> >>- manage copy of file >>- if file has changed trigger: >> - mv -v $managedfile $realfile >> >>Greets, >>Stefan Thanks Stefan, works fine indeed. I really didn't known/remember that /etc/pve was not atomic without

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Stefan Priebe - Profihost AG
Hi Alexandre, Am 08.01.19 um 21:55 schrieb Alexandre DERUMIER: >>> But, file_set_contents - which save_clusterfw_conf uses - does this >>> already[0], >>> so maybe this is the "high-level fuse rename isn't atomic" bug again... >>> May need to take a closer look tomorrow. > > mmm, ok. > > In

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Alexandre DERUMIER
>>But, file_set_contents - which save_clusterfw_conf uses - does this >>already[0], >>so maybe this is the "high-level fuse rename isn't atomic" bug again... >>May need to take a closer look tomorrow. mmm, ok. In my case, it was with a simple file copy (cp /tmp/cluster.fw

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Thomas Lamprecht
Hi, On 1/8/19 7:37 PM, Alexandre DERUMIER wrote: > I'm able to reproduce with: > --- > on 1 host: > > cluster.fw: > [OPTIONS] > > enable: 1 > policy_in: ACCEPT > > > > > #!/usr/bin/perl > > use IO::File; > use PVE::Firewall; > use Data::Dumper; > use Time::HiRes qw ( time

Re: [pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Alexandre DERUMIER
I'm able to reproduce with: --- on 1 host: cluster.fw: [OPTIONS] enable: 1 policy_in: ACCEPT #!/usr/bin/perl use IO::File; use PVE::Firewall; use Data::Dumper; use Time::HiRes qw ( time alarm sleep usleep ); while(1){ $filename = "/etc/pve/firewall/cluster.fw";

[pve-devel] firewall : possible bug/race when cluster.fw is replicated and rules are updated ?

2019-01-08 Thread Alexandre DERUMIER
Hi, I'm currently debugging a possible firewalling problem. I'm running some cephfs client in vm, firewalled by proxmox. cephfs client are really sensitive to network problem, and mainly with packets logss or dropped packets. I'm really not sure, but I have currently puppet updating my

[pve-devel] applied: [PATCH ifupdown2 0/3] bump to 1.2.2

2019-01-08 Thread Thomas Lamprecht
On 12/31/18 11:20 AM, Alexandre Derumier wrote: > This need to update mirror_ifupdown2 to tag 1.2.2-1 > > Alexandre Derumier (3): > Makefile : bump to 1.2.2 > changelog : dump to 1.2.2 > update config tuning patch > > Makefile| 2 +- > debian/changelog