[pve-devel] [PATCH 3/7] clone live vm : add support for multiple jobs

[Alexandre DERUMIER]

>>I wonder if we can make clone_vm more consistent when the VM 
>>is running? A correct workflow would be 

>>- suspend VM 
>>- start all mirror jobs 
>>- resume VM 
>>
>>Is it worth to implement that? 

Maybe it could work with simply pause before block-job-complete, then resume after block-job-complete ?

Maybe without pause, calling qga freezefs, before doing the block-job-complete could help too ?



----- Mail original -----
De: "dietmar" <dietmar at proxmox.com>
À: "Alexandre Derumier" <aderumier at odiso.com>, "pve-devel" <pve-devel at pve.proxmox.com>
Envoyé: Vendredi 30 Décembre 2016 10:28:42
Objet: Re: [pve-devel] [PATCH 3/7] clone live vm : add support for multiple jobs

I wonder if we can make clone_vm more consistent when the VM 
is running? A correct workflow would be 

- suspend VM 
- start all mirror jobs 
- resume VM 

Is it worth to implement that? 


> On December 28, 2016 at 9:24 AM Alexandre Derumier <aderumier at odiso.com> 
> wrote: 
> 
> 
> Signed-off-by: Alexandre Derumier <aderumier at odiso.com> 
> --- 
> PVE/API2/Qemu.pm | 12 +++++++++++- 
> 1 file changed, 11 insertions(+), 1 deletion(-) 
> 
> diff --git a/PVE/API2/Qemu.pm b/PVE/API2/Qemu.pm 
> index d0070a6..d21f3df 100644 
> --- a/PVE/API2/Qemu.pm 
> +++ b/PVE/API2/Qemu.pm 
> @@ -2443,21 +2443,29 @@ __PACKAGE__->register_method({ 
> my $upid = shift; 
> 
> my $newvollist = []; 
> + my $jobs = {}; 
> 
> eval { 
> local $SIG{INT} = $SIG{TERM} = $SIG{QUIT} = $SIG{HUP} = sub { die 
> "interrupted by signal\n"; }; 
> 
> PVE::Storage::activate_volumes($storecfg, $vollist, $snapname); 
> 
> + my $total_jobs = scalar(keys %{$drives}); 
> + my $i = 1; 
> + my $skipcomplete = 1; 
> + 
> foreach my $opt (keys %$drives) { 
> + 
> my $drive = $drives->{$opt}; 
> + $skipcomplete = undef if $total_jobs == $i; #finish after last drive 
> 
> my $newdrive = PVE::QemuServer::clone_disk($storecfg, $vmid, $running, 
> $opt, $drive, $snapname, 
> - $newid, $storage, $format, $fullclone->{$opt}, $newvollist); 
> + $newid, $storage, $format, $fullclone->{$opt}, $newvollist, $jobs, 
> $skipcomplete); 
> 
> $newconf->{$opt} = PVE::QemuServer::print_drive($vmid, $newdrive); 
> 
> PVE::QemuConfig->write_config($newid, $newconf); 
> + $i++; 
> } 
> 
> delete $newconf->{lock}; 
> @@ -2478,6 +2486,8 @@ __PACKAGE__->register_method({ 
> if (my $err = $@) { 
> unlink $conffile; 
> 
> + eval { PVE::QemuServer::qemu_blockjobs_cancel($vmid, $jobs) }; 
> + 
> sleep 1; # some storage like rbd need to wait before release volume - 
> really? 
> 
> foreach my $volid (@$newvollist) { 
> -- 
> 2.1.4 
> 
> _______________________________________________ 
> pve-devel mailing list 
> pve-devel at pve.proxmox.com 
> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel