Backup tooks so long to finish

G

Gilberto Ferreira

Renowned Member
Aug 21, 2014
371
50
93
Joinville, Santa Catarina, Brazil, Brazil
Hello
I have a VM with Windows 2012 running nicely in PVE 3.2.
This VM has a 300 GB disk size that was re-partition with 5 or 6 different parts, in order to allocate different databases...
The issue is that the backup task takes 6 hours to generate a lzo file with 13 GB size!
However, I have another VM, which is Windows 2008 with disk 120 GB size.
The backup makes 74 GB file of size, and takes approximately 30 minutes!!!
I can't see why the first task took so long time to finish!!!
Can anybody assist me??
Thanks
 
Hello Gilberto


Gilberto Ferreira said:
I have a VM with Windows 2012 running nicely in PVE 3.2.
This VM has a 300 GB disk size that was re-partition with 5 or 6 different parts, in order to allocate different databases...
The issue is that the backup task takes 6 hours to generate a lzo file with 13 GB size!
Click to expand...

What kind (snapshot, suspend, stop) of backup did you make?

In case of snapshot:

did you have a lot of disk accesses in the guest during backup? In that case the backup is extremely slow - in oder to demonstrate it an example. I started a backup during a lot of disk accesses
in mode "snapshot":

Code: 
INFO: starting new backup job: vzdump 700 --remove 0 --mode snapshot --compress lzo --storage ebu --node ebux
INFO: Starting Backup of VM 700 (qemu)
INFO: status = running
INFO: update VM 700: -lock backup
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: snapshots found (not included into backup)
INFO: creating archive '/3/eb3/dump/vzdump-qemu-700-2014_08_28-09_23_59.vma.lzo'
INFO: started backup task '73eaff0c-4236-4acd-b272-16540fad2684'
INFO: status: 0% (31391744/273804165120), sparse 0% (24522752), duration 3, 10/2 MB/s
INFO: status: 1% (2882666496/273804165120), sparse 0% (2734432256), duration 77, 38/1 MB/s
INFO: status: 2% (5511446528/273804165120), sparse 1% (5289598976), duration 119, 62/1 MB/s
INFO: status: 3% (8365015040/273804165120), sparse 2% (7833141248), duration 190, 40/4 MB/s
ERROR: interrupted by signal
INFO: aborting backup job
ERROR: Backup of VM 700 failed - interrupted by signal
ERROR: Backup job failed - interrupted by signal
TASK ERROR: interrupted by signal

After 3 minutes I gave up - now the same in "suspend" mode:

Code: 
INFO: starting new backup job: vzdump 700 --remove 0 --mode suspend --compress lzo --storage ebu --node ebux
INFO: Starting Backup of VM 700 (qemu)
INFO: status = running
INFO: update VM 700: -lock backup
INFO: backup mode: suspend
INFO: ionice priority: 7
INFO: suspend vm
INFO: snapshots found (not included into backup)
INFO: creating archive '/3/eb3/dump/vzdump-qemu-700-2014_08_28-09_48_17.vma.lzo'
INFO: started backup task 'd64bef30-9a84-450b-8b9c-a7bdae75e8a4'
INFO: status: 0% (1740242944/273804165120), sparse 0% (1644736512), duration 3, 580/31 MB/s
INFO: status: 1% (3126198272/273804165120), sparse 1% (2977558528), duration 6, 461/17 MB/s
INFO: status: 2% (5578620928/273804165120), sparse 1% (5354336256), duration 12, 408/12 MB/s
INFO: status: 3% (8314552320/273804165120), sparse 2% (7782805504), duration 20, 341/38 MB/s
........
INFO: status: 44% (120572018688/273804165120), sparse 41% (113780695040), duration 189, 156/70 MB/s
........
INFO: status: 100% (273804165120/273804165120), sparse 94% (257679671296), duration 425, 4026/0 MB/s
INFO: transferred 273804 MB in 425 seconds (644 MB/s)
INFO: archive file size: 10.86GB
INFO: resume vm
INFO: vm is online again after 427 seconds
INFO: Finished Backup of VM 700 (00:07:07)
INFO: Backup job finished successfully
TASK OK

You see it´s more than 10 times faster.

There can be other reasons too: network disk access, processor overload, etc. Difficult to say without more information. Observe the backup log and (in host) the CPU load, memory usage and process lzop with top during backup:

Code: 
root@ebux:~# top

top - 09:24:31 up 5 min,  1 user,  load average: 0.51, 0.28, 0.11
Tasks: 278 total,   1 running, 277 sleeping,   0 stopped,   0 zombie
%Cpu(s):  0.7 us,  0.3 sy,  0.0 ni, 95.5 id,  3.5 wa,  0.0 hi,  0.0 si,  0.0 st
KiB Mem:   8081628 total,  1943572 used,  6138056 free,     7244 buffers
KiB Swap:   917496 total,        0 used,   917496 free,   184772 cached

    PID USER      PR  NI  VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND                                                                                         
   7325 root      20   0 3456m 1.2g 3964 S   4.3 15.6   0:25.14 kvm                                                                                             
   4611 www-data  20   0  284m  60m 3760 S   1.0  0.8   0:00.40 pveproxy worker                                                                                 
   4584 root      20   0  283m  60m 6288 S   0.7  0.8   0:00.45 pvedaemon worke                                                                                 
   4605 root      20   0  193m  36m 2524 S   0.7  0.5   0:00.47 pvestatd                                                                                        
   7479 root      20   0  6828 1244  464 S   0.7  0.0   0:00.13 lzop                                                                                            
   4293 root      20   0  164m 1656  892 S   0.3  0.0   0:00.03 rrdcached                                                                                       
      1 root      20   0 10604  836  704 S   0.0  0.0   0:00.82 init

Kind regards

Mr.Holmes
 
You must log in or register to reply here.
Top Bottom