Feature #3718
Get disk space usage (real and virtual)
| Status: | Closed | Start date: | 03/24/2015 | |
|---|---|---|---|---|
| Priority: | Normal | Due date: | ||
| Assignee: | - | % Done: | 0%  | |
| Category: | Drivers - VM | |||
| Target version: | Release 4.14 | |||
| Resolution: | fixed | Pull request: | 
Description
File based storage not always use the maximum virtual size of the disk. For example qcow2 or sparse raw files.
Create a probe that gets this information from the VM disks.
Related issues
Associated revisions
feature #3718: add disk usage to kvm probe
feature #3718: add disk usage to xen probe
feature #3718: Add disk_cost, use disk_actual_size (needs to revisit)
feature #3718: Rename update_info method to set_deploy_id
feature #3718: New template for MONITORING data, it can accomodate any
data, and support the general OpenNebula Template format.
feature #3718: Added missing include file
Feature #3718: replaces old monitoring disk attrs names with new ones
Feature #3718: Change NET_TX, NET_RX, CPU and MEMORY for all hypervisors
Feature #3718: Use new disk probes for Xen
Feature #3718: Fixes the poll attribute definition
for locally executed monitor drivers
Feature #3718: Replace NET_TX and NET_RX w/ NETTX and NETRX
in the monitorization probes for VMs
Feature #3718: Update xpaths for the new probes in onevm and oneacct
Feature #3718: Display size of disk and snapshots in onevm
Feature #3718: top output has changed
Feature #3718: fix consistency for disk size output
feature #3718: Compute disk total costs in showback
feature #3718: Updated dummy driver to new monitor names
Feature #3718: Fix dummy probe reporting for snapshot_size
Feature #3718: Use Monitoring data for snapshot size in cli commands
Feature #3718: Remove TOTAL_DISK_SIZE from the probes
Feature #3718: remove total disk size from onevm and calculate it for oneacct
feature #3718: Fix wrong variable name
Feature #3718: Rename oneacct TOTAL_DISK_SIZE column to DISK
Feature #3718: Update oneacct.yaml columns
Feature #3718: disable i18n in the monitorization poll
Feature #3718: Move monitoring attributes in onedb upgrade
Feature #3718: Update flow grammar to look for monitoring attributes
feature #3718: Show MONITORING info in ovevm show
    TODO: Remove CPU, MEMORY, NETTX, NETRX, DISK_SIZE from MONITORING
feature #3718: Fix external IPs path
feature #3718: Fix paths for monitoring graphs
feature #3718: Fix cpu and memory monitoring in vms table
feature #3718: Show monitoring info in vm info panel
feature #3718: Add Size to the disks table in the vm storage panel
Feature #3718: Implement part of the DB upgrade with oned running
Feature #3718: Monitor regular disk snapshots
Feature #3718: Add hybrid hosts monitoring attr. to onedb upgrade files
Feature #3718: Remove USEDCPU from monitoring docs
feature #3718: Remove STATE from monitor values
feature #3718: Add DISK on oned.conf comment. DISK_COST is now a restricted
attribute
Feature #3718: Add units to the onevm show disk and snapshot size columns
Feature #3718: Clean the output of onevm show
Feature #3718: Change ESX_HOST attribute location for VNC
Feature #3718: Display the monitorization info in onevm show
Feature #3718: Remove snapshots as well as main image for DS_MAD=fs
feature #3718: Reduce monitoring info in the monitoring DB table
feature #3718: VM disk tables width (less than 80c)
Feature #3718: Scheduler takes snapshot size into account
History
#1
    
    Updated by Javi Fontan over 6 years ago
    - Related to Backlog #2626: Monitor datastores IO added
 
#2
    
    Updated by Javi Fontan over 6 years ago
    - Related to Feature #2921: (Per VM) DISKIO IO information in Sunstone added
 
#3
    
    Updated by Javi Fontan about 6 years ago
    - Status changed from Pending to Closed
 - Resolution set to fixed
 
#4
    
    Updated by Ruben S. Montero about 6 years ago
    - Status changed from Closed to New
 - Resolution deleted (
fixed) 
Needs to
- Implement this in vCenter Probe
 - Include support in oned to parse the disk usage information
 - Include this information in history records
 - Evalute adding it to showback/accounting
 - Add Sunstone support for viewing this metric
 
#5
    
    Updated by Javi Fontan about 6 years ago
    - Status changed from New to Closed
 - Resolution set to fixed
 
Probes now return DISK_ACTUAL_SIZE and DISK_VIRTUAL_SIZE for file based images in KVM and Xen.
#6
    
    Updated by Ruben S. Montero about 6 years ago
    - Status changed from Closed to New
 - Assignee deleted (
Javi Fontan) - Resolution deleted (
fixed) 
#7
    
    Updated by Ruben S. Montero about 6 years ago
    Implement this for VMware
#8
    
    Updated by Ruben S. Montero about 6 years ago
    Review behavior on LVM, Ceph and DEV datastores
#9
    
    Updated by Ruben S. Montero about 6 years ago
    Commits:
#552e6a8f96cca917a191d1eab6aa872b9472e19a
#552e6a8f96cca917a191d1eab6aa872b9472e19a
Needs migrator to add <DISK_ACTUAL_SIZE> <DISK_VIRTUAL_SIZE>
Needs to update Sunstone views for accounting
#10
    
    Updated by Ruben S. Montero about 6 years ago
    Added <DISK_COST> to <SHOWBACK_RECORD> it may require migrator.
Sunstone needs DISK_COST in templates.
Currently DISK_ACTUAL_SIZE is used as a place holder, but DISK_VIRTUAL_SIZE can be used also. Moreover we need to deal with situation where snapshosts are removed within the same history record window. Probably, either add a new record (as we do for attach) or if snapshoting remains only for poweroff that will add the record.
#11
    
    Updated by Ruben S. Montero about 6 years ago
    Probes needs to use the same names, and use for known/required information:
POLL_ATTRIBUTE = {                                                                                           :usedmemory  => "USEDMEMORY",                                                                            :usedcpu     => "USEDCPU",                                                                       
        :nettx       => "NETTX",                                                                         
        :netrx       => "NETRX",                                                                         
        :state       => "STATE",                                                                         
        :disk_actual => "DISK_ACTUAL_SIZE",                                                                      :disk_virtual=> "DISK_VIRTUAL_SIZE"                                                                  }#12
    
    Updated by Ruben S. Montero about 6 years ago
    Needs to update xsd's
Ruben S. Montero wrote:
Probes needs to use the same names, and use for known/required information:
POLL_ATTRIBUTE = { :usedmemory => "USEDMEMORY", :usedcpu => "USEDCPU",
:nettx => "NETTX",
:netrx => "NETRX",
:state => "STATE",
:disk_actual => "DISK_ACTUAL_SIZE", :disk_virtual=> "DISK_VIRTUAL_SIZE" }
#13
    
    Updated by Jaime Melis about 6 years ago
    New parameters:
DISK_SIZE = [ ID=<disk_id>, SIZE = <actual_size_in_mb> ]
SNAPSHOT_SIZE = [ ID=<snapshot_id>, DISK_ID=<disk_id>, SIZE = <actual_size_in_mb> ]
TOTAL_DISK_SIZE = <sum of the sizes of all disks and all snapshots in MB>
Use this ticket to update these VM probes too:
USEDMEMORY => MEMORY
USEDCPU => CPU
#14
    
    Updated by Ruben S. Montero almost 6 years ago
    - Status changed from New to Closed
 - Resolution set to fixed