ODAVP: How to monitor the guest domU performance

Asset ID:	1-71-1672694.1
Update Date:	2014-05-19
Keywords:

Solution Type Technical Instruction Sure

Solution 1672694.1 : ODAVP: How to monitor the guest domU performance

Applies to:

Oracle Database Appliance Software - Version 2.5.0.0 and later
Oracle Database Appliance - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.

Goal

It's important to remember as ODA VP (Virtualized Platform) is an environment capable of supporting complex virtualization configurations.
Deploying virtualization requires an understanding of the tools and techniques necessary to monitor the running environment, identify performance issues and track down problems.

Solution

How to monitor the guest domU performance

All commands described here are working from dom0 only.

- Lists information about the active guest VM domains

The status of the guest domain may be viewed at any time using the list option of the xm tool. For example:

# xm list
Name                                        ID   Mem VCPUs      State   Time(s)
Domain-0                                     0 4096    24     r----- 2418818.0
MsWin7                                       5 8192     2     -b----   6006.8
OL6U4_1                                      6 16384     4     -b----    515.1
oakDom1                                      1 49152    12     -b---- 990763.8
ol6u5                                        4 2048     2     -b----    225.6

where

Name is the name of the domain.
ID the numeric domain id.
Mem is the desired amount of memory to allocate to the domain (although it may not be the currently allocated amount).
VCPUs is the number of virtual CPUs allocated to the domain.
State is the run state (see below).
Time is the total run time of the domain

STATES
The State field lists 6 states for a guest domain, and which ones the current domain is in:

r - running
   The domain is currently running on a CPU.

b - blocked
    Whenever the guest is idle or when it is awaiting user input, the status shows blocked. It does not mean the VM is blocked

p - paused
    The domain has been paused. When in a paused state the domain will still consume allocated resources like memory, but will not be eligible for scheduling by the hypervisor.

s - shutdown
   The guest has requested to be shutdown, rebooted or suspended, and the domain is in the process of being destroyed in response.

c - crashed
    The domain has crashed, which is always a violent ending. Usually this state can only occur if the domain has been configured not to restart on crash.

d - dying
   The domain is in process of dying, but hasn't completely shutdown or crashed.

- Monitoring Guest Domain Performance with XenMon

The XenMon tool is useful for monitoring the performance guest domains, particularly when identifying which domains are responsible for the highest I/O or processing loads.
XenMon is started from the command-line using the xenmon.py command. The following figure shows a typical XenMon session:

xenmon

The XenMon display shows two sets of data. On the left hand side are statistics captured over the preceding 10 seconds and on the right is the data for the last 1 second.

For each domain three sets of data are provided.
   - The first row for each domain is the amount of time the domain as spent executing.
   - The Blocked row shows statistics for idle time.
   - The Waited row indicates the amount of time the domain has been in a wait state.
For each category
   - the amount of time spent in the particular mode
   - the time as a percentage of overall time during the corresponding period (i.e 1 or 10 seconds) is displayed.
The final value depends on the category:
   - Gotten this represents processor time
   - Blocked the average blocked time
   - Wait the average waiting time.

By default XenMon displays information for CPU 0. The p and n keys can be used to page through the data for each ODA CPU-core.

When XenMon is exited (using the q key) a summary of data collected during the monitoring session is displayed:

# xenmon.py
ms_per_sample = 100
Initialized with 24 cpu's
CPU Frequency = 3059.06
Event counts:
250026343    Other
00000000    Add Domain
00000000    Remove Domain
00000000    Sleep
21107314    Wake
20595388    Block
39645476    Switch
00000000    Timer Func
39645487    Switch Prev
39645480    Switch Next
00000214    Page Map
00000214    Page Unmap
00000000    Page Transfer
processed 410665916 total records in 496 seconds (827955 per second)
woke up 18439 times in 496 seconds (37 per second)

XenMon accepts a range of command-line options which control various aspects of the monitoring. To get information of these options simply pass the --help argument through to xenmon.py:

# xenmon.py --help
usage: xenmon.py [options]

options:
-h, --help            show this help message and exit
-l, --live            show the ncurses live monitoring frontend (default)
-n, --notlive         write to file instead of live monitoring
-p PREFIX, --prefix=PREFIX
                        prefix to use for output files
-t DURATION, --time=DURATION
                        stop logging to file after this much time has elapsed
                        (in seconds). set to 0 to keep logging indefinitely
-i INTERVAL, --interval=INTERVAL
                        interval for logging (in ms)
--ms_per_sample=MSPERSAMPLE
                        determines how many ms worth of data goes in a sample
--cpu=CPU             specifies which cpu to display data for
--allocated           Display allocated time for each domain
--noallocated         Don't display allocated time for each domain
--blocked             Display blocked time for each domain
--noblocked           Don't display blocked time for each domain
--waited              Display waiting time for each domain
--nowaited            Don't display waiting time for each domain
--excount             Display execution count for each domain
--noexcount           Don't display execution count for each domain
--iocount             Display I/O count for each domain
--noiocount           Don't display I/O count for each domain

- Monitoring Performance with XenTop

Everyone coming from UNIX or Linux know the top command. Top is used to display information, such as CPU and memory usage, about processes running on a particular system. One of the best features of top is that is puts the process making the heaviest use of a particular resource at the top of the list. When a system is exhibiting performance degradation the top command is often the first check an experienced system administrator is doing.

XenTop is essentially a hypervisor version of the original top utility and is used to show information about all the domains running on a particular system. The XenTop tool is launched by typing xentop as root at the dom0 command-line. It is worth knowing that a range of options are available and can be listed using the --help flag

# xentop -h
Usage: xentop [OPTION]
Displays ongoing information about xen vm resources

-h, --help           display this help and exit
-V, --version        output version information and exit
-d, --delay=SECONDS seconds between updates (default 3)
-n, --networks       output vif network data
-x, --vbds           output vbd block device data
-r, --repeat-header repeat table header before each domain
-v, --vcpus          output vcpu data
-b, --batch         output in batch mode, no user input accepted
-i, --iterations     number of iterations before exiting
-f, --full-name      output the full domain name (not truncated)

The following is an example of how xentop is working:

xentop

Attachments

This solution has no attachment