<< | [up] | >> |
dim_STAT User's Guide. by Dimitri |
Starting On-Line collecting |
Before starting any STAT collect, check first STAT-service is up and running on every server you want to monitor!!! At least you will be sure you are not in case of the most common error :))Another point - if you want to monitor any Linux server: be sure you're installed Linux STAT Add-Ons before start any collect (see special Linux section in this document).
Now from dim_STAT Main Page you may just follow Start New Collect link. (Note: since v.8.0 there is no more separation on single or multi host collect).
IMPORTANT:
- Any STAT collect for any host is independent of all others, so it can be stopped and/or restarted at any moment independently to others...
- Your collect options always saved into special script(s) with name based on the "Collect Base Name", so using customized names you may pre-load different set of options according your needs...
- Any collect you may start via browser on-line, or just make a starting script (to run by hand, via cron, as batch, etc.)
Main Steps |
There are 4 main points in Starting STAT Collect:
- - choose/give host name(s)
- - set collect attributes (title, id, etc.)
- - choice of collected statistics
- - start now, or prepare a script for manual/delayed execution
1.) Host name(s)
Since v.8.0 you choose host(s) first. You may easily setup a list of frequently used host names via 'Preferences' intrefece (host list). This list as well all other used host names are kept via browser coockies. Before you start any STAT collect, for each given host name tool will indicate you host STAT-service state by LED color - I hope it'll avoid any potential misconfiguration issues for new users for experimented as well. For the moment there are 3 LED colors:
- Red: host is not running STAT-service on the default port, or host is inaccessible from the network, or host is down, etc.
- Orange: host is running STAT-service but the old version
- Green: ok! STAT-service is running and has all needed features!
NOTE: since v.8.0 STAT-service has a new 'stat publish' feature, and tool knows exactly now which kind of STATs you may or not collect from each given host and protect you from choosing wrong or unavailable data.
2.) Set Collect Attributes
Collect BaseName -- all selected options are saved in special start script; the name of this script is composed of BaseName + some context extentions; when you start new collect next time you may pre-load previously selected options by giving previous BaseName + click on "Preload" (by default last given BaseName keeping via cookies in your browser)
Stat ID -- all data in database referenced to this ID, the ID is not assigned automatically to give a choice to use personalized range numbers (your project id; etc.)
Stat Title -- title description you give for starting collect
Time Interval -- frequency interval (in sec.) you want to be used by statistics programs (default 30sec is quite good in many cases)
Client Log File -- name of file on the "hostname" side you want to watch: all text lines appearing in this file will be automatically copied into STAT database and timestamped; during analyze of collected STATs you may at any time visualize all Log messages corresponding to analyzed interval (may be very useful to trace auto-starting jobs, night batches, etc...). As well they give you a simple and fast way to find a right time position during data analyzing (ex: show N minutes before/after/around of selected message).
STAT-service Port -- "hostname" port number on which STAT-service is running (by default tool will used the port number given during installation, and it's a good rule to use then the same port on every host to avoid complications :))
3.) Choose Statistics
Simply select all statistics you want (and may) to collect (help bullets showing a full description of each STAT (if you have JavaScript enabled in your web browser)). Please, think before to select - probably you don't need everything! :))
Generally good rule may be to start:
- VMSTAT
- MPSTAT
- IOSTAT
- netLOAD (avoid to use 'netstat')
- ProcLOAD
these STATs will already give you a quite useful general view on your system, and once having and analyzing them - you may go more in depth according needs.
4.) Start Mode
Make Start script only -- don't start collect, just make a script
Start Now! -- start the new STAT collect right now
Show Debug output -- in case you want to see debug messages from starting collect output...
Few screenshots... |
Choose STATs
You may see here several servers:
- neel, fourrier - Solaris hosts running upgraded STAT-service
- localhost - Linux box, upgraded STAT-service
- sting - Solaris host, old STAT-service
- fudji - Solaris host, powered off
I select neel, fourrier, localhost and sting and click on [Continue] button...
Choose STATs, next
So well, hosts chosen, let's select STATs to collect now!
You may remark for hosts:
- Linux stats are not proposed for any 'green' Solaris hosts
- Solaris stats are absent for Linux 'green' host selection
- not configured or disabled stats are not present for any 'green' host
- 'orange' host (sting) has all stats present and it's up to you to keep in mind which commands will run or not on this host (as it was before v.8.0)
Hope new feature will make pleasure to everybody ;-))
Load collect from output files |
If by any reason you cannot collect data directly from your hosts and all you have is a set of several statistics output files - you may still download them via Web interface as one STAT collect and analyze your data later. Just fill need parameters and go! :))However, if your output files representing a quite big volume - it may take much more time to load, and your browser may simply loose connection by timeout and you'll never see the final result...
In such cases it'll be better to vote for EasySTAT solution (simplified) or BatchLOAD (for more experimented users) - see following sections for more details...
Standalone configuration |
Before thinking about collecting stats via any kind of scripts, don't forget about 'standalone' dim_STAT possibility: there is _no_ restriction to:
- install dim_STAT on host A
- starts STAT-service on host A
- collect data from host A into host A
- be aware: on 4CPU machine (very small Sun machine) 20sec intervall collect of [vmstat + mpstat + iostat + psSTAT + netLOAD] will generate only 0.2% CPU Usage! (yes!)'Collecting' CPU usage of dim_STAT is very low. It highly uses CPU only during 'analyze' requests or global export/import/etc. actions. So, don't forget about such simple solution: install dim_STAT on the same host you want to collect from, collect locally all data you need, then simply backup whole collected database and restore it on the another machine for further analyze!
<< | [up] | >> |