Find the answer to your Linux question:
Results 1 to 5 of 5
Hi, I'm using GNU/Linux, how can I connect another Linux (same distro) and get stats like CPU, Disk, also using df? So that I don't have to go thru all ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
  1. 02-04-2013 #1
    Gideon2
    Gideon2 is offline
    Just Joined!
    Join Date
    Feb 2013
    Location
    Manila
    Posts
    17

    Question Using SAR to get stats from other linux machine

    Hi,

    I'm using GNU/Linux, how can I connect another Linux (same distro) and get stats like CPU, Disk, also using df? So that I don't have to go thru all 20 of them to check every morning.
    Reply With Quote Reply With Quote
  2. 02-04-2013 #2
    Irithori
    Irithori is online now
    Trusted Penguin Irithori's Avatar
    Join Date
    May 2009
    Location
    Munich
    Posts
    2,974
    Hi and welcome usually monitoring and graphing webtools are used. Have a look at e.g. nagios, xymon for monitoring and e.g. cacti, collectd for graphing.
    You must always face the curtain with a bow.
    Reply With Quote Reply With Quote
  3. 02-09-2013 #3
    Rubberman
    Rubberman is offline
    Linux Guru Rubberman's Avatar
    Join Date
    Apr 2009
    Location
    I can be found either 40 miles west of Chicago, or in a galaxy far, far away.
    Posts
    10,144
    Quote Originally Posted by Gideon2 View Post
    Hi,

    I'm using GNU/Linux, how can I connect another Linux (same distro) and get stats like CPU, Disk, also using df? So that I don't have to go thru all 20 of them to check every morning.
    I use sar, and process that data to push in real-time into a Hadoop/OpenTSDB/HBase database, which allows me to analyze performance data from 1000's of servers in real-time. IE, I have a data collector that takes data from SAR, converts that to OpenTSDB (OpenTSDB - A Distributed, Scalable Monitoring System) put commands, and sends it to OpentTSDB. 500 servers that collect SAR data in 1 minute intervals will generate about 350M data points per day, and this will let you chart your systems' performance (or problems) quite nicely. My Hadoop cluster (6 data nodes w/ 1TB each for data), can handle about 1 year of data from 500 nodes.
    Sometimes, real fast is almost as good as real time.
    Just remember, Semper Gumbi - always be flexible!
    Reply With Quote Reply With Quote
  4. 02-09-2013 #4
    Irithori
    Irithori is online now
    Trusted Penguin Irithori's Avatar
    Join Date
    May 2009
    Location
    Munich
    Posts
    2,974
    Wow, new toy
    Thanks for the info about OpenTsdb, looks promising.
    You must always face the curtain with a bow.
    Reply With Quote Reply With Quote
  5. 02-09-2013 #5
    Rubberman
    Rubberman is offline
    Linux Guru Rubberman's Avatar
    Join Date
    Apr 2009
    Location
    I can be found either 40 miles west of Chicago, or in a galaxy far, far away.
    Posts
    10,144
    Quote Originally Posted by Irithori View Post
    Wow, new toy
    Thanks for the info about OpenTsdb, looks promising.
    I've spent the better part of the past year working on this cruft. We are now in beta production (in 1 data center w/ 600 servers), and as soon as I configure a "real" production cluster, we will go into world-wide production data collection. With SAR, application stats, and SNMP data, each data center (4 or 5 world-wide) will generate about 1B data points per day, each. Our first DC went live in December, and is running like a clock! It has already saved more $$ than it costs in detecting system issues that our operations and network operations staff did not, such as servers that were running, but not getting any data loads, etc.
    Sometimes, real fast is almost as good as real time.
    Just remember, Semper Gumbi - always be flexible!
    Reply With Quote Reply With Quote
« Previous Thread | Next Thread »

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

Forum Rules