I have over 1200 machines chacking in to my K1200S, and I want to be certain that I can detect all patches on all computers safely. Can this have an impact on network bandwidth? Will the end user see slowdowns? This is the very first time for these machines to be detected.

Answer Summary:
"With 1200 machines, I would recommend breaking it down by labels." - ShawnCarson "[...]instead of having one monolithic job of 1200 machines, it would be 4-5 different jobs, defined by your machine labels. Ideal target for a job would be in the neighborhood of 250 machines." - ShawnCarson
Cancel
1 Comment   [ + ] Show Comment

Comments

  • You're not using patch labels? I haven't tested it, but I'm pretty sure detecting all patches would slow down everything.
    • I haven't created ones yet. I have been going through all of the KKE videos and the presenter suggested Detect ALL on ALL. Given that our numbers are so high, i guess it would impact speed. But it is only checking the signature files, right? With those being so small, could it work? Or should I detect specific machine labels, or patch labels? Also, does detect use the replication shares for signature files or is that direct to the K1000?
      • I would use patch labels. I only download patches that are associated to my patch labels to save on space. If you were going to do "all and all" I would only do it on a couple of machines.
  • This content is currently hidden from public view.
    Reason: Removed by member request
    For more information, visit our FAQ's.
Please log in to comment

Community Chosen Answer

5

usually you don't see relevant slowdown, except you have already a slow network.
he detect may run long on the clients but usually there is no problem

Answered 08/06/2013 by: Nico_K
Red Belt

  • if you see an impact, simply run the detects in different runs for smaller groups.
  • easterdaymatt,

    Nico is correct here. Detections are generally not resource-intensive on the K1000, your network, or the machines processing them. You should, easily, be able to run one Detect job on 1200 machines simultaneously. If you really need confirmation, you could always try a selective sample while running a network sniffer, and perfmon on the machines.

    Ron Colson
    KACE Koach
Please log in to comment

Answers

2

With 1200 machines, I would recommend breaking it down by labels.  If you do this, you can still patch all of your machines but instead of having one monolithic job of 1200 machines, it would be 4-5 different jobs, defined by your machine labels.  Ideal target for a job would be in the neighborhood of 250 machines.

Answered 08/07/2013 by: ShawnCarson
White Belt

Please log in to comment
Answer this question or Comment on this question for clarity