Hitachi Vantara Pentaho Community Forums

View Poll Results: What integration method do you prefer for monitoring?

Voters
15. You may not vote on this poll
  • SNMP

    6 40.00%
  • JMX

    2 13.33%
  • HTTP (returning HTML or XML as it is today)

    9 60.00%
  • Other integration points

    2 13.33%
Multiple Choice Poll.
Results 1 to 7 of 7

Thread: Poll: Monitoring via SNMP / JMX / HTTP

  1. #1
    Join Date
    Nov 1999
    Posts
    459

    Default Poll: Monitoring via SNMP / JMX / HTTP

    For the new release we will add a couple of monitoring features and we would like to get your feedback to the following questions.

    These questions apply most to larger enterprises, but feel free to add any comment you have!

    1) What integration for monitoring with Kettle, Data-Integration- or Carte-Server would you prefer:
    a) SNMP
    b) JMX
    c) HTTP (returning HTML or XML as it is today)
    d) other integration points?

    2) What monitoring tools do you use or prefer?

    3) What metrics are you measuring or do you need to measure?

    Thanks a lot for helping with this!

  2. #2
    Join Date
    Apr 2007
    Posts
    2,010

    Default

    so, who voted jmx and why? What tools are you using that make jmx useful for monitoring?
    Also; As for "other integration points" such as..?

    I voted snmp as it works well with our existing nagios systems.

  3. #3

    Default

    I use Zenoss and struggle a bit with monitoring. I wrote a UDJC to use syslog using unix domain sockets rather than tcp. It's still not great though. It would be great if I had easy access to things like "facility" (see http://search.cpan.org/~saper/Sys-Sy...og.pm#EXAMPLES) and could wrap an entire job in a monitoring routine that would write to syslog with an indication of what failed. I'd like to avoid cluttering my jobs/steps with individual monitoring steps - so a job or transformation setting would be ideal that complains to syslog with a meaningful error would be nice.

    I use heartbeat with SNMP traps calling snmp from a shell step. Perhaps there are some better approaches - we'll be looking at monitoring in more detail over the next few weeks. I may have some additional details as an update.

    DB
    Last edited by dbron0000; 11-15-2011 at 10:40 PM.

  4. #4

    Default

    SNMP v2c and v3 on IP v4 and v6 would be very helpful, as many IT resources uses this for performance data collection.
    The solution should allow load of MIBs and configuration of authentication per polled interface.

  5. #5
    Join Date
    Apr 2008
    Posts
    4,696

    Default

    I would want to know a little bit more on the SNMP question - is this intended to be SNMP Polling (eg. Nagios queries the DI server every 5 seconds) or SNMP Trap (PDI sends a trap for ETL Start, ETL End)

    I like the Trap idea, I'm just not sure how that plays through.
    Perhaps we can build a list of requirements for what we want this to do, and submit it as a JIRA?

  6. #6
    Join Date
    Nov 1999
    Posts
    459

    Default

    Thanks all for the great insights. In combination from the customer feedback we got, we will go with supporting SNMP at first and see if the JMX user stories come in the future.
    Feel free to add your requirements to this JIRA http://jira.pentaho.com/browse/PDI-6909
    This will be more detailed soon.

    Thanks again!
    Jens

  7. #7

    Default

    Wondered if this is a dead request? Been checking in on it for some time now...

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.