Comments

  • 1-10 of 10
  • Makarand Malandkar

    Really useful post. Thanks for sharing Dan.

  • Makarand Malandkar

    Hi Nagendra,

    Please review the out of office accelerator https://cx.rightnow.com/app/answers/detail/a_id/5436.

    Although this is not exactly what you are looking for but something very close to it.

     

    Regards,

    Makarand

  • Makarand Malandkar

    Thank you all for your insights/suggestions. We are already following the suggestions provided by Luis. But we have 1.4 million contacts and around 70 SLAs and there is no restriction i.e. A contact can have any of the 70 SLAs available. So every time an SLA is modified we need to update large volume of contacts. And that is where we thought of using kind of dummy Organization to link all contacts to single Organization so we just need to update one record. But then the recommendation to limit number of contacts assigned to organization is potential blocker there. So we created contact and case workspace removing all the reference to Organization field and also did some tests by assigning over 25000 contacts to a single Org record. We didn't find any performance issues with loading of console as advised in this article. And this is where I am looking for inputs from experts/Product or Community managers that if the org field is not part of workspace then will it really slow down console even if there are over million contacts assigned to that Org.

    From product feature point of view, I think simply having ability from incident business rule to terminate contact/Org SLA instance would make this process dynamic. We can have rules in following order whenever case is created or updated:

    1. Terminate existing SLA instance
    2. Apply SLA insance with create instance if one does not exist checkbox ticked.

    Regards,

    Makarand.

  • Makarand Malandkar
    Edson Junior said:

    Makarand,

    Very interest discussion. Thank you for bring this point, and here are my contribution.

    The following answer is not an Oracle documented best practice or recommendation. I just like to share here few experience which help me find and troubleshoot performance issue with other customer in the past. Hope this help.

    Workspace load and save Performance is a combination of three essential facts, which are Client (workstation capacity), Network (network throughput) and Server (response time). In most case I’ve experienced issues were found in client and network such as bad configuration / customization, complex network firewall rules (remove the proxy to verify the differences), and/or not enough hardware to attend a customer’s requirement (complex workspace rules).

    How the project team found bottlenecks to make significant improvements to the implementation?

    Essential the project team want to use free tools in order to help a project implementation, for this reason the following 2 tools were used:

    • PSR (Problem Step Record): PSR is a free windows tools (win + R and type “psr”). With this tools the project team was able to record all agent steps and times spent for each click. The project team wants to identify processes and navigation flow improvements. In addition, it will help to measure the overall time.
    • Fiddler: With fiddler the project team were able to identify the network times, between post and gets. Remember OSvC is a SaaS application, where data back and forth between client and network.

    PSR has a result a .mht file with all screen and time captured, and Fiddler has a result a .SAZ file with traffic network information. With these two files the project team used an spreadsheet and an equation to understand the client, network and server time. Here are the equation used. 

    • Client = (PSR Total (Start - End) – Fiddler Times) * Total of Server + Total of Network + Total of Internal Time
    • Network = (ServerGotRequest - ClientDoneRequest) + (ServerDoneResponse – ServerBeginResponse)
    • Server = (ServerBeginResponse - ServerGotRequest)

    The equation was created based on the fiddler documentation (recommendation), in which I wanted to summary in the attached img.

    As a conclusion, if you use results provided by PSR and Fiddler you will able to identify your bottleneck and potential issues to make an improvement.

    Thank you, Edson

    View original

    Thanks Edson for the detailed approach. We used fiddler but will have to look into PSR tool. I would also like Oracle to provide filtering on the VCIO average admin page load report, which can make the ongoing monitoring lot easier.

    Regards,

    Makarand.

  • Makarand Malandkar

    A masterpiece recipe. Not sure if anyone else experienced it but I noticed a strange behavior for output column "% of average" and it wasn't showing correct value. This could be due to difference of data type of average incidents (decimal)  and # Incidents Today (number). After truncating the average incidents column the % was calculated as expected.

    Thanks Bastiaan for your great recipes/posts, they save the day for us.

    Regards,

    Makarand.

  • Makarand Malandkar

    Hi Erica,

    Links in this post are broken. Any chance you can point to me relevant content?

    Regards,

    Makarand.

  • Makarand Malandkar

    I hope you have the profile permissions setup correctly. In addition the following answer might help.

    https://cx.rightnow.com/app/answers/detail/a_id/5884/kw/cobrowse%20desktop/related/1

    Regards,

    Makarand.

  • Makarand Malandkar

    Great. Thanks for the update Keri.

    Regards,

  • Makarand Malandkar

    Hi Keri,

    Any further progress on this?

    Regards,

    Makarand.

  • Makarand Malandkar

    Thank you so much VIP and Keri for your responses. Keri, the document is perfect, exactly what I was looking for.

    Regards,

    Makarand.