Comments

  • Charlie Mopps

    Perfect, thank you! I'll try and piece it together from there.

  • Charlie Mopps

    The ROQL request isn't an issue. Writing the chaining in C# is the issue. I've been playing around with the demo code but the way those source and destination variables work is what's confusing me. Do you have any links/examples of an ROQL request that provides the data required for the update? This whole thing is building an XML request that somehow tells the API "Make the results of X the input of Y" and that XML is my real goal.

  • Charlie Mopps

    I think you're actually making our point for us. You've got 5 customers in this thread that all had the same issue, at the same time, on the same day. What exactly are the criteria for issuing an SEO?

    Are there any plans to move Rightnow support into the Oracle support structure any time soon?

     

  • Charlie Mopps

    There is an Analytics audit log table: ac_audit_log

    It does not appear to log the id of the report correctly, so I'm not sure of how much use that would be to you. You should test it out yourself, you might be on a different version of the product where it works better. If not you could also try opening a ticket with support to get the table fixed. If the table worked properly you could querry all records of the type "Generated" from that table, for that user, and that report id and that'd give you a count.

     

  • Charlie Mopps

    Did you ever get an answer on this? We had the same sort of thing happen to us yesterday, and they won't tell us what happened. They wouldn't submit a VCIO, but then they said they performed some maintenance last night to fix it, and now wont tell us what the outage was about because there was no VCIO. I've no idea if whatever happened was actually resolved or not.

  • Charlie Mopps

    Ohh... it's in the SOAP Developer guide?

    So this brings up another point. If it's in there... it's completely unsearchable. From your website, from google, it just doesn't exist. As well as any other information in the guide for that matter. That explains why I couldn't find it. If it were a standard HTML page, I could have found it with ease.

    Ok, I'm done bothering you now. I can read through it now that I know where it's at. Thanks!

  • Charlie Mopps

    ah... see, I had no idea that even existed.  How do I access those notes? I can't find them anywhere on your site. I even did multiple searches and nothing comes up. I can't find them anywhere.

  • Charlie Mopps

    This provides an excellent opportunity for me to help Oracle/Rightnow understand some of the difficulties we have with your documentation.

    Here's a link to your release notes: https://documentation.custhelp.com/euf/assets/docs/may2014/olh/wwhelp/wwhimpl/js/html/wwhelp.htm?href=May2014_release_notes.3.1.html

    Without using whatever internal tools you have, try and read through that and tell me where and how I'd find this information. I found 2 mentions of the "DESCRIBE" function in the entire document. On one, it's shown in a screenshot of a help window. In the other it's in an example query. I can't find any mention of "We added DESCRIBE to ROQL!" at all. This is the only direct mention of it that I can find: https://documentation.custhelp.com/euf/assets/docs/may2014/olh/wwhelp/wwhimpl/common/html/wwhelp.htm?context=crm&file=admin.73.27.html

    And I know exactly what I'm looking for.

    ...and this is assuming I'm upgrading to May 2014. We didn't. We skipped to a newer one, never touched this release.... When we upgrade I have to collate a list of all of the updates from all of the releases between our current version and the one we're upgrading to. Given the way these release notes are written, there's no way I could ever find something like this. It's just not possible.

    We use a lot of software packages from a lot of other sources, including Oracle, and most release notes read like a bullet pointed list:

    http://dev.mysql.com/doc/relnotes/mysql/5.6/en/

    http://docs.oracle.com/cd/E11882_01/relnotes.112/e23558/toc.htm#CHDIAEBD

    This is what we need. You sort of have a "Major changes" section that's bullet pointed, but that often leaves out some fairly major enhancements like this one.

     

     

     

  • Charlie Mopps

    I figured this out on my own. I wish there was a way to see patch notes or something... or get on a mailing list for this sort of thing.

  • Charlie Mopps

    My problem is that when it hits 10,000 records, the report throws a join error and doesn't return any data. Further, I have over 10 years of data. Returning that in 1min increments isn't going to work. We're only allowed so many API calls per our contract. I'm not sure how valuable 10yrs worth of transaction data is so I'm currently trying to get the time frame requirements reduced.

  • Charlie Mopps

    Unfortunately I've personally tested Progress Direct offering... and no, it would definitely not help with this situation.

  • Charlie Mopps

    I want to know which contacts are currently associated with the incident, and which have changed, without having to go through all 8 million rows.

    Am I correct in assuming the only way to do this is look at incidents.updated? And just know that after an incident is updated, I need to drop and re-import every related inc2contact record that has the same i_id?

  • Charlie Mopps

    This is part of our ODBC replacement project. So we're trying to replicate the entire table locally. It's obviously likely that our queries crash at some point during the 100 million+ rows of initial load. So we need a unique key so we won't duplicate data.

    Since there isn't a unique key, and there's no way to get a unique key, we're going to declare the entire row as unique, and if it happens to not be unique we'll just let it error on the unique constraint and toss out the duplicate row (did we really need it anyway?)

    Unfortunately the size of the primary key when we concat the entire row is problematic. So we thought to reduce its size with an MD5 hash. This isn't perfect, a 32 bit hash isn't entirely unique, but it's mathematically unlikely that we'd get a collision.

    Unfortunately, any change you were to make in a future release as a result of something posted to the idea lab isn't going to help us. You're turning off ODBC in November and we're in a time crunch. We have reporting that cannot lose access to the data. If there's no way to add a unique key, or fire a php script to create one, we're going to just have to accept that this is going to be messy and let our reporting folks know that the data is not perfect.

  • Charlie Mopps

    MB said:


    Jerry,




    I had to hard code Created and Closed but am now being asked for a way to allow a date range to be entered for the cases created and cases closed




    I imported your report def but not sure if I understood it clearly especially what you meant by "hardcode created and closed...". An example would help 



    Thanks,



    MB


    View original


    You replied to the wrong post. JLFolkman was the one that gave you the report ;-)

  • Charlie Mopps

    The end goal of this is to copy the entire database. So I need to get ALL fields on ALL objects. I'm just trying to get it working with a single record. In the end this will be a 1 time iterative loop through the entire database followed by by periodic lookups for updated records.

    In this case, the Primary object I'm querying is the opportunities table. This specific item is the custom field tds_lead_src. Depending on which part of rightnow I'm in that field can be called "tds_lead_src" or "c$tds_lead_src" I've tried both in this code and they both fail. In the end, I want the entire row, with all fields, as it appears in the application.

    I've found a way to dynamically build namedID's for primary objects so it returns all that data correctly. The only thing I'm having trouble with are the custom fields. Doing a Get request is undesirable, because of the number of records we're talking about.

    If we had access to the labels table like we do with ODBC, I'd just copy that as well and do the lookups via joins. But it appears Lables has been turned into something different that I can't query normally. An alternative solution, if it's possible, would be to get a blanket copy of the Labels table so we could do our joins externally.

    Barring all that, we'll have to build this entire thing with QueryCSV, but the CSV format is really messy to parse.

    We'd prefer just to keep ODBC honestly.