<div dir="ltr"><div class="gmail_extra"><div class="gmail_quote">On Wed, Mar 1, 2017 at 1:40 PM, Chris Hutten-Czapski <span dir="ltr"><<a href="mailto:chutten@mozilla.com" target="_blank">chutten@mozilla.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div>So, for each channel, having a line for each of the current and N previous versions' crash rates would be helpful? (where N is small... say 2)<br></div></div></div></blockquote><div><br>It's closer. If it's possible to have a single line that is the aggregate, that might smooth out adoption noise.<br> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><br><div>crash_aggregates does its aggregation by version, not build, so if crash-rates-per-build is necessary this will require quite a rewrite. (I'll have to join main_summary and crash_summary on dates) If crash-rates-per-version is sufficient (or is worth the effort of exploring), then that can be adopted within the existing architecture.<br></div></div></div></blockquote><div><br></div><div>This is incorrect. build_id is one of the dimensions, specifically because we needed it for betas with e10s. Example from stmo:<br><br>activity_date 
      
        dimensions 
      
        stats 
      
        submission_date 
      
    

    
    
    
      24/08/16{"build_id":"<wbr>20150804030204","os_name":"<wbr>Windows_NT","os_version":"6.2"<wbr>,"country":"NP","application":<wbr>"Firefox","architecture":"x86-<wbr>64","build_version":"42.0a1","<wbr>channel":"nightly","e10s_<wbr>enabled":"False"}{"usage_<wbr>hours_squared":0.<wbr>01585779320987654,"main_<wbr>crashes":0,"content_shutdown_<wbr>crashes":0,"usage_hours":0.<wbr>15416666666666667,"content_<wbr>shutdown_crashes_squared":0,"<wbr>gmplugin_crashes":0,"content_<wbr>crashes":0,"content_crashes_<wbr>squared":0,"plugin_crashes":0,<wbr>"plugin_crashes_squared":0,"<wbr>ping_count":3,"gmplugin_<wbr>crashes_squared":0,"main_<wbr>crashes_squared":0}25/08/16<br><br><br></div><div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div><br></div><div>One thing to consider may be that crash-rates-per-build will look messy on anything pre-Beta. <br></div></div></div></blockquote><div><br></div><div>Could be, but here's an example way to present this for nightly: <a href="http://benjamin.smedbergs.us/blog/2013-04-22/graph-of-the-day-empty-minidump-crashes-per-user/" target="_blank">http://benjamin.smedbergs.us/<wbr>blog/2013-04-22/graph-of-the-<wbr>day-empty-minidump-crashes-<wbr>per-user/</a><br><br></div><div>Here's also a draft of something a while back in STMO: <a href="https://sql.telemetry.mozilla.org/queries/192/source#309">https://sql.telemetry.mozilla.org/queries/192/source#309</a><br><br></div><div>--BDS</div></div><br></div></div>