In Ninja's defense, they cannot fix an issue that they can't replicate.
While I do believe this problem is real and not imagined, the circumstances that cause it have been impossible to replicate at will. This means I have to wait for this intermittent problem to happen again before I can take another look, and limits the guidance I have been able to offer, to purely "symptomatic" .... which is showing on my charts that I occasionally get a 'false' IsFirstBarofSession boolean when it should be true and making that code available to NT Support (along with log/trace files).
So far none of these have pinpointed the underlying cause....
At one point, I thought I had this issue trapped....and sent my workspace to Ninja Support....only to find that the issue which clearly show up on charts at 9am CDT, cleared itself when Ninja loaded it up and tried to test that evening. When I went back to that PC and popped the workspace up again....it had corrected on my PC as well.
Whatever is going on is clearly transient, which makes it very difficult to isolate.
Additionally, the fact that NinjaSupport does not...(well..at least does not from my limited perspective)....seem to be buried in similar problem tickets from other users, implies that the circumstances are quite specific.
My anecdotal data suggests the problem only occurs when loading loading multiple days data on a PC that hasn't been attached to a live data stream for a while, running a workspace with multiple charts all running indicators with secondary dataseries...(perhaps some 'lock' contention on the update threads ... who knows)....which sometimes causes only one...(usually yesterday's)...IsFirstBarofSession boolean to flag 'false' on a bar that is clearly the beginning bar of a new session.
Making this issue even less clear cut, the same chart with the one 'false' boolean, does shows the 'session break' line...(which implies that at NT8 code 'found the session break').... Why that one single IsFirstBarofSession boolean returns false has me flummoxed. Why it seems to correct itself later in the day (sometimes) and corrects on the next new day (mostly) and will always eventually correct if I delete the db cache and reload data, also has me flummoxed.
I imagine this is not quite the answer you hoped for....but so far it is the best I've been able do. Hope springs eternal that something will "click" and I can isolate what is going on....or.....that some one of the other changes/corrections Ninja installs, will also fix this.
While I do believe this problem is real and not imagined, the circumstances that cause it have been impossible to replicate at will. This means I have to wait for this intermittent problem to happen again before I can take another look, and limits the guidance I have been able to offer, to purely "symptomatic" .... which is showing on my charts that I occasionally get a 'false' IsFirstBarofSession boolean when it should be true and making that code available to NT Support (along with log/trace files).
So far none of these have pinpointed the underlying cause....
At one point, I thought I had this issue trapped....and sent my workspace to Ninja Support....only to find that the issue which clearly show up on charts at 9am CDT, cleared itself when Ninja loaded it up and tried to test that evening. When I went back to that PC and popped the workspace up again....it had corrected on my PC as well.
Whatever is going on is clearly transient, which makes it very difficult to isolate.
Additionally, the fact that NinjaSupport does not...(well..at least does not from my limited perspective)....seem to be buried in similar problem tickets from other users, implies that the circumstances are quite specific.
My anecdotal data suggests the problem only occurs when loading loading multiple days data on a PC that hasn't been attached to a live data stream for a while, running a workspace with multiple charts all running indicators with secondary dataseries...(perhaps some 'lock' contention on the update threads ... who knows)....which sometimes causes only one...(usually yesterday's)...IsFirstBarofSession boolean to flag 'false' on a bar that is clearly the beginning bar of a new session.
Making this issue even less clear cut, the same chart with the one 'false' boolean, does shows the 'session break' line...(which implies that at NT8 code 'found the session break').... Why that one single IsFirstBarofSession boolean returns false has me flummoxed. Why it seems to correct itself later in the day (sometimes) and corrects on the next new day (mostly) and will always eventually correct if I delete the db cache and reload data, also has me flummoxed.
I imagine this is not quite the answer you hoped for....but so far it is the best I've been able do. Hope springs eternal that something will "click" and I can isolate what is going on....or.....that some one of the other changes/corrections Ninja installs, will also fix this.
Comment