I think that we already spoke about that. It would be great if you could filter out all snapshots which were unsuccessfully uploaded due to network troubles or cancelled action by user.
You can take a look i.e. onto issue #195278. There was almost 50% of all investigated snapshots broken.
Thanks a lot...
Sorry Tomas, I have to increase priority. I evaluated next about 40 npss and more than 50% of them was broken again. Anything what could be done in that?
Thanks a lot...
Can you specify what do you mean by 'broken snapshot'?
I am sorry, but I do not see any broken snapshot there.
If you see this in the exception report
Profiler Statistics : Samples count: 305
Std. deviation: 6.0
it means that the snapshot was successfully read.
Jardo, it seems like there's something wrong on mechanism which generates snapshot that is later uploaded. e.g. http://statistics.netbeans.org/exceptions/exception.do?id=479982 stand for ~4s slowness, but npss snapshot collected only ~60ms of data. Could you please take a look?
Not a part of the IDE codebase and IDE users are not affected -> NO701 keyword.
See also issue #199774.
Sorry Tomas, for my bad explaining of the problem. I'm worried about the same problem mentioned by Tomas Musil in comment #3. There are collected shorted data than should be collected (often lower than ~100ms).
Thanks for the issue #199774, this will help at least in half of cases.
Any plans to address this for 7.3 (or in the near future)?
*** Bug 179182 has been marked as a duplicate of this bug. ***