cnescatlab/i-CodeCNES

Too many open files

Closed this issue · 0 comments

Describe the bug
When i-code analyses projects with a very large number of files (+2000), a memory overflow occurs when the files are opened by the I-code lexers. This implies that some files that the user wants to analyze will not be processed. The files which undergo the bug are random with each new analysis. This implies unpredictable and random results. Thus, a whole project cannot be analysed.

To reproduce

  • Run an i-code analysis on a big project with more than 2000 files in it.
  • Environment needed: can't define. This bug occurs in a SonarQube server at EUMETSAT. You can't reproduce the bug easily, you need to find a "weak" environment that can't handle all this amount of files.

Expected behavior
The scan should process all files and be constant.

Screenshots & log
image

User environment (please complete the following information):

  • OS: CentOS Linux 7
  • Java version: 11.0.15
  • i-Code version: 4.1.0