Tune Tesseract & Sidekiq to (hopefully) prevent locks

tesseract-ocr/tesseract#898
Minitex · Dec 17, 2021 · 96df0f9 · 96df0f9
1 parent 87c11b2
commit 96df0f9
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 2 deletions.
diff --git a/config/sidekiq.yml b/config/sidekiq.yml
@@ -1,6 +1,6 @@
 ---
 :verbose: false
-:concurrency: 5 # Sidekiq default
+:concurrency: 4
 :timeout: 25
 
 # Sidekiq will run this file through ERB when reading it so you can

diff --git a/lib/mdl/process_document_for_search.rb b/lib/mdl/process_document_for_search.rb
@@ -196,8 +196,9 @@ def manifest_uri
 
     def run_ocr_on(candidate)
       log("Running OCR for canvas #{candidate.canvas_id}")
+      env = { 'OMP_THREAD_LIMIT' => '1' }
       command = "tesseract #{candidate.image_file.path} #{candidate.ocr_temp_file_path} -l eng hocr"
-      _output, error, status = Open3.capture3(command)
+      _output, error, status = Open3.capture3(env, command)
       if status.success?
         log("OCR successful for image on canvas #{candidate.canvas_id}")
       else