diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 03ee898..cc784fa 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -28,20 +28,16 @@ jobs:
         id: datasets
         working-directory: qadb
         run: |
-          ls -d pass*/* | jq -Rs '{"dataset": split("\n")[:-1]}' > list.json
-          echo "### List of Datasets" >> $GITHUB_STEP_SUMMARY
-          echo '```json' >> $GITHUB_STEP_SUMMARY
-          cat list.json | xargs -0 -I{} echo {} >> $GITHUB_STEP_SUMMARY
-          echo '```' >> $GITHUB_STEP_SUMMARY
-          echo datasets=$(jq -c . list.json) >> $GITHUB_OUTPUT
+          ls -d pass*/* | jq -Rs '{"dataset": split("\n")[:-1]}' | tee datasets.json
+          echo datasets=$(jq -c . datasets.json) >> $GITHUB_OUTPUT
 
   # check consistency between Groovy and C++ APIs
-  groovy_vs_cpp:
+  test_dataset:
     needs:
       - get_datasets
     runs-on: ubuntu-latest
     strategy:
-      fail-fast: true
+      fail-fast: false
       matrix: ${{ fromJson(needs.get_datasets.outputs.datasets) }}
     steps:
       - name: checkout
@@ -57,21 +53,29 @@ jobs:
           source environ.sh
           echo "QADB=${QADB}" >> $GITHUB_ENV
           echo "JYPATH=${JYPATH}" >> $GITHUB_ENV
-      - name: compile_cpp_tests
+      - name: make sure all the table files exist # since pre-commit.ci auto-fix bot will not `git add` *new* files, we fallback to checking their existence here
+        working-directory: qadb/${{ matrix.dataset }}
         run: |
-          cd srcC/tests
-          make
-      - name: test_diff_groovy_cpp
+          for f in qaTree.json.table miscTable.md; do
+            if [ ! -f $f ]; then
+              echo "missing table file '$f'; run pre-commit hook manually and commit new file(s)"
+              exit 1
+            fi
+          done
+      - name: sync check
+        run: stdbuf -i0 -e0 -o0 util/syncCheck.rb ${{ matrix.dataset }}
+      - name: diff the Groovy and C++ QADB dumps
         run: |
-          tests/test_diffGroovyCpp.loop.sh ${{matrix.dataset}}
-      - name: concatenate_artifacts
+          make -C srcC/tests
+          tests/test_diffGroovyCpp.loop.sh ${{ matrix.dataset }}
+      - name: concatenate artifacts
         id: artifacts
         run: |
-          mkdir -p artifacts/${{matrix.dataset}}
+          mkdir -p artifacts/${{ matrix.dataset }}
           for lang in cpp groovy ; do
-            cat tmp/${lang}*.out > artifacts/${{matrix.dataset}}/${lang}.txt ;
+            cat tmp/${lang}*.out > artifacts/${{ matrix.dataset }}/${lang}.txt ;
           done
-          echo "artifact_name=groovy_vs_cpp__$(echo ${{matrix.dataset}} | sed 's;/;_;g')" | tee -a $GITHUB_OUTPUT
+          echo "artifact_name=groovy_vs_cpp__$(echo ${{ matrix.dataset }} | sed 's;/;_;g')" | tee -a $GITHUB_OUTPUT
       - name: upload_artifacts
         uses: actions/upload-artifact@v4
         with:
@@ -79,10 +83,10 @@ jobs:
           retention-days: 3
           path: artifacts/*
 
-  # report status for github status check (successful only if all `groovy_vs_cpp` jobs pass)
+  # report status for github status check (successful only if all `test_dataset` jobs pass)
   report:
     runs-on: ubuntu-latest
     needs:
-      - groovy_vs_cpp
+      - test_dataset
     steps:
       - run: echo success
diff --git a/util/chargeSumRuns.groovy b/util/chargeSumRuns.groovy
index 90c3017..e4c0639 100644
--- a/util/chargeSumRuns.groovy
+++ b/util/chargeSumRuns.groovy
@@ -1,6 +1,6 @@
 // calculate total analyzed charge for a run period,
 // with specified QA cuts enabled
-// note: if syncCheck.groovy errors are present in the run range,
+// note: if syncCheck.rb errors are present in the run range,
 // the final charge value might be a bit wrong...
 
 import org.jlab.io.hipo.HipoDataSource
diff --git a/util/syncCheck.groovy b/util/syncCheck.groovy
deleted file mode 100644
index a396ca6..0000000
--- a/util/syncCheck.groovy
+++ /dev/null
@@ -1,42 +0,0 @@
-// highlights where we have a QADB syncing problem:
-// eventnumMax of DST file N is larger than eventnumMin of DST file N+1
-// (more correctly, of N+5, since these are 5-files)
-
-
-// open QADB
-import clasqa.QADB
-QADB qa = new QADB("latest")
-
-
-def printSep
-
-// loop through qaTree runs (sorted by run number)
-qa.getQaTree().sort{a,b -> 
-  a.key.toInteger() <=> b.key.toInteger()
-}.each{ runnum,runTree ->
-  printSep = false
-
-  // loop through run's files (sorted by file number)
-  runTree.sort{c,d ->
-    c.key.toInteger() <=> d.key.toInteger()
-  }.each { filenum,fileTree ->
-    def evnumMin = fileTree['evnumMin']
-    def evnumMax = fileTree['evnumMax']
-    //println "CHECK: $runnum $filenum $evnumMin $evnumMax"
-    def filenumNxt = filenum.toInteger() + 5
-    if(runTree["$filenumNxt"]!=null) {
-      def fileTreeNxt = runTree["$filenumNxt"]
-      def evnumMinNxt = fileTreeNxt['evnumMin']
-      def evnumMaxNxt = fileTreeNxt['evnumMax']
-      if( evnumMax > evnumMinNxt) {
-        def overlap = evnumMax - evnumMinNxt
-        println "SYNC ERROR: runnum=$runnum"
-        println "  file $filenum\tevnumMin = $evnumMin\tevnumMax = $evnumMax"
-        println "  file $filenumNxt\tevnumMin = $evnumMinNxt\tevnumMax = $evnumMaxNxt"
-        println "  overlap = $overlap"
-        printSep = true
-      }
-    }
-  }
-  if(printSep) println "=================================="
-}
diff --git a/util/syncCheck.rb b/util/syncCheck.rb
new file mode 100755
index 0000000..8049f0e
--- /dev/null
+++ b/util/syncCheck.rb
@@ -0,0 +1,77 @@
+#!/usr/bin/env ruby
+# highlights where we have a QADB syncing problem:
+# eventnumMax of bin N is larger than eventnumMin of bin N+1
+
+require 'json'
+
+if ARGV.empty?
+  puts "USAGE: #{$0} [dataset]"
+  exit(2)
+end
+raise 'source environment variables first' if ENV['QADB'].nil?
+dataset = ARGV.first
+
+qa_tree_file = File.join ENV['QADB'], 'qadb', dataset.split('/'), 'qaTree.json'
+raise "#{qa_tree_file} does not exist" unless File.exist? qa_tree_file
+qa_tree = JSON.load_file qa_tree_file
+
+# true if any issue found
+found_issue_anywhere = false
+
+# loop over runs
+qa_tree.sort{ |a,b| a.first.to_i <=> b.first.to_i }.each do |runnum, run_tree|
+
+  binnum_prev = nil
+  found_issue_in_run = false
+
+  # loop over QA bins
+  run_tree.sort{ |a,b| a.first.to_i <=> b.first.to_i }.each do |binnum, bin_tree|
+
+    # check that min event number < max
+    evnumMin = bin_tree['evnumMin']
+    evnumMax = bin_tree['evnumMax']
+    if evnumMin > evnumMax
+      $stderr.puts "ERROR run=#{runnum} bin=#{binnum}: evnumMin > evnumMax: #{evnumMin} > #{evnumMax}"
+      found_issue_anywhere = true
+      found_issue_in_run = true
+    end
+
+    # check for overlap with previous bin
+    unless binnum_prev.nil?
+      evnumMin_prev = run_tree[binnum_prev]['evnumMin']
+      evnumMax_prev = run_tree[binnum_prev]['evnumMax']
+      if evnumMax_prev > evnumMin
+        $stderr.puts """SYNC ERROR: run=#{runnum} bin=#{binnum}: previous bin's evnumMax > this bin's evnumMin:
+  prev bin:  binnum=#{binnum_prev}\tevnumMin=#{evnumMin_prev}\tevnumMax=#{evnumMax_prev}
+  this bin:  binnum=#{binnum}\tevnumMin=#{evnumMin}\tevnumMax=#{evnumMax}"""
+        found_issue_anywhere = true
+        found_issue_in_run = true
+      end
+    end
+
+    binnum_prev = binnum
+
+  end
+
+  puts "RUN #{runnum} #{found_issue_in_run ? "has issues (see stderr)" : "is okay"}"
+end
+
+if found_issue_anywhere
+  if [ # datasets which used DST 5-files as QA bins
+      "pass1/rga_fa18_inbending",
+      "pass1/rga_fa18_outbending",
+      "pass1/rga_sp19",
+      "pass1/rgb_fa19",
+      "pass1/rgb_sp19",
+      "pass1/rgb_wi20",
+      "pass1/rgk_fa18_6.5GeV",
+      "pass1/rgk_fa18_7.5GeV",
+      "pass1/rgm_fa21",
+  ].include? dataset
+    $stderr.puts "WARNING: this dataset was done using DST 5-files as QA bins, which inherently causes SYNC ERRORS; now exitting with 0"
+    exit 0
+  else
+    $stderr.puts "ERROR: this dataset should not have any errors"
+    exit 1
+  end
+end