diff --git a/.pylintrc b/.pylintrc
index 7b9da25..2932039 100644
--- a/.pylintrc
+++ b/.pylintrc
@@ -145,7 +145,10 @@ disable=print-statement,
         global-statement,
         bad-continuation,
         too-few-public-methods,
-        no-else-return
+        no-else-return,
+        consider-using-f-string,
+        use-implicit-booleaness-not-comparison,
+        duplicate-code
 
 # Enable the message, report, category or checker with the given id(s). You can
 # either give multiple identifier separated by comma (,) or put this option
diff --git a/Dockerfile b/Dockerfile
index cc22d8b..4b9ffb8 100644
--- a/Dockerfile
+++ b/Dockerfile
@@ -1,7 +1,7 @@
 # We do a multi-stage build (requires Docker >= 17.05) to install everything, then copy it all
 # to the final target image.
 
-FROM centos:8 AS base
+FROM rockylinux:8 AS base
 
 # update packages and install Python
 RUN dnf -y upgrade-minimal && \
diff --git a/asl_articles/__init__.py b/asl_articles/__init__.py
index fd1ba88..e938e9f 100644
--- a/asl_articles/__init__.py
+++ b/asl_articles/__init__.py
@@ -75,7 +75,7 @@ _load_config( _cfg, _fname, "Debug" )
 # initialize logging
 _fname = os.path.join( config_dir, "logging.yaml" )
 if os.path.isfile( _fname ):
-    with open( _fname, "r" ) as fp:
+    with open( _fname, "r", encoding="utf-8" ) as fp:
         logging.config.dictConfig( yaml.safe_load( fp ) )
 else:
     # stop Flask from logging every request :-/
diff --git a/asl_articles/db_report.py b/asl_articles/db_report.py
index 498fdab..901ff2e 100644
--- a/asl_articles/db_report.py
+++ b/asl_articles/db_report.py
@@ -65,16 +65,15 @@ def check_db_link():
     """Check if a link appears to be working."""
     url = request.args.get( "url" )
     try:
-        resp = urllib.request.urlopen(
-            urllib.request.Request( url, method="HEAD" )
-        )
+        req = urllib.request.Request( url, method="HEAD" )
+        with urllib.request.urlopen( req ) as resp:
+            resp_code = resp.code
     except urllib.error.URLError as ex:
-        code = getattr( ex, "code", None )
-        if code:
-            abort( code )
-        abort( 400 )
-    if resp.code != 200:
-        abort( resp.code )
+        resp_code = getattr( ex, "code", None )
+        if not resp_code:
+            resp_code = 400
+    if resp_code != 200:
+        abort( resp_code )
     return "ok"
 
 # ---------------------------------------------------------------------
diff --git a/asl_articles/images.py b/asl_articles/images.py
index 7d9765e..42a8e58 100644
--- a/asl_articles/images.py
+++ b/asl_articles/images.py
@@ -21,5 +21,5 @@ def get_image( image_type, image_id ):
         abort( 404 )
     return send_file(
         io.BytesIO( img.image_data ),
-        attachment_filename = img.image_filename # nb: so that Flask can set the MIME type
+        download_name = img.image_filename # nb: so that Flask can set the MIME type
     )
diff --git a/asl_articles/main.py b/asl_articles/main.py
index 459c2b1..34f83b9 100644
--- a/asl_articles/main.py
+++ b/asl_articles/main.py
@@ -1,7 +1,5 @@
 """ Main handlers. """
 
-from flask import request
-
 from asl_articles import app
 
 # ---------------------------------------------------------------------
@@ -10,11 +8,3 @@ from asl_articles import app
 def ping():
     """Let the caller know we're alive (for testing porpoises)."""
     return "pong"
-
-# ---------------------------------------------------------------------
-
-@app.route( "/shutdown" )
-def shutdown():
-    """Shutdown the server (for testing porpoises)."""
-    request.environ.get( "werkzeug.server.shutdown" )()
-    return ""
diff --git a/asl_articles/tests/test_articles.py b/asl_articles/tests/test_articles.py
index 0957980..00af88f 100644
--- a/asl_articles/tests/test_articles.py
+++ b/asl_articles/tests/test_articles.py
@@ -278,8 +278,9 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             btn = find_child( ".row.image .remove-image", dlg )
             assert btn.is_displayed()
             # make sure the article's image is correct
-            resp = urllib.request.urlopen( image_url ).read()
-            assert resp == open( expected, "rb" ).read()
+            with urllib.request.urlopen( image_url ) as resp:
+                with open( expected, "rb" ) as fp:
+                    assert resp.read() == fp.read()
         else:
             # make sure there is no image
             img = find_child( ".row.image img.image", dlg )
@@ -290,7 +291,8 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             # make sure the article's image is not available
             url = flask_app.url_for( "get_image", image_type="article", image_id=article_id )
             try:
-                resp = urllib.request.urlopen( url )
+                with urllib.request.urlopen( url ):
+                    pass
                 assert False, "Should never get here!"
             except urllib.error.HTTPError as ex:
                 assert ex.code == 404
@@ -350,7 +352,8 @@ def test_parent_publisher( webdriver, flask_app, dbconn ):
         # check that the parent publication was updated in the database
         article_id = sr.get_attribute( "testing--article_id" )
         url = flask_app.url_for( "get_article", article_id=article_id )
-        article = json.load( urllib.request.urlopen( url ) )
+        with urllib.request.urlopen( url ) as resp:
+            article = json.load( resp )
         if expected_parent:
             if article["pub_id"] != expected_parent[0]:
                 return None
diff --git a/asl_articles/tests/test_authors.py b/asl_articles/tests/test_authors.py
index 5bc67c2..f1b3ce2 100644
--- a/asl_articles/tests/test_authors.py
+++ b/asl_articles/tests/test_authors.py
@@ -92,5 +92,6 @@ def _check_authors( flask_app, all_authors, expected ):
 
     # check the authors in the database
     url = flask_app.url_for( "get_authors" )
-    authors = json.load( urllib.request.urlopen( url ) )
+    with urllib.request.urlopen( url ) as resp:
+        authors = json.load( resp )
     assert set( a["author_name"] for a in authors.values() ) == all_authors
diff --git a/asl_articles/tests/test_import_roar_scenarios.py b/asl_articles/tests/test_import_roar_scenarios.py
index 6c8b1bb..c0c20a4 100644
--- a/asl_articles/tests/test_import_roar_scenarios.py
+++ b/asl_articles/tests/test_import_roar_scenarios.py
@@ -18,7 +18,8 @@ def test_import_roar_scenarios( dbconn ):
     # initialize
     session = init_tests( None, None, dbconn )
     roar_fname = os.path.join( os.path.split(__file__)[0], "fixtures/roar-scenarios.json" )
-    roar_data = json.load( open( roar_fname, "r" ) )
+    with open( roar_fname, "r", encoding="utf-8" ) as fp:
+        roar_data = json.load( fp  )
 
     # do the first import
     _do_import( dbconn, session, roar_fname,
diff --git a/asl_articles/tests/test_publications.py b/asl_articles/tests/test_publications.py
index db2be13..18bdc3e 100644
--- a/asl_articles/tests/test_publications.py
+++ b/asl_articles/tests/test_publications.py
@@ -246,8 +246,9 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             btn = find_child( ".row.image .remove-image", dlg )
             assert btn.is_displayed()
             # make sure the publication's image is correct
-            resp = urllib.request.urlopen( image_url ).read()
-            assert resp == open( expected, "rb" ).read()
+            with urllib.request.urlopen( image_url ) as resp:
+                with open( expected, "rb" ) as fp:
+                    assert resp.read() == fp.read()
         else:
             # make sure there is no image
             img = find_child( ".row.image img.image", dlg )
@@ -258,7 +259,8 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             # make sure the publication's image is not available
             url = flask_app.url_for( "get_image", image_type="publication", image_id=pub_id )
             try:
-                resp = urllib.request.urlopen( url )
+                with urllib.request.urlopen( url ):
+                    pass
                 assert False, "Should never get here!"
             except urllib.error.HTTPError as ex:
                 assert ex.code == 404
@@ -318,7 +320,8 @@ def test_parent_publisher( webdriver, flask_app, dbconn ):
         # check that the parent publisher was updated in the database
         pub_id = sr.get_attribute( "testing--pub_id" )
         url = flask_app.url_for( "get_publication", pub_id=pub_id )
-        pub = json.load( urllib.request.urlopen( url ) )
+        with urllib.request.urlopen( url ) as resp:
+            pub = json.load( resp )
         if expected_parent:
             if pub["publ_id"] != expected_parent[0]:
                 return None
@@ -672,8 +675,11 @@ def test_default_image( webdriver, flask_app, dbconn ):
         f: os.path.join( os.path.split(__file__)[0], "fixtures/images/"+f )
         for f in images
     }
+    def read_image_data( fname ):
+        with open( fname, "rb" ) as fp:
+            return fp.read()
     image_data = {
-        f: open( image_fnames[f], "rb" ).read()
+        f: read_image_data( image_fnames[f] )
         for f in images
     }
 
@@ -690,8 +696,8 @@ def test_default_image( webdriver, flask_app, dbconn ):
         if img:
             assert expected
             image_url = img.get_attribute( "src" )
-            resp = urllib.request.urlopen( image_url ).read()
-            assert resp == image_data[ expected ]
+            with urllib.request.urlopen( image_url ) as resp:
+                assert resp.read() == image_data[ expected ]
         else:
             assert not expected
 
diff --git a/asl_articles/tests/test_publishers.py b/asl_articles/tests/test_publishers.py
index fdb4f6b..c3fd963 100644
--- a/asl_articles/tests/test_publishers.py
+++ b/asl_articles/tests/test_publishers.py
@@ -176,8 +176,9 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             btn = find_child( ".row.image .remove-image", dlg )
             assert btn.is_displayed()
             # make sure the publisher's image is correct
-            resp = urllib.request.urlopen( image_url ).read()
-            assert resp == open(expected,"rb").read()
+            with urllib.request.urlopen( image_url ) as resp:
+                with open( expected, "rb" ) as fp:
+                    assert resp.read() == fp.read()
         else:
             # make sure there is no image
             img = find_child( ".row.image img.image", dlg )
@@ -188,7 +189,8 @@ def test_images( webdriver, flask_app, dbconn ): #pylint: disable=too-many-state
             # make sure the publisher's image is not available
             url = flask_app.url_for( "get_image", image_type="publisher", image_id=publ_id )
             try:
-                resp = urllib.request.urlopen( url )
+                with urllib.request.urlopen( url ):
+                    pass
                 assert False, "Should never get here!"
             except urllib.error.HTTPError as ex:
                 assert ex.code == 404
diff --git a/asl_articles/tests/test_scenarios.py b/asl_articles/tests/test_scenarios.py
index 25b32ba..9bcd315 100644
--- a/asl_articles/tests/test_scenarios.py
+++ b/asl_articles/tests/test_scenarios.py
@@ -104,7 +104,8 @@ def _check_scenarios( flask_app, all_scenarios, expected ):
 
     # check the scenarios in the database
     url = flask_app.url_for( "get_scenarios" )
-    scenarios = json.load( urllib.request.urlopen( url ) )
+    with urllib.request.urlopen( url ) as resp:
+        scenarios = json.load( resp )
     assert set( _make_scenario_display_name(a) for a in scenarios.values() ) == all_scenarios
 
 def _make_scenario_display_name( scenario ):
diff --git a/asl_articles/tests/test_tags.py b/asl_articles/tests/test_tags.py
index 5091503..442f7f8 100644
--- a/asl_articles/tests/test_tags.py
+++ b/asl_articles/tests/test_tags.py
@@ -145,10 +145,12 @@ def _check_tags( flask_app, expected ): #pylint: disable=too-many-locals
         if sr.text.startswith( "publication" ):
             pub_id = sr.get_attribute( "testing--pub_id" )
             url = flask_app.url_for( "get_publication", pub_id=pub_id )
-            pub = json.load( urllib.request.urlopen( url ) )
+            with urllib.request.urlopen( url ) as resp:
+                pub = json.load( resp )
             assert expected[ pub["pub_name"] ] == fixup_tags( pub["pub_tags"] )
         elif sr.text.startswith( "article" ):
             article_id = sr.get_attribute( "testing--article_id" )
             url = flask_app.url_for( "get_article", article_id=article_id )
-            article = json.load( urllib.request.urlopen( url ) )
+            with urllib.request.urlopen( url ) as resp:
+                article = json.load( resp )
             assert expected[ article["article_title"] ] == fixup_tags( article["article_tags"] )
diff --git a/asl_articles/tests/utils.py b/asl_articles/tests/utils.py
index e002bce..76800ff 100644
--- a/asl_articles/tests/utils.py
+++ b/asl_articles/tests/utils.py
@@ -49,7 +49,8 @@ def init_tests( webdriver, flask_app, dbconn, **kwargs ):
     # re-initialize the search engine
     if flask_app:
         url = flask_app.url_for( "init_search_for_test" )
-        _ = urllib.request.urlopen( url ).read()
+        with urllib.request.urlopen( url ) as resp:
+            _ = resp.read()
 
     # initialize the documents directory
     dname = kwargs.pop( "docs", None )
@@ -70,7 +71,10 @@ def init_tests( webdriver, flask_app, dbconn, **kwargs ):
         if to_bool( kwargs.pop( "disable_confirm_discard_changes", True ) ):
             kwargs[ "disable_confirm_discard_changes" ] = 1
         webdriver.get( webdriver.make_url( "", **kwargs ) )
-        wait_for_elem( 2, "#search-form" )
+        # FUDGE! Since we switched from running the test Flask server with app.run() to make_server().serve_forever(),
+        # stopping and starting the server seems to be much quicker, but refreshing the page can be slower when
+        # running multiple tests :shrug:
+        wait_for_elem( 10, "#search-form" )
 
     return session
 
@@ -83,7 +87,8 @@ def load_fixtures( session, fname ):
     if fname:
         dname = os.path.join( os.path.split(__file__)[0], "fixtures/" )
         fname = os.path.join( dname, fname )
-        data = json.load( open( fname, "r" ) )
+        with open( fname, "r", encoding="utf-8" ) as fp:
+            data = json.load( fp )
     else:
         data = {}
 
@@ -318,21 +323,21 @@ def wait_for_not_elem( timeout, sel ):
 def find_child( sel, parent=None ):
     """Find a child element."""
     try:
-        return (parent if parent else _webdriver).find_element_by_css_selector( sel )
+        return (parent if parent else _webdriver).find_element( By.CSS_SELECTOR, sel )
     except NoSuchElementException:
         return None
 
 def find_children( sel, parent=None ):
     """Find child elements."""
     try:
-        return (parent if parent else _webdriver).find_elements_by_css_selector( sel )
+        return (parent if parent else _webdriver).find_elements( By.CSS_SELECTOR, sel )
     except NoSuchElementException:
         return None
 
 def find_parent_by_class( elem, class_name ):
     """Find a parent element with the specified class."""
     while True:
-        elem = elem.find_element_by_xpath( ".." )
+        elem = elem.find_element( By.XPATH, ".." )
         if not elem:
             return None
         classes = set( elem.get_attribute( "class" ).split() )
@@ -497,7 +502,8 @@ def call_with_retry( func, expected_exceptions, max_retries=10, delay=0.1 ):
 def change_image( dlg, fname ):
     """Click on an image to change it."""
     # NOTE: This is a bit tricky since we started overlaying the image with the "remove image" icon :-/
-    data = base64.b64encode( open( fname, "rb" ).read() )
+    with open( fname, "rb" ) as fp:
+        data = base64.b64encode( fp.read() )
     data = "{}|{}".format( os.path.split(fname)[1], data.decode("ascii") )
     elem = find_child( ".row.image img.image", dlg )
     _webdriver.execute_script( "arguments[0].scrollTo( 0, 0 )", find_child( ".MuiDialogContent-root", dlg ) )
diff --git a/conftest.py b/conftest.py
index dbcc54c..8cdc35a 100644
--- a/conftest.py
+++ b/conftest.py
@@ -9,6 +9,7 @@ from urllib.error import URLError
 
 import pytest
 import flask
+import werkzeug
 import sqlalchemy
 from flask_sqlalchemy import SQLAlchemy
 import alembic
@@ -99,19 +100,22 @@ def flask_app( request ):
         # the *configured* database connection string (since it will fail to start if there's a problem).
         asl_articles._disable_db_startup = True #pylint: disable=protected-access
         # yup - make it so
+        server = werkzeug.serving.make_server(
+            _FLASK_SERVER_URL[0], _FLASK_SERVER_URL[1],
+            app, threaded=True
+        )
         thread = threading.Thread(
-            target = lambda: app.run(
-                host=_FLASK_SERVER_URL[0], port=_FLASK_SERVER_URL[1],
-                use_reloader=False
-            )
+            target = server.serve_forever,
+            daemon=True
         )
         thread.start()
         # wait for the server to start up
         def is_ready():
             """Try to connect to the Flask server."""
             try:
-                resp = urllib.request.urlopen( app.url_for( "ping" ) ).read()
-                assert resp == b"pong"
+                url = app.url_for( "ping" )
+                with urllib.request.urlopen( url ) as resp:
+                    assert resp.read() == b"pong"
                 return True
             except URLError:
                 return False
@@ -125,7 +129,7 @@ def flask_app( request ):
     finally:
         # shutdown the local Flask server
         if not flask_url:
-            urllib.request.urlopen( app.url_for("shutdown") ).read()
+            server.shutdown()
             thread.join()
 
 # ---------------------------------------------------------------------
@@ -142,10 +146,7 @@ def webdriver( request ):
         options = wb.FirefoxOptions()
         if headless:
             options.add_argument( "--headless" ) #pylint: disable=no-member
-        driver = wb.Firefox(
-            options = options,
-            service_log_path = os.path.join( tempfile.gettempdir(), "geckodriver.log" )
-        )
+        driver = wb.Firefox( options=options )
     elif driver == "chrome":
         options = wb.ChromeOptions()
         if headless:
diff --git a/requirements-dev.txt b/requirements-dev.txt
index 7d96d7f..d6870ca 100644
--- a/requirements-dev.txt
+++ b/requirements-dev.txt
@@ -1,5 +1,5 @@
-pytest==6.2.1
-selenium==3.141.0
-pylint==2.6.0
+pytest==6.2.5
+selenium==4.1.0
+pylint==2.12.2
 pylint-flask-sqlalchemy==0.2.0
 pytest-pylint==0.18.0
diff --git a/requirements.txt b/requirements.txt
index cc6bb4e..b429219 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,13 +1,13 @@
 # python 3.8.7
 
-flask==1.1.2
+flask==2.0.3
 # NOTE: Newer versions of SQLAlchemy contain a change that breaks Flask-SQLALchemy :-/
 #   https://stackoverflow.com/a/66652728
 # This wasn't a problem on vm-linux-dev, but manifested itself on the rPi4 (probably because
 # the virtualenv on vm-linux-dev was built before this became a problem).
 flask-sqlalchemy==2.5.1
-psycopg2-binary==2.8.6
-alembic==1.4.3
-pyyaml==5.3.1
-lxml==4.6.2
+psycopg2-binary==2.9.3
+alembic==1.7.6
+pyyaml==6.0
+lxml==4.8.0
 waitress==2.0.0
diff --git a/run_server.py b/run_server.py
index 62e78fd..a14b8c5 100755
--- a/run_server.py
+++ b/run_server.py
@@ -48,7 +48,8 @@ def _force_init():
             host = host[:-1]
         url = "{}:{}{}".format( host, flask_port, url )
         # make the request
-        _ = urllib.request.urlopen( url ).read()
+        with urllib.request.urlopen( url ) as resp:
+            _ = resp.read()
     except Exception as ex: #pylint: disable=broad-except
         print( "WARNING: Startup ping failed: {}".format( ex ) )
 threading.Thread( target=_force_init ).start()
diff --git a/setup.py b/setup.py
index 6ff1be9..fadd3aa 100644
--- a/setup.py
+++ b/setup.py
@@ -16,11 +16,12 @@ def parse_requirements( fname ):
     """Parse a requirements file."""
     lines = []
     fname = os.path.join( os.path.split(__file__)[0], fname )
-    for line in open(fname,"r"):
-        line = line.strip()
-        if line == "" or line.startswith("#"):
-            continue
-        lines.append( line )
+    with open( fname, "r", encoding="utf-8" ) as fp:
+        for line in fp:
+            line = line.strip()
+            if line == "" or line.startswith("#"):
+                continue
+            lines.append( line )
     return lines
 
 # ---------------------------------------------------------------------
diff --git a/tools/import_roar_scenarios.py b/tools/import_roar_scenarios.py
index e1c166a..00d142c 100755
--- a/tools/import_roar_scenarios.py
+++ b/tools/import_roar_scenarios.py
@@ -62,7 +62,8 @@ def import_roar_scenarios( dbconn, roar_data, progress=None ):
     # load the ROAR scenarios
     if isinstance( roar_data, str ):
         log_progress( "Loading scenarios: {}", roar_data )
-        roar_data = json.load( open( roar_data, "r" ) )
+        with open( roar_data, "r", encoding="utf-8" ) as fp:
+            roar_data = json.load( fp )
     else:
         assert isinstance( roar_data, dict )
     log_progress( "- Last updated: {}".format( roar_data.get("_lastUpdated_","(unknown)") ) )