diff --git a/synda/sdt/sdnexturl.py b/synda/sdt/sdnexturl.py
index f0233640..3d91f4d3 100755
--- a/synda/sdt/sdnexturl.py
+++ b/synda/sdt/sdnexturl.py
@@ -14,9 +14,9 @@
 # multiple changes by JfP to make fallback more flexible.
 
 import argparse
-import sdlog
-import sdquicksearch
-import sdexception
+from synda.sdt import sdlog
+from synda.sdt import sdquicksearch
+from synda.sdt import sdexception
 import sqlite3
 
 from synda.source.config.file.user.preferences.models import Config as Preferences
@@ -43,8 +43,10 @@ def run(tr):
                   (tr.url, tr.file_functional_id) )
         conn.commit()
     except sqlite3.IntegrityError as e:
-        # url is already in the failed_url table
-        sdlog.info("SDNEXTUR-001","During database operations, IntegrityError %s"%(e,))
+        # url,file_id is already in the failed_url table
+        sdlog.info("SDNEXTUR-001","During database operations, IntegrityError %s on %s with new file_id=%s"
+                   %(e,url,tr.file_id))
+        conn.commit()
     except Exception as e:
         sdlog.info("SDNEXTUR-002","During database operations, unknown exception %s"%(e,))
         return False
@@ -79,14 +81,11 @@ def next_url(tr,conn):
     sdlog.info("SDNEXTUR-007","failed_urls= %s"%(failed_urls,))
     urlps = [urlp for urlp in all_urlps if urlp[0] not in failed_urls]
     # ... Note that list comprehensions preserve order.
-
-    # At this point urls is just a list of urls.  We no longer have to keep track of the
-    # protocol because only http and gsiftp are possible (OpenDAP is a different protocol
-    # but it also uses a http url).
+    urls = [urlp[0] for urlp in urlps]
     
-    if len(urlps)>0:
+    if len(urls)>0:
         old_url=tr.url
-        new_url=urlps[0]
+        new_url=urls[0]
         tr.url=new_url
         sdlog.info("SDNEXTUR-008","Url successfully switched (file_functional_id=%s,old_url=%s,new_url=%s)"%(tr.file_functional_id,old_url,new_url))
     else:
@@ -103,11 +102,11 @@ def get_urls(file_functional_id):
                        file_functional_id],
             post_pipeline_mode=None )
     except Exception as e:
-        sdlog.debug("SDNEXTUR-015", "exception %s.  instance_id=%s"%(e,file_functional_id))
+        sdlog.debug("SDNEXTUR-010", "exception %s.  instance_id=%s"%(e,file_functional_id))
         raise e
 
     li=result.get_files()
-    sdlog.info("SDNEXTUR-016","sdquicksearch returned %s sets of file urls: %s"%(len(li),li))
+    sdlog.info("SDNEXTUR-011","sdquicksearch returned %s sets of file urls: %s"%(len(li),li))
     if li==[]:
         # No urls found. Try again, but wildcard the file id. (That leads to a string search on all
         # fields for the wildcarded file id, rather than a match of the instance_id field only.)
@@ -116,7 +115,7 @@ def get_urls(file_functional_id):
                        file_functional_id+'*'],
             post_pipeline_mode=None )
         li=result.get_files()
-        sdlog.info("SDNEXTUR-017","sdquicksearch 2nd call %s sets of file urls: %s"%(len(li),li))
+        sdlog.info("SDNEXTUR-012","sdquicksearch 2nd call %s sets of file urls: %s"%(len(li),li))
     # result looks like
     # [ {protocol11:url11, protocol12:url12, attached_parameters:dict, score:number, type:'File',
     #    size:number} }, {[another dict of the same format}, {another dict},... ]
@@ -160,6 +159,8 @@ def priurl(url):
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
 
-    file_functional_id='cmip5.output1.LASG-CESS.FGOALS-g2.decadal1985.day.atmos.day.r1i1p1.v1.va_day_FGOALS-g2_decadal1985_r1i1p1_19880101-19881231.nc'
+    #file_functional_id='cmip5.output1.LASG-CESS.FGOALS-g2.decadal1985.day.atmos.day.r1i1p1.v1.va_day_FGOALS-g2_decadal1985_r1i1p1_19880101-19881231.nc'
+    #file_functional_id='cmip5.output1.CMCC.CMCC-CMS.piControl.mon.atmos.Amon.r1i1p1.v20120717.pr_Amon_CMCC-CMS_piControl_r1i1p1_416401-417312.nc'
+    file_functional_id='cmip5.output1.NOAA-GFDL.GFDL-CM2p1.historical.mon.atmos.Amon.r3i1p1.v20110601.cl_Amon_GFDL-CM2p1_historical_r3i1p1_190601-191012.nc'
     from pprint import pprint
-    pprint( get_urls(file_functional_id) )
+    pprint( get_urls(file_functional_id,'esgf-node.llnl.gov') )