We are running FreeRADIUS v. 1.0.2
2x load-balanced LDAP servers - Sun ONE DS 5.2 on W2k3 Ent. with network load balancing.  One of the LDAPs is the primary and is handling the auth traffic.

Here is the issue we're seeing:

Approximately 10-20 times per day users are unable to authenticate - despite using correct credentials.  The radius server reports bind failed because it "Can't contact LDAP server"  The LDAP logs show the bind, search, and reply for the "does this user exist" request.  Sometimes this search is repeated a couple of times.  However, there is no follow-up bind as this user for checking the creds.  If the user tries again in 30secs or more, they succeed - with the same creds as before.  

Any ideas?  Thanks for any help!

Below are excerpts from the logs:

------------------------   Radius log entry ------------------------

rlm_ldap: - authorize
rlm_ldap: performing user authorization for someuser
radius_xlat:  '(uid=someuser)'
radius_xlat:  'ou=people,dc=uttyler,dc=edu'
rlm_ldap: ldap_get_conn: Checking Id: 0
rlm_ldap: ldap_get_conn: Got Id: 0
rlm_ldap: performing search in ou=people,dc=uttyler,dc=edu, with filter (uid=someuser)
rlm_ldap: ldap_search() failed: LDAP connection lost.
rlm_ldap: Attempting reconnect
rlm_ldap: attempting LDAP reconnection
rlm_ldap: closing existing LDAP connection
rlm_ldap: (re)connect to ldap.uttyler.edu:389, authentication 0
rlm_ldap: bind as uid=radiususer,ou=special users,dc=uttyler,dc=edu/radius_password to ldap.uttyler.edu:389
rlm_ldap: waiting for bind result ...
rlm_ldap: Bind was successful
rlm_ldap: performing search in ou=people,dc=uttyler,dc=edu, with filter (uid=someuser)
rlm_ldap: looking for check items in directory...
rlm_ldap: looking for reply items in directory...
rlm_ldap: user someuser authorized to use remote access
rlm_ldap: ldap_release_conn: Release Id: 0
  modcall[authorize]: module "ldap" returns ok for request 987
modcall: group authorize returns ok for request 987
  rad_check_password:  Found Auth-Type LDAP
auth: type "LDAP"
  Processing the authenticate section of radiusd.conf
modcall: entering group Auth-Type for request 987
rlm_ldap: - authenticate
rlm_ldap: login attempt by "someuser" with password "04191987"
rlm_ldap: user DN: uid=someuser,ou=People,dc=uttyler,dc=edu
rlm_ldap: (re)connect to ldap.uttyler.edu:389, authentication 1
rlm_ldap: bind as uid=someuser,ou=People,dc=uttyler,dc=edu/04191987 to ldap.uttyler.edu:389
rlm_ldap: uid=someuser,ou=People,dc=uttyler,dc=edu bind to ldap.uttyler.edu:389 failed: Can't contact LDAP server
rlm_ldap: ldap_connect() failed
  modcall[authenticate]: module "ldap" returns fail for request 987
modcall: group Auth-Type returns fail for request 987
auth: Failed to validate the user.
Login incorrect: [someuser/04191987] (from client AireSpace port 0 cli 10.3.1.72)
Delaying request 987 for 1 seconds
Finished request 987
Going to the next request


--------------------  LDAP Log ------------------------------

[23/Jan/2006:07:47:13 -0600] conn=886 op=1 msgId=2 - SRCH base="ou=people,dc=uttyler,dc=edu" scope=2 filter="(uid=someuser)" attrs="radiusexpiration acctflags ntpassword lmpassword radiuscallingstationid radiuscalledstationid radiussimultaneoususe radiusauthtype radiuscheckitem radiusloginlatport radiusportlimit radiusframedappletalkzone radiusframedappletalknetwork radiusframedappletalklink radiusloginlatgroup radiusloginlatnode radiusloginlatservice radiusterminationaction radiusidletimeout radiussessiontimeout radiusclass radiusframedipxnetwork radiuscallbackid radiuscallbacknumber radiuslogintcpport radiusloginservice radiusloginiphost radiusframedcompression radiusframedmtu radiusfilterid radiusframedrouting radiusframedroute radiusframedipnetmask radiusframedipaddress radiusframedprotocol radiusservicetype radiusreplyitem"
[23/Jan/2006:07:47:13 -0600] conn=886 op=1 msgId=2 - RESULT err=0 tag=101 nentries=1 etime=0
[23/Jan/2006:07:47:24 -0600] conn=886 op=2 msgId=3 - SRCH base="ou=people,dc=uttyler,dc=edu" scope=2 filter="(uid=someuser)" attrs="radiusexpiration acctflags ntpassword lmpassword radiuscallingstationid radiuscalledstationid radiussimultaneoususe radiusauthtype radiuscheckitem radiusloginlatport radiusportlimit radiusframedappletalkzone radiusframedappletalknetwork radiusframedappletalklink radiusloginlatgroup radiusloginlatnode radiusloginlatservice radiusterminationaction radiusidletimeout radiussessiontimeout radiusclass radiusframedipxnetwork radiuscallbackid radiuscallbacknumber radiuslogintcpport radiusloginservice radiusloginiphost radiusframedcompression radiusframedmtu radiusfilterid radiusframedrouting radiusframedroute radiusframedipnetmask radiusframedipaddress radiusframedprotocol radiusservicetype radiusreplyitem"
[23/Jan/2006:07:47:24 -0600] conn=886 op=2 msgId=3 - RESULT err=0 tag=101 nentries=1 etime=0
[23/Jan/2006:07:47:25 -0600] conn=887 op=-1 msgId=-1 - fd=1132 slot=1132 LDAP connection from 198.213.57.20 to 198.213.56.5
[23/Jan/2006:07:47:25 -0600] conn=887 op=0 msgId=1 - BIND dn="uid=someuser,ou=People,dc=uttyler,dc=edu" method=128 version=3
[23/Jan/2006:07:47:25 -0600] conn=887 op=0 msgId=1 - RESULT err=0 tag=97 nentries=0 etime=0 dn="uid=someuser,ou=people,dc=uttyler,dc=edu"
[23/Jan/2006:07:47:25 -0600] conn=887 op=1 msgId=2 - UNBIND


Tim Crouch
Systems Administrator
Campus Computing Services
903-566-7476