[RADIATOR] Intermittent "User Database Access Error"

[Bjoern A. Zeeb]

On 16 Dec 2018, at 18:10, rohan.henry at cwjamaica.com wrote:

>> On December 16, 2018 at 6:14 AM Christian Kratzer <ck at cksoft.de> 
>> wrote:

>> this could be related to timeouts in openldap connections.
>>
>> Check following parameter for openldap slapd
>>
>>         idletimeout <integer>
>>                Specify the number of seconds to wait before forcibly 
>> closing an
>>                idle client  connection.   A  idletimeout  of  0  
>> disables  this
>>                feature.   The  default  is  0.  You  may  also  want 
>> to set the
>>                writetimeout option.
>>
>> If set openldap will close the socket upon timeout and radiator will 
>> only notice on the next request.
>
> I am using CGP LDAP. Unfortunately, I still haven't located where to 
> adjust the timeouts you specified.
>
> But my AuthBy LDAP2 clause has following.
>
>         Host            ldap1_IP ldap2_IP ldap3_IP
>
>         Version         3
>         HoldServerConnection
>         Timeout         10
>
> I also notice that despite specifying multiple LDAP servers on the 
> Host line Radiator does not switch to the alternate servers when the 
> first becomes unavailable (physically offline). I had to implement 
> multiple AuthBy LDAP2 clause and add the clauses to the Handler as a 
> work-around.


I thought earlier today, what Christian describes I implemented a 
work-around to re-try;  now I wonder, maybe it was for a different bit 
other than LDAP.

Doing the multiple-host should be easy to implement if it’s not there 
yet.  I’ll see if I’ll find a few minutes over the holidays.

/bz