Segmentation Fault

[Paul Harvey <pharvey@cern.ch>]

Hello all!


I am currently trying to use syncrepl to maintain a set of databases 
around the world.

I am currently doing some testing to see how things would cope over 
large distances, between Europe and Australia. The current set up is 
that there is a master provider in Europe which is being periodically 
modified, and two consumers: one also in Europe and one in Australia.

Things were going fine - databases were in sync on both consumers - 
until both consumers are now segfaulting.

The output even with full debugging gives no clue as to an error, simply 
segfault is popping up. Short of getting the c source and opening up gdb 
i am at a bit of a loss as to why.

=> access_allowed: search access to "o=grid" "entry" requested
<= root access granted
=> access_allowed: search access granted by manage(=mwrscxd)
=> access_allowed: search access to 
"Mds-Vo-name=ru-Moscow-GCRAS-LCG2,Mds-Vo-name=local,o=grid" "entry" 
requested
<= root access granted
=> access_allowed: search access granted by manage(=mwrscxd)
=> access_allowed: search access to 
"Mds-Vo-name=ru-Moscow-GCRAS-LCG2,Mds-Vo-name=local,o=grid" 
"objectClass" requested
<= root access granted
=> access_allowed: search access granted by manage(=mwrscxd)
Segmentation fault


Restarting slapd does not solve the problem, the refresh phase begins 
and then segfaults on both machines.

I (somewhat jumping the gun) wiped the consumer databases and started 
again and after a day things are going ok.

I have been doing a bit of hunting and can't seem to find any other 
cases of this.


Does anyone have any suggestions, and what more information do i need to 
provide.