[Date Prev][Date Next] [Chronological] [Thread] [Top]

Re: test008 failed in HEAD



Aaron Richton wrote:
Given recent checkins, HEAD bdb test008 has ran successfully all day in an
infinite loop. Will let hdb run overnight just in case it's different...

Thanks. Yes, hdb is slightly different but not as far as test008 is concerned. I had a problem in test020 which showed a regression that's been there ever since 2.3 was released, but that was also fixed with the last round of checkins. (hdb had a race that could allow multiple entries with the same DN to be created. Never happened in 2.2.)

On Wed, 5 Dec 2007, Howard Chu wrote:

Hallvard B Furuseth wrote:
2 Quad Core CPUs. Intel Xeon E5335. 2GHz.
The code in HEAD has passed 560 iterations on my quad-socket server and is
still going. It failed after 309 iterations on my single-socket dual-core
machine. On the advice of the BDB folks I've added some more logging inside
the BDB library to try to track it further.
--
  -- Howard Chu
  Chief Architect, Symas Corp.  http://www.symas.com
  Director, Highland Sun        http://highlandsun.com/hyc/
  Chief Architect, OpenLDAP     http://www.openldap.org/project/