Dave Page wrote: > For info, the buildfarm script failed to leave the broken tree behind > again so I was unable to get a dump from the affected index. > > Andrew; My run logs show that the script did think it was leaving the > tree behind (it included the 'leaving error trees' text that you asked > me to add to the script). I'm assuming it does leave the tree under the > buildroot for the branch and doesn't try to copy them into /tmp or > somewhere?
darn. Yes, they are just renamed as pgsqlkeep.<timestr> and instkeep.<timestr>. I will send you a more comprehensive patch to try later today. cheers andrew > > Regards, Dave > > PG Build Farm wrote: >> The PGBuildfarm member narwhal had the following event on branch HEAD: >> >> Status changed from OK to Check failure >> >> The snapshot timestamp for the build that triggered this notification >> is: 2007-05-02 20:00:01 >> >> The specs of this machine are: >> OS: Windows Server 2003 R2 / 5.2.3790 >> Arch: i686 >> Comp: GCC / 3.4.2 (mingw-special) >> >> For more information, see >> http://www.pgbuildfarm.org/cgi-bin/show_history.pl?nm=narwhal&br=HEAD > ---------------------------(end of broadcast)--------------------------- TIP 4: Have you searched our list archives? http://archives.postgresql.org
