[Bug 1030310] Regression in WAL and COMMIT times for pgioperf-bench on ext3

http://bugzilla.suse.com/show_bug.cgi?id=1030310 http://bugzilla.suse.com/show_bug.cgi?id=1030310#c21 --- Comment #21 from Jan Kara <jack@suse.com> --- Ah, now I realized what could be the problem you possibly observe! The output from pgioperf processes is cached in glibc stdio buffers. These can contain significant amount of output. What I have in my private version of pgioperf is that I fflush() the output after each line which deals with the issue for me... -- You are receiving this mail because: You are on the CC list for the bug.