https://bugzilla.kernel.org/show_bug.cgi?id=203247 --- Comment #8 from pkb07815 (pankajcabs@xxxxxxxxx) --- issue is reproducible: I have use dd command to generate file in /data area to compete with the postgres operation and using iostat -x to monitor the io condition. Here is how I create the dd with variation of count value and sleep time: for i in {1..100}; do nice -n 10 dd if=/dev/zero of=/data/tmp/testonly/tmpfile$i.txt count=1024 bs=1048576; echo "done tmpfile $i"; sleep 10; done & I use postgresql utility pgbench paired with other in-house sql commands for test of transactions: su -s /bin/bash postgres -c "/usr/postgresql/bin/pgbench -T 90000 -c 8 bench2" & When I execute the pgbench and dd both..I see the processes going into D state and stuck there for long (atleast for 140 seconds). -- You are receiving this mail because: You are watching the assignee of the bug.
