Fix a posible hang during health checking
authorYugo Nagata <nagata@sraoss.co.jp>
Wed, 8 Jun 2016 11:59:03 +0000 (20:59 +0900)
committerYugo Nagata <nagata@sraoss.co.jp>
Wed, 8 Jun 2016 11:59:03 +0000 (20:59 +0900)
commited9f2900f1b611f5cfd52e8f758c3616861e60c0
tree575ea5e4f9373f4b60b19b42c391e6d2bbc398c1
parent81b8071f0e0d8ce28c611d963e1c37251c33153b
Fix a posible hang during health checking

Helath checking was hang when any data wasn't sent
from backend after connect(2) succeeded. To fix this,
pool_check_fd() returns 1 when select(2) exits with
EINTR due to SIGALRM while health checkking is performed.

Reported and patch provided by harukat and some modification
by Yugo. Per bug #204.
src/protocol/pool_process_query.c