Hi, During a jhalfs run of LFS 7.4, I has my attention caught by an unusual duration of the util-linux build. I waited one more hour, then tried to investigate (whithout interrupting the build). Looking at the logs and the test logs, the tests had hanged at "script: race conditions".
A 'ps aux | grep script' command returned: 99 2580 0.0 0.0 9696 1532 ? S 15:15 0:00 /bin/bash /sources/util-linux-2.23.2/tests/ts/script/race --srcdir=/sources/util-linux-2.23.2 --builddir=/sources/util-linux-2.23.2 99 2624 0.0 0.0 9828 808 ? S 15:15 0:00 /bin/bash /sources/util-linux-2.23.2/tests/ts/script/race --srcdir=/sources/util-linux-2.23.2 --builddir=/sources/util-linux-2.23.2 99 4903 0.0 0.0 8360 688 ? S 15:15 0:00 /sources/util-linux-2.23.2/script -q -c printf 'Bingo\n' /sources/util-linux-2.23.2/tests/output/script/race-bingo 99 4904 0.0 0.0 8368 340 ? S 15:15 0:00 /sources/util-linux-2.23.2/script -q -c printf 'Bingo\n' /sources/util-linux-2.23.2/tests/output/script/race-bingo I killed the second (PID 4904) 'script' process, and the build went to completion almost immediately. Amazingly, the test log says: --------------------------------------------------------------------- All 113 tests PASSED --------------------------------------------------------------------- Actually, looking at tests/run.sh, I cannot understand how it is that two '/sources/util-linux-2.23.2/tests/ts/script/race' processes have been launched. OTH, each 'race' script launches 1000 'script' processes, so it is normal to have more. But it seems that the race condition test has failed in this case... Well, I guess it is a race condition test, according to the name of the script ;-) Do not know if it is something to worry about, but thought I should share Pierre -- http://linuxfromscratch.org/mailman/listinfo/lfs-dev FAQ: http://www.linuxfromscratch.org/faq/ Unsubscribe: See the above information page
