Messages in this thread | | | Date | Thu, 8 Jul 1999 14:47:04 +0200 | Subject | something terribly rotten with scheduler? | From | Marc Lehmann <> |
| |
Scenario:
- linux 2.2.10 on an dual SMP intel machine (gcc-2.7.2.3, no patches) - crack in network mode (i.e. two processes with high user load):
000 0 614 1 20 19 1836 1188 - RN ? 117:21 cracker -kill run/RK1-localhost -from 127 -nice 20 000 0 692 1 19 19 1528 880 - RN ? 117:14 cracker -kill run/RK2-localhost -from 127 -nice 20
As you can see, they are niced. No problem, usually.
- X-server runs in real time priority, if that is a concern
Now _sometimes_ (rarely), other processes hang. Like this:
0 9352 0000000000000102 0000000000000000 0000000080000000 0000000000000000 D ttyp7 0:00 xv -nodecor pgcc.png
As you can see from the pending signals mask (102) kill -9 was not very effective in killing it. However, a "killall -STOP cracker" _immediately_ released the locked processes. How can that be? How can two nice 20 processes totally block normal processes?
BTW, this only happened with 2.2.10, but I didn't run crack for a long time so it might not be version dependent.
-- -----==- | ----==-- _ | ---==---(_)__ __ ____ __ Marc Lehmann +-- --==---/ / _ \/ // /\ \/ / pcg@goof.com |e| -=====/_/_//_/\_,_/ /_/\_\ XX11-RIPE --+ The choice of a GNU generation | |
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |