lkml.org 
[lkml]   [1996]   [Sep]   [19]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
Subjectipx support problem
As of 2.0.19, there's a problem with ipx support that wipes out the some process
queue. The stalled processes look something like this: (excerpt from ps -ajx):

29083 29091 8 8 ? -1 SW 1000 0:00 (M-POSTSCRIPT)
29083 29092 8 8 ? -1 S 1000 0:00 (print)
29092 29093 8 8 ? -1 DW 1000 0:00 (nprint)
29158 29159 8 8 ? -1 SW 1000 0:00 (printmessage)
29159 29160 8 8 ? -1 SW 1000 0:00 (awk)
29159 29161 8 8 ? -1 S 1000 0:00 (print)
29161 29169 8 8 ? -1 SW 1000 0:00 (M-POSTSCRIPT)
29161 29170 8 8 ? -1 S 1000 0:00 (print)
29170 29171 8 8 ? -1 DW 1000 0:00 (nprint)
29286 29287 8 8 ? -1 SW 1000 0:00 (printmessage)
29287 29289 8 8 ? -1 SW 1000 0:00 (awk)
29287 29290 8 8 ? -1 S 1000 0:00 (print)
29290 29298 8 8 ? -1 SW 1000 0:00 (M-POSTSCRIPT)
29290 29299 8 8 ? -1 S 1000 0:00 (print)
29299 29300 8 8 ? -1 DW 1000 0:00 (nprint)

Unfortunately, this is a rather easy problem to reproduce (significant use of
ipx resources causes it). Also, unfortunately, once ipx is stuck, any process
which attempts to use it also gets stuck. Also, unfortunately, any process
which attempts to wait on a stuck process also gets stuck (this does bad
things to init).

Let me know if there's anything I should explicitly test for.

[Oh, one other thing: any process which attempts to ptrace a stuck process gets
stuck.]

--
Raul

\
 
 \ /
  Last update: 2005-03-22 13:37    [W:0.023 / U:1.444 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site