[lkml]   [2003]   [Jan]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectProcesses hang in get_request_wait (2.4.18-19-7.xbigmem)

I have a database (IBM/Informix/Ardent UniVerse) that is running on a
Dell PowerEdge 6600 (stats below). Sometimes when there is a very high
I/O, any database command shell will hang when trying to exit. A ps
shows that the process is in a get_request_wait.

When the process that is causing the high I/O finishes what it is doing,
then about 5-10 seconds later the command shell that was trying to exit
will finish. I've read some threads on the list and elsewhere and seen
some patches that refer to "fixing" get_request_wait and
wait_on_page/wait_on_buffer "bugs."

Could mounting file systems onto directories that are only accessible
after another file system is mounted cause an issue (nested mount
points)? For example I have fs /dev/sdd1 that contains a directory
called 'dbms,' I then mount other filesystems onto directories contained
within dbms/:

/dev/sdd1 /cubs
/dev/sdc1 /cubs/dbms/ACCOUNT
/dev/sde1 /cubs/dbms/16kb

System specs are below. I would try 2.4.20 vanilla, but for some reason
I get a panic WRT some SMP things when 2.4.20 boots, but that's another
issue that I will try to tackle on my own.

Is there anything that I can try (patches, /proc settings, etc.) to help
alleviate this issue? Should I reconfigure my mount points? If more
information is needed, please let me know.

Thanks in advance,

Andrew Rechenberg
Infrastructure Team, Sherman Financial Group

Dell PE6600
(4) Xeon 1.4GHz with HT enabled
Red Hat 7.3
2.4.18-19-7.xbigmem with SD_TIMEOUT in sd.c upped to 90*HZ and
(2) Dell PERC3/QC (megaraid2 2.00.2) with most recent firmware
(1) Dell PERC3/DC (megaraid2 2.00.2)

[root@box ~]# ps -eo pid,stat,pcpu,nwchan,wchan=WIDE-WCHAN-COLUMN -o
args | grep wait
20 DW 0.1 19e8a6 get_request_wait [bdflush]
21 DW 0.3 19e8a6 get_request_wait [kupdated]
1440 S 0.0 1217c8 wait4 -bash
1580 S 0.0 1217c8 wait4 su -
1581 S 0.0 1217c8 wait4 -bash
1748 S 0.0 1217c8 wait4 -bash
1850 S 0.0 1217c8 wait4 /bin/bash
1918 S 0.0 1217c8 wait4 su -
1919 S 0.0 1217c8 wait4 -bash
2158 S 0.0 1217c8 wait4 /bin/bash
2430 S 0.0 1217c8 wait4 -bash
2611 S 0.0 1217c8 wait4 su -
2612 S 0.0 1217c8 wait4 -bash
2869 S 0.0 1217c8 wait4 -bash
3013 S 0.0 1217c8 wait4 su -
3014 S 0.0 1217c8 wait4 -bash
3425 S 0.0 1217c8 wait4 su -
3427 S 0.0 1217c8 wait4 -bash
3534 S 0.0 1217c8 wait4 -bash
3673 S 0.0 1217c8 wait4 su -
3674 S 0.0 1217c8 wait4 -bash
3753 S 0.0 1217c8 wait4 uv
3853 D 6.7 19e8a6 get_request_wait RESIZE -N ALECREDCBI1 18 29826
3882 S 0.0 1217c8 wait4 uv
3913 D 0.0 19e8a6 get_request_wait uv
3929 D 4.4 19e8a6 get_request_wait RESIZE -N PFMODDEBTOR3 2 342313
3939 D 0.2 19e8a6 get_request_wait uv
3954 D 0.1 19e8a6 get_request_wait uv
3956 S 0.0 14d79c pipe_wait grep wait
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
More majordomo info at
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:32    [W:0.024 / U:0.816 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site