biod

[Nobody's Hero]

I am running an oracle import onto some older HP 12H autoraid drives. Its been running forever. Also, I have 2 biod processes taking up 2 cpus. I've never seen this before during an import. We are using an import file from an NFS directory on another server, an Nclass. Not sure what I can do to resolve this issue.

Load averages: 0.88, 0.92, 0.98
139 processes: 126 sleeping, 12 running, 1 zombie
Cpu states:
CPU LOAD USER NICE SYS IDLE BLOCK SWAIT INTR SSYS
0 0.79 2.4% 0.0% 7.8% 89.8% 0.0% 0.0% 0.0% 0.0%
1 1.00 0.0% 0.0% 97.1% 2.9% 0.0% 0.0% 0.0% 0.0%
2 0.87 12.6% 0.0% 1.0% 86.4% 0.0% 0.0% 0.0% 0.0%
3 0.87 0.0% 0.0% 98.5% 1.5% 0.0% 0.0% 0.0% 0.0%
--- ---- ----- ----- ----- ----- ----- ----- ----- -----
avg 0.88 3.9% 0.0% 51.2% 44.9% 0.0% 0.0% 0.0% 0.0%

Memory: 271584K (227156K) real, 373564K (232128K) virtual, 1546632K free Page# 1/5

CPU TTY PID USERNAME PRI NI SIZE RES STATE TIME %WCPU %CPU COMMAND
3 ? 1255 root 154 20 0K 16K run 858:22 83.49 83.35 biod
1 ? 1256 root 154 39 0K 16K run 857:15 83.30 83.15 biod
2 ? 24670 oracle 149 0 31368K 30624K sleep 349:44 16.73 16.70 imp

[Michael Steele_2]

First, lets see if 'biod' is waiting on an open file / I/O:

lsof -p pid
-or-
lsof -p 1255
lsof -p 1256
lsof -p 24670

Second, check for bottlenecks:

sar -u 5 5
sar -d 5 5
sar -v 5 5
sar -b 5 5
vmstat 5 5
swapinfo -tam
kmtune -q dbc_max -l
kmtune -q dbc_min -l

[Nobody's Hero]

dbc max and min are 7 and 2.
No disk bottle necks. Just a lot of system resources pegged out, but not memory. Memory seems ok.

Also, not waiting on an open file. I did check that before I posted this. Im at a loss here.

[Nobody's Hero]

[ihshp2:/sbin/init.d]# vmstat 5 5
procs memory page faults cpu
r b w avm free re at pi po fr de sr in sy cs us sy id
4 0 0 58073 387787 113 37 0 0 47 0 0 1194 -713 370 8 15 78
6 1 0 59041 387813 283 56 2 0 57 0 0 1292 6480 418 9 69 22
5 1 0 57730 387887 261 49 0 0 51 0 0 1276 5464 358 8 72 20
6 0 0 56395 387860 273 63 0 0 63 0 0 1268 5130 368 8 71 21
3 0 0 56427 387662 387 66 0 0 67 0 0 1289 6579 441 22 60 18
[ihshp2:/sbin/init.d]#

[Robert-Jan Goossens]

Hi,

Take a look at this doc,

Date: 5/5/00
Document description: HP-UX 10.X/11.0 NFS Performance Assessment Tips - November 2000
Document id: KBAN00000261

search inside the doc for biod,

http://www5.itrc.hp.com/service/cki/docDisplay.do?docLocale=en_US&docId=200000062684158

Kind regards,

Robert-Jan.

[Elena Leontieva · Accepted solution]

biods are the primary NFS client daemon. on HP-UX 11.00 the default number is 4, on 11.i - 16.
The number of biods is configured via the NUM_NFSIOD in the /etc/rc.config.d/nfsconf file. The number can be increased on the fly by N:
/usr/sbin/biod N ( will give you total=current +N); to decrease the number of biods, - you need to reboot the server.

Since you never seen this problem before, it could be an NFS issue, try the following:
nfsstat -c
nfsstat -m

[Massimo Bianchi]

Hi,
do you have enough psace on your local disks to host the dump ?

It may well worth it...

Massimo

[Nobody's Hero]

Thanks Elena,

increasing the number of biod 's helped level the problem, and now I am gett more io.

Thanks,
Bob