Browsing No - causes high CPU use on client

[angelb]

Hello all.

Under CUPS 1.2.4, when Browsing is set to No in a client system, CPU
use spikes to the roof:

ROOT @ SAPTI02 # ps -ef|grep -i cups
    root 708680      1 101 10:49:24      -  1:05 /opt/TWWfsw/cups124/sbin/cupsd -c /etc/opt/TWWfsw/cups124/cupsd.conf
    cups 725180 708680   0 10:49:25      -  0:00 cups-polld occam522 631 2700 631
    cups 811092 708680   0 10:49:25      -  0:00 cups-polld occam503 631 2700 631

ROOT @ SAPTI02 # vmstat 2
System Configuration: lcpu=2 mem=6144MB
kthr     memory             page              faults        cpu
----- ----------- ------------------------ ------------ -----------
 r  b   avm   fre  re  pi  po  fr   sr  cy  in   sy  cs us sy id wa
 1  1 705545 278980   0   0   0   4   17   0 315 6111 134  3  6 90  1
 1  0 705550 278975   0   0   0   0    0   0 315 5357 160  4 47 48  1
 1  0 705550 278975   0   0   0   0    0   0 310 8532 150  6 47 47  1
 1  0 705550 278975   0   0   0   0    0   0 314 5371 150  7 46 47  1
 1  0 705550 278975   0   0   0   0    0   0 321 5421 167  8 44 48  1
 1  0 705550 278975   0   0   0   0    0   0 328 6797 163  7 45 48  0
 1  0 705550 278975   0   0   0   0    0   0 342 7552 169  6 48 46  1
 1  0 705550 278975   0   0   0   0    0   0 308 6397 165  4 47 48  1


If Browsing is set to Yes, CPU use is normal:

System Configuration: lcpu=2 mem=6144MB
kthr     memory             page              faults        cpu
----- ----------- ------------------------ ------------ -----------
 r  b   avm   fre  re  pi  po  fr   sr  cy  in   sy  cs us sy id wa
 1  1 705546 279072   0   0   0   4   17   0 315 6111 134  3  6 90  1
 0  0 705551 279067   0   0   0   0    0   0 321 2673 121  2  2 96  1
 0  0 705551 279067   0   0   0   0    0   0 312 1108 100  1  1 97  1
 0  0 705551 279067   0   0   0   0    0   0 307 1115 105  1  2 94  3
 0  0 705551 279067   0   0   0   0    0   0 313 2167 110  1  2 97  1
 0  0 705551 279067   0   0   0   0    0   0 308 2138 103  1  2 96  1
 0  0 705551 279067   0   0   0   0    0   0 310 2178 111  2  2 95  1
 1  0 705551 279064   0   0   0   0    0   0 310 15370 122 13  5 80  1
 0  0 705551 279064   0   0   0   0    0   0 315 1184 112  1  1 97  0
 1  0 712209 272406   0   0   0   0    0   0 420 13444 298  2 21 74  4
 1  0 705551 279064   0   0   0   0    0   0 330 2044 128  0  5 94  1
 1  0 705551 279064   0   0   0   0    0   0 321 3246 114  3  2 95  0
 0  0 705551 279064   0   0   0   0    0   0 331 1283 129  2  3 94  2

This problem can be duplicated under AIX or Solaris systems.

Has anyone else seen the same?

On another note, load-balance and fail-over work as wonderfuly now.

Thanks,
Angel