I'm experimenting with a UDP broadcast block on a couple of Sun Blade 150s running Solaris 9 and have come up with the following rule which seems to drop the traffic:

block in on eri proto udp from any to 192.168.1.255

However, the available system memory gradually depletes, eventually resorting to swap, and finally resulting in a hard freeze after a day or so. The degradation can be monitored with "netstat -m", showing rising mblk and dblk counts, along with a growing amount of kbytes allocated for streams data.

I've tried adding a /32 mask to the rule, as well as a port range, and neither seem to effect the problem. If anyone can offer some guidance, it'd be much appreciated. Below is some (possibly) relevant information.


Thanks, Roy


----------

system1# uname -a
SunOS system1 5.9 Generic_117171-02 sun4u sparc SUNW,Sun-Blade-100

system2# uname -a
SunOS system2 5.9 Generic_112233-08 sun4u sparc SUNW,Sun-Blade-100

system1&2# isainfo -vk
64-bit sparcv9 kernel modules

system1&2# modinfo | grep ipf
107 7807e000  37532 xxx   1  ipf (IP Filter: v4.1.3)

system2# ipfstat [about a quarter of the way towards oblivion]
bad packets: in 0 out 0
IPv6 packets: in 0 out 0
input packets: blocked 201973 passed 528580 nomatch 24752 counted 0 short 0
output packets: blocked 411 passed 312940 nomatch 0 counted 0 short 0
input packets logged: blocked 0 passed 423197
output packets logged: blocked 409 passed 251471
packets logged: input 0 output 0
log failures: input 111403 output 149720
fragment state(in): kept 0 lost 0 not fragmented 0
fragment state(out): kept 0 lost 0 not fragmented 0
packet state(in): kept 45 lost 162
packet state(out): kept 3700 lost 2
ICMP replies: 0 TCP RSTs sent: 0
Invalid source(in): 0
Result cache hits(in): 126943 (out): 324
IN Pullups succeeded: 0 failed: 0
OUT Pullups succeeded: 0 failed: 0
Fastroute successes: 0 failures: 0
TCP cksum fails(in): 0 (out): 0
IPF Ticks: 196468
Packet log flags set: (0)
none


system2# netstat -m [about a quarter of the way towards oblivion]
streams allocation:
                                         cumulative  allocation
                      current   maximum       total    failures
streams                   195       252        6084           0
queues                    441       552       11768           0
mblk                   206662    207391      284267           0
dblk                   206658    208145     5903250           0
linkblk                     6       169          11           0
strevent                   11       169      110077           0
syncq                      15        48          64           0
qband                       0         0           0           0

144280 Kbytes allocated for streams data



Reply via email to