Lot of traffic after installing bind 8.4.3 on sparc

Paige Stafford staffordp1 at ornl.gov
Thu Dec 4 14:32:42 UTC 2003


Sorry, different time zone...  After reading through the threads...

We upgraded from 8.4.1 to 8.4.3 on our Sun, Solaris 8 (one CPU).  Since
then, we too are seeing a higher CPU usage (54.49% vs 0.18%) and alot of
AAAA queries.  Here's a snapshot of a tcpdump from yesterday:

13:07:49.207725 ns1.ornl.gov.61723 > ns1.bbbdns.com.domain:  58581 AAAA? ns1.bbbdns.com. (32) (DF)
13:07:49.208005 ns1.ornl.gov.61723 > g.gtld-servers.net.domain:  24573 AAAA? ns2.bbbdns.com. (32) (DF)
13:07:49.208662 ns1.ornl.gov.61723 > 12.129.90.126.domain:  47480 [1au] AAAA? ns1.reallifeanswers.com. OPT  UDPsize=4096 (52) (DF)
13:07:49.208839 ns1.ornl.gov.61723 > prod-ns-knox01.ctimi.com.domain:  56047 [1au] AAAA? ns4.reallifeanswers.com. OPT  UDPsize=4096 (52) (DF)
13:07:49.209552 ns1.ornl.gov.61723 > g.gtld-servers.net.domain:  54029 AAAA? ns1.reallifeanswers.com. (41) (DF)
13:07:49.210288 ns1.ornl.gov.61723 > alice.adsnet.com.domain:  3856 AAAA? alice.adsnet.com. (34) (DF)
13:07:49.210381 ns1.ornl.gov.61723 > ns1.adsnet.com.domain:  46494 AAAA? ns1.adsnet.com. (32) (DF)
13:07:49.210728 ns1.ornl.gov.61723 > 69.57.163.218.domain:  59594 AAAA? ns2.vgeek.net. (31) (DF)
13:07:49.210961 ns1.ornl.gov.61723 > 69.57.163.217.domain:  62185 AAAA? ns1.vgeek.net. (31) (DF)
13:07:49.211180 ns1.ornl.gov.61723 > 69.57.163.217.domain:  55105 AAAA? ns2.vgeek.net. (31) (DF)
13:07:49.211400 ns1.ornl.gov.61723 > 69.57.163.218.domain:  29135 AAAA? ns1.vgeek.net. (31) (DF)
13:07:49.212055 ns1.ornl.gov.61723 > alice.adsnet.com.domain:  44331 AAAA? ns1.adsnet.com. (32) (DF)
13:07:49.212148 ns1.ornl.gov.61723 > ns1-auth.sprintlink.net.domain:  24652 AAAA? alice.adsnet.com. (34) (DF)
13:07:49.212693 ns1.ornl.gov.61723 > prod-ns-knox01.ctimi.com.domain:  15932 [1au] AAAA? ns1.reallifeanswers.com. OPT  UDPsize=4096 (52) (DF)
13:07:49.213006 ns1.ornl.gov.61723 > 66.220.30.195.domain:  44494 AAAA? ns1.kalsey.com. (32) (DF)
13:07:49.213362 ns1.ornl.gov.61723 > g.gtld-servers.net.domain:  65060 AAAA? ns2.kalsey.com. (32) (DF)
13:07:49.215161 ns1.ornl.gov.61723 > 12.129.90.126.domain:  10783 AAAA? ns4.reallifeanswers.com. (41) (DF)

Sounds like we have a good work around by turning off IPv6 type queries,
the '-4' option.

We extend our appreciation to the list in solving this problem.


----------------------------

On Thu, 4 Dec 2003, Osmo Laitinen wrote:

 > On Wed, Dec 03, 2003 at 06:48:33PM -0500, Paige Stafford wrote:
 > >
 > > We at ORNL are experiencing the same problem.
 >
 > I can also confirm this on Solaris 8.
 >
 > Without "-4" on commandline:
 > PID USERNAME THR PRI NICE  SIZE   RES STATE    TIME    CPU COMMAND
 > 2288 dns        1  10    0   21M   14M run      2:01 83.42% named
 >
 > And with "-4":
 > PID USERNAME THR PRI NICE  SIZE   RES STATE    TIME    CPU COMMAND
 > 2319 dns        1  22    0   27M   21M sleep    4:09 27.66% named
 >
 > --
 > T:Ossi
 >
 >


More information about the bind-users mailing list