netatalk and the dreaded nbp_rgstr
Andy McFadden
GnoRPM. No problems there. However, several seconds after the system
comes up, I get:
nbp_rgstr: Connection timed out
Can't register quake:Workstation@*
nbp_rgstr: Connection timed out
Can't register quake:netatalk@*
I found related Q&A on the netatalk FAQ-O-Matic, but neither of the two
suggestions (put an explicit "eth0" in /etc/atalk/atalkd.conf, and
uninstall/reinstall) had any effect.
I found one item on a netatalk message digest, but it was just a question
from 1.5 years ago from a guy whose machine crashed and then started
emitting these messages after the machine rebooted.
Has anybody figured out what this is all about? Internet searches on
"nbp_rgstr" turn up few hits (two hits in altavista, three hits from
hotbot, half a dozen articles on dejanews), but nobody seems to have an
answer.
The Ethernet card is a 3c59x "vortex"; ifconfig -a shows:
eth0 Link encap:Ethernet HWaddr 00:60:97:D8:CC:7D
inet addr:172.17.131.58 Bcast:172.17.131.63 Mask:255.255.255.248
EtherTalk Phase 2 addr:65280/156
UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
RX packets:3856 errors:0 dropped:0 overruns:0 frame:0
TX packets:3509 errors:0 dropped:0 overruns:0 carrier:0
collisions:0
lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
EtherTalk Phase 2 addr:0/0
UP LOOPBACK RUNNING MTU:3924 Metric:1
RX packets:125309 errors:0 dropped:0 overruns:0 frame:0
TX packets:125309 errors:0 dropped:0 overruns:0 carrier:0
collisions:0
Red Hat 6.0 appears to be Linux 2.2.5.
I believe the NBP registration is the machine talking to itself, which
makes me wonder if the lo config has more to do with it than the eth0. I
suppose it depends on how nbprgstr works.
Suggestions would be appreciated.
--
Send mail to fad...@netcom.com (Andy McFadden)
CD-Recordable FAQ - http://www.fadden.com/cdrfaq/ (a/k/a www.spies.com/~fadden)
Fight Internet Spam - http://spam.abuse.net/spam/ & news.admin.net-abuse.email
Andy McFadden
Andy McFadden <fad...@netcom.com> wrote:
>I've installed netatalk-1.4b2+asun2.1.3-4.i386.rpm under Red Hat 6.0 using
>GnoRPM. No problems there. However, several seconds after the system
>comes up, I get:
>
>nbp_rgstr: Connection timed out
>Can't register quake:Workstation@*
>nbp_rgstr: Connection timed out
>Can't register quake:netatalk@*
Michael Casteel pointed me in the right direction. I have an Ascend
Pipeline 25 that I use to bridge into work. During the AppleTalk init
stuff, it sends out some messages that wake up the connection. When it
goes looking for "rtmp_packet router"s, it finds one, and tries to do
something clever. This cleverness fails, probably because the Pipeline 25
is configured to ignore AppleTalk traffic (to prevent broadcasts from
holding the connection up unnecessarily).
By unplugging the LAN or WAN connection from the Pipeline 25, I was able
to go from this:
Aug 1 14:07:58 quake kernel: NET4: AppleTalk 0.18 for Linux NET4.0
Aug 1 14:07:58 quake atalkd[478]: restart (1.4b2+asun2.1.3)
Aug 1 14:07:58 quake rc: Starting linuxconf succeeded
Aug 1 14:07:58 quake rc: Starting local succeeded
Aug 1 14:07:59 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:08 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:15 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:18 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:25 quake last message repeated 4 times
Aug 1 14:08:28 quake atalkd[478]: config for no router
Aug 1 14:08:29 quake atalkd[478]: ready 0/0/0
Aug 1 14:08:29 quake atalk: atalkd startup succeeded
Aug 1 14:08:33 quake atalkd[478]: rtmp_packet router has become available
Aug 1 14:08:33 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:44 quake last message repeated 6 times
Aug 1 14:08:45 quake atalk: papd startup succeeded
Aug 1 14:08:45 quake papd[634]: restart (1.4b2+asun2.1.3)
Aug 1 14:08:45 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:45 quake atalk: afpd startup succeeded
Aug 1 14:08:48 quake atalkd[478]: zip_getnetinfo for eth0
Aug 1 14:08:54 quake last message repeated 3 times
Aug 1 14:08:55 quake afpd[645]: Can't register quake:AFPServer@*
Aug 1 14:08:55 quake afpd[645]: ASIP started on 172.17.131.58:548(0) (1.4b2+asu
n2.1.3)
to this:
Aug 1 14:04:59 quake kernel: NET4: AppleTalk 0.18 for Linux NET4.0
Aug 1 14:04:59 quake atalkd[472]: restart (1.4b2+asun2.1.3)
Aug 1 14:04:59 quake rc: Starting linuxconf succeeded
Aug 1 14:04:59 quake rc: Starting local succeeded
Aug 1 14:05:00 quake atalkd[472]: zip_getnetinfo for eth0
Aug 1 14:05:09 quake atalkd[472]: zip_getnetinfo for eth0
Aug 1 14:05:19 quake atalkd[472]: zip_getnetinfo for eth0
Aug 1 14:05:29 quake atalkd[472]: config for no router
Aug 1 14:05:30 quake atalkd[472]: ready 0/0/0
Aug 1 14:05:30 quake atalk: atalkd startup succeeded
Aug 1 14:05:42 quake atalk: papd startup succeeded
Aug 1 14:05:43 quake papd[533]: restart (1.4b2+asun2.1.3)
Aug 1 14:05:43 quake atalk: afpd startup succeeded
Aug 1 14:05:49 quake afpd[544]: quake:AFPServer@* started on 65280.77:128 (1.4b2+asun2.1.3)
Aug 1 14:05:49 quake afpd[544]: ASIP started on 172.17.131.58:548(1) (1.4b2+asun2.1.3)
If this router stuff is only checked when the system boots, it might explain
why some people experience a situation where their server runs fine for a
long time, then starts failing when they reboot. Something appeared on the
network since the last boot that confuses atalkd during initialization.