Bug #3518
closedSometimes DHCP hostname registration does not work for a newly registered host
0%
Description
Even if the setup works most of the time (i.e. hosts register their hostname through DHCP, which can then be resolved with hostname.domain.com), sometimes it fails to work (a dig @routerip fails).
Failed resolve:
host myhost Host myhost not found: 3(NXDOMAIN)
By watching the systems logs (resolver), normally when a new host registers, you can see "dnsmasq73263: read /etc/hosts - 220 addresses", but when it fails nothing happens anymore. I checked that /var/dhcpd/var/db/dhcpd.leases contains the new lease, but not the /etc/hosts nor /var/etc/hosts.
I am guessing that for an unknown reason, the HUP signal (by dhcpd ?) is not being sent properly. I don't see anything obvious in the logs.
My current workaround is to create a cron job, which triggers a dnsmasq dhcp hostfile refresh, by sending a HUP signal every minute. Not nice but seems to work.
1 * * * * root /bin/pkill -HUP -f dnsmasq
Our setup is quite standard i guess (SOHO type), with a 192.168.1-100 reserved range (for static & static dhcp), 100-254 range available for dhcp, and some static DNS overrides defined in the DNS forwarder.
Settings:- DHCP server enabled
- DNS Forwarder enabled, with the following subfeatures:
- Register DHCP leases in DNS forwarder
- Register DHCP static mappings in DNS forwarder
Updated by Chris Buechler over 11 years ago
- Status changed from New to Feedback
- Priority changed from High to Normal
- Target version deleted (
2.1.1)
Generally speaking, it works. Do you have a specific scenario where it's replicable?
Updated by Florent Thiery over 11 years ago
Unfortunately, i don't have other deployment or replication method. I am suspecting that maybe one of the hostnames triggers a character/parsing issue which breaks the automatic update of the db.
Updated by Chris Buechler over 10 years ago
- Status changed from Feedback to Duplicate
- Affected Version deleted (
2.1)
root problem is #4931