I'm taking another crack at debugging my DNS update problem with my
local scalr installation. I've noticed a couple things which seem
peculiar, but I'm not sure how it is supposed to function. So, I've a
couple questions of the experts :) :
1. On Ubuntu, I didn't actually add the zone information to named.conf
directly, as it suggests to place it in named.conf.local. So, in /etc/
bind/named.conf.local I've got:
zone "cloud.example.com" {
type master;
file "/etc/bind/cloud.example.com.db";
allow-transfer { none; };
};
I've told scalr that the file to update is named.conf.local.
Now, sometimes it manages to update the cloud.example.com.db file just
fine, but then it often fails to do so and remains that way and
nothing I can do from the web UI will force it.
To attempt to force it I issued the SQL command:
update zones set hosts_list_updated=0
and sure enough the cron task DNSZoneListUpdateProcess.php actual did
something, but not the right thing!
It:
a. changed the zone file parameter above to
"cloud.example.com.db" (i.e. removed the absolute path), which caused
this to appear in the syslog:
Jul 30 01:21:04 (none) named[21213]: received control channel command
'reload'
Jul 30 01:21:04 (none) named[21213]: loading configuration from '/etc/
bind/named.conf'
Jul 30 01:21:04 (none) named[21213]: max open files (1024) is smaller
than max sockets (4096)
Jul 30 01:21:04 (none) named[21213]: using default UDP/IPv4 port
range: [1024, 65535]
Jul 30 01:21:04 (none) named[21213]: using default UDP/IPv6 port
range: [1024, 65535]
Jul 30 01:21:04 (none) named[21213]: zone cloud.example.com/IN:
(master) removed
Jul 30 01:21:04 (none) named[21213]: reloading configuration succeeded
Jul 30 01:21:04 (none) named[21213]: zone cloud.example.com/IN:
loading from master file cloud.example.com.db failed: file not found
Jul 30 01:21:04 (none) named[21213]: reloading zones succeeded
(notice the "file not found")
b. It didn't touch the cloud.example.db file at all. So, it still
reads:
; !Automatically generated by LibWebta (www.webta.net) - do not edit!
; Zone file for cloud.apez.biz.
$ORIGIN cloud.apez.biz.
@ 300 IN SOA ns1.example.com.
example.com.secretregistration.com. (
2009072920 ; serial, todays date+todays
600 ; refresh, seconds
7200 ; retry, seconds
3024000 ; expire, seconds
60 ) ; minimum, seconds
cloud.example.com. 14400 IN NS ns1.example.com.
www 60 IN CNAME ext-www.cloud.example.com.
(obviously I've changed my domain to example.com to protect the
innocent :) )
However - the thing to notice is that there is no A record for www (!)
(www is the only role in the application currently, though that is not
significant as it occurs with a typical compliment of
app,www,mysqllvms)
Can someone shed some light on the best way to setup the bind files so
that Scalr can do the right thing?
2. I noticed in the Scalr system log:
927096 INFO 2009-07-30 01:21:04 Open stream to: ssh2.sftp://
CONNECTION/etc/bind/named.conf.local
927097 INFO 2009-07-30 01:21:04 Reading: /etc/bind/named.conf.local
927098 INFO 2009-07-30 01:21:04 Updating list of allowed hosts for
'cloud.example.com' on 'ns1.example.com'
927102 INFO 2009-07-30 01:21:04 Reloading bind on
'ns1.example.com'!
927103 INFO 2009-07-30 01:21:04 Execute rndc reload
927104 INFO 2009-07-30 01:21:04 RNDC reload result: server reload
successful
Is that "CONNECTION" correct? or should it say "localhost" or
something. If so, where it that set? (it appears to be able to modify
the bind file, so it must be actually connecting I suppose)
Thank you for any help in advance.
-David.
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"scalr-discuss" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/scalr-discuss?hl=en
-~----------~----~----~----~------~----~------~--~---
