: Using domain name addressing (e.g. www.mysubdomain1.com
), my virtually hosted domains are resolving correctly some days and then not resolving at other times. When they don't resolve, I can always access them via direct ip addressing (e.g. http://126.96.36.199/~username
). So it's not that the web server is down at the time.
All of the virtually hosted domains use the same nameservers, which are, the namerservers of the parent reseller account I own (e.g. the nameservers for mysubdomain1.com are set to ns1.myparentdomain.com and ns2.myparentdomain.com; for mysubdomain2.com they are also set to ns1.myparentdomain.com and ns2.myparent domain.com, etc.)
I never have problems accessing myparentdomain.com via domain name addressing. But all of my subdomains play peek-a-boo if addressed as www.mysubdomainx.com
: I run all of my sites on a VPS hosted by Hostican. I have a reseller account there (arcalinea.com) underwhich I host several subdomains. Hostican made some server upgrades a while back, moved my domains to another server, changing the IP address of the parent reseller account in the process. Since then, I've had the problem described above. Before Hostican's server upgrade, the approach I had used to set nameservers at my various subdomain registrars had worked correctly.
My sites are setup as follows. My main account on the VPS is www.arcalinea.com
(this is the reseller account). I then have several virtually hosted domains under arcalinea's IP address (this is how I was setup previously with no problems).
After Hostican changed arcalinea's IP address, I updated the nameservers at arcalinea's registrar (which is domainsatcost.com) and registered the new IP addresses at this registrar. So at domainsatcost.com, arcalinea.com is defined as having nameservers ns1.arcalinea.com and ns2.arcalinea.com and those are registered to 188.8.131.52 and 184.108.40.206 respectively.
At this stage of the process, everything works as planned. www.arcalinea.com
resolves correctly in a browser. Also,running a trace (say at www.network-tools.com
) returns a valid DNS records and a complete trace for www.arcalinea.com
The main problem is with all of the domains hosted virtually under account arcalinea.com at hostican (those are the domains i refer to above as mysubdomain1.com, mysubdomain2.com ,etc.). For example: cadstudio3.com, healingparadigms.ca and maryjaneporter.ca are some of the domains hosted under arcalinea.
Everyone of those virtually hosted domains has its nameservers set the same way: they each are set to ns1.arcalinea.com and ns2.arcalinea.com
Even at times where those virtually hosted domains are resolving correctly in my browser, if I run a check on that domain at www.network-tools.com
(using say the Express option), I get several errors such as:
'IP address: Error, try again'
'TraceRoute' is empty.
'Attempt to get a DNS server for maryjaneporter.ca failed. The query returned a server failure'
But, the Whois query for maryjaneporter.ca in the same express report shows the nameservers to be correctly defined as:
This of course begs the question, if there are no DNS records available, how is the browser able to correctly resolve (some of the times) to www.maryjaneporter.ca
(or any of the other subdomains in question)?
It seems to me that the problem is with the way the nameservers, hosted at Hostican, are behaving or are configured. Trying to get this resolved by/with Hostican has been an exercise in frustration, ranging from them denying there was a problem to then acknowledging there was a problem, to then suggesting the problem was with my ISP (yet other friends and colleagues on different ISPs than mine have the same issue trying to resolve to my virtually hosted domains; some days it works, other days it doesn't).
The latest from Hostican is "Judging by the symtoms of this, most likely named/bind is dieing on your VPS. You will want to look through your named/bind logs to determine the root cause.
You will want to look at /var/log/messages or /var/log/named specifically around the time your domains stop resolving.".
This is the first, after many less than useful answers that hints at a possible problem area. Unfortunately, II don't know where to find those logs (a friend has suggested I may not even have access to them as they may be higher up in the server's root, beyond my reseller account's root).
I've tried finding these logs on my reseller account via the WHM as well as on the hosted accounts via their respective CPanels but no luck (they might be there but I can't find them).
I have WHM access to my reseller account's root which includes DNS Functions such as 'Edit DNS Zone, Perform a DNS Cleanup, Synchronize DNS Records' and more, along with the ability to restart the DNS Server (BIND/NSD) but I am reluctant to just start initiating functions when I'm not certain of what I'm doing,
I'm in the middle of some projects and can't afford to lose access to those sites if something screws up.
Can anyone help me sort this out?