Can't get to my site or main knownhost site
- Thread starter Abstr
- Start date
We haven't had any network connectivity problems yesterday and I'm also not seeing and drops in graphs at around the time when this thread was created.
I guess you both might be on the same ISP or route was (is) going thourgh the same transit provider and some sort of local problem happened somewhere on the way. What can use useful in such cases is the output of the "tracert domain.com" command where domain.com needs to be replaced with the actual name of the domain you're trying to reach. Output generated by the traceroute utility executed at the time when problem is experience most likely will show where the problem is.
Trust all your VPS hosting needs to US based KnownHost.
I guess you both might be on the same ISP or route was (is) going thourgh the same transit provider and some sort of local problem happened somewhere on the way. What can use useful in such cases is the output of the "tracert domain.com" command where domain.com needs to be replaced with the actual name of the domain you're trying to reach. Output generated by the traceroute utility executed at the time when problem is experience most likely will show where the problem is.
Trust all your VPS hosting needs to US based KnownHost.
I doubt that Paul. The reason being is that I'm on WildBlue(in the middle of nowhere Mississippi) and my customer is in Colorado. I know she isn't on Wildblue and I had her try knownhost.com as well to see what the problem was. I can give you more information if you wish but the odds of Abstr being on WildBlue is slim to none. Lol, it bites but when it is all you have besides dial up...............
It was strange though....billing.knownhost.com and forums.knownhost.com were accessible. Knownhost.com and support.knownhost.com were not for almost an hour. I got my customer's server back with a reboot and then had to manually restart the services. Fluke maybe? Maybe someone else will chime in with what they experienced. You had several logged in that viewed this thread but didn't respond.
It was strange though....billing.knownhost.com and forums.knownhost.com were accessible. Knownhost.com and support.knownhost.com were not for almost an hour. I got my customer's server back with a reboot and then had to manually restart the services. Fluke maybe? Maybe someone else will chime in with what they experienced. You had several logged in that viewed this thread but didn't respond.
forums.knownhost.com is hosted elsewere outside of all 3 of our networks but billing.knownhost.com and knownhost.com are hosted on the very same system in Dallas, TX so it is quite strange that you were able to access billing but not the knownhost.com site itself.
Is there any chance you were able to run traceroute at the moment when problem happened?
If reboot was done this would mean that you were able to establish the network connection to the system, so it sounds even stranger than before.
What I'm seeing so far is:
- network usage graphs do not show any drops yesterday;
- hyperspin didn't pickup / report any problems with any of the servers and/or network devices yesterday;
- I reviewed all tickets from yesterday and found 3 tickets submitted at around the time when this thread was created. These 3 tickets were about timed out error happening for customer attemting to access the sites hosted. All tickets were processed, customers sites / systems were checked and no problems were found.
Based on these 3 points above I, again, would think about local problems with single ISP (doesn't seem to be the case taking in account your location) and/or localized problem with one of the transit providers. Again, all of this is a pure speculation at this point. What could have helped is traceroute output which suppose to show where timeout occurs.
Another easy way to see if "everything is down" or if there is a trouble with the specific location is to run the site through these resources:
- http://downforeveryoneorjustme.com/
- http://alertra.com/spotcheck.php
Is there any chance you were able to run traceroute at the moment when problem happened?
If reboot was done this would mean that you were able to establish the network connection to the system, so it sounds even stranger than before.
What I'm seeing so far is:
- network usage graphs do not show any drops yesterday;
- hyperspin didn't pickup / report any problems with any of the servers and/or network devices yesterday;
- I reviewed all tickets from yesterday and found 3 tickets submitted at around the time when this thread was created. These 3 tickets were about timed out error happening for customer attemting to access the sites hosted. All tickets were processed, customers sites / systems were checked and no problems were found.
Based on these 3 points above I, again, would think about local problems with single ISP (doesn't seem to be the case taking in account your location) and/or localized problem with one of the transit providers. Again, all of this is a pure speculation at this point. What could have helped is traceroute output which suppose to show where timeout occurs.
Another easy way to see if "everything is down" or if there is a trouble with the specific location is to run the site through these resources:
- http://downforeveryoneorjustme.com/
- http://alertra.com/spotcheck.php
I didn't run traceroute because she was burning up my phone and I wanted her gone asap. Since a simple reboot will take care of processes gone wild, that was my first step. She was reporting "out of memory" errors but I got the standard connection timed out error. I have several VPS(s) with you guys and all were up but that one and of course the no access to knownhost.com. I wish now that I'd have done further checking but we may have to wait and see if it happens again.
When I was repairing phone systems, we had a saying "the wind did it." that we used when we didn't have a clue what went wrong. Maybe the wind did it.
When I was repairing phone systems, we had a saying "the wind did it." that we used when we didn't have a clue what went wrong. Maybe the wind did it.
Sounds like this is related, at least time frame matches.
http://isc.sans.org/diary.html?storyid=5569
http://www.webhostingtalk.com/showthread.php?t=746090
Appears the routing problems was isolated to one (some?) of the Level3 POPs.
http://isc.sans.org/diary.html?storyid=5569
http://www.webhostingtalk.com/showthread.php?t=746090
Appears the routing problems was isolated to one (some?) of the Level3 POPs.