Two things, have you checked your connection speeds when this happens? (Especially wireless). Do you see the speeds being low prior to the host error message?
Second, have you updated the firmware on your router? Even though other systems aren't seeing the same problem it might be helpful to update that firmware just in case. If it's happening both wired and wireless then maybe resetting the tcp/ip stack would help? Not sure.
Let me know about the connection speed and when the last time is that you updated the firmware on the router. Doesn't seem like a driver issue since it's two different network cards showing the same issue.
Hey Todd - Thanks for the reply. I haven't noticed any change in the connection speed prior to this occurring. The router is actually TWC property, so I may need to check with them. Next time it happens I'll try resetting the ip stack. I'll let you know. Thanks :)
TWC usually gives you the admin account and password to you when they set up the router, and allow you to change the password. (I have a TWC modem myself). Updating the firmware is a pretty standard troubleshooting step and their techs can walk you through the process over the phone.
Well, they wouldn't give me access to it, but they claim to have updated the firmware themselves. TWC always gives me the mysterious "I'm sending signals to your device."
Still having the same issue. I should put up a little context here, my background is in network design and management, so I've eliminated the basics. TWC always wants me to do the whole unplug this, unplug that thing. I'm fairly confident that the physical, data link, and network layers are fine. This has to be something at the transport layer or higher. It just has to be. If it isn't I'll eat my hat.
I'll give those a try. I did try static DNS before with the Google dns servers. I'll try these two and if that doesn't work I m ay just go static config all around.
3 or 4 users in our office are seeing this problem, which just started appearing within the last maybe 3 weeks. The only thing that seems fairly consistent is that those with the problems have relatively new Dells (<10 months old). Each of these is running Windows 7, though as our standard OS, all others (except for Mac users) are also running Windows 7.
Most in the office connect to the network wirelessly via the same SonicWall wireless router; others aren't seeing the problem.
Sorry I have no answers, but since it seems that these problems are related, I thought it was worth sharing this information.
Excellent info, thanks for sharing. This is particularly relevant as I placed a wireshark on my network and found that DNS is resolving! I get DNS response packets back to both requests from the browser and from an nslookup. But both those apps hang up as if DNS is not resolving - web browsers say "look for host," and nslookup just hangs. Note that it does not time out - it hangs and the command window won't close, even by ending the process manually. So ... something else is happening here. Next I'm going to see if I can use some Winternals tools to see what's happening under the hood on Win7.
After further investigation, I have some additional information. What is happening is that the DNS request is not being sent immediately. Watching the packet trace live, I observed the following:
1. Web site requested via browser (or nslookup, or any other DNS triggering event)
2. Between 5 and 10 minutes later, the outbound DNS request is sent and the address resolved normally.
I tried flushing the DNS cache, but an empty DNS cache did not cause the issue to reoccur. Possibly something to do with the way Win7 handles TTLs of the DNS records?
This is happening sporadically, sometimes I'll go 24 hours between, other times it will happen twice within an hour. Since the DNS records will expire their TTL at different times, I'm considering it has something to do with that. But that is mere speculation.
Do you have any access to DNS on two or three domains where you could adjust the TTL, with one set extremely short (say, <600), another set to 3600, and the third set to something larger? I haven't thought through exactly how you'd test, but it seems that at a minimum, you need to know how the TTLs are set, and they need to be set in such a way that it can be tested. Not much help, perhaps...
Unfortunately this is happening on my home network, so I don't have access to any of the DNS servers I'm accessing. And ... TWC doesn't allow DNS to non-TWC servers. At least not here, anyway.
Hmm.. Well, I was thinking about this last night and trying to come up with some plausible explanation for this, and why only some systems in our office are seeing it, when I thought to myself that we have a couple of different AntiVirus type products in use here. I'm using McAfee, and I think all of the machines with problems are as well. There are several other users using Norton. I'm seeing some potential in this simply because it's software that kind of injects itself into the discussion when it comes to requests in/out.
I really don't know that it's related, but do you have McAfee AV installed by any chance?
I do. I'm running McAfee Live Safe on all computers here (there are 4 of them). Not experiencing this on any of the others, but they are also all different brands and this is the only one with Win7.
If McAfee is the culprit, it would explain the randomness of the events. Let me elaborate there. At different times I've been sure I've isolated it down to one thing. First it was Google Drive. The DNS issue was happening, I shut down Google Drive, an it immediately resolved itself. Uninstalled Google Drive Sync, problem solved. Until it wasn't. Then the same thing happened with the Trello web page. Close the tab, voila. Then it was SugarSync. Each time I would shut down whatever it was and the issue would immediately resolve. But then later something else would seem to be causing it.
Since McAfee does, as you say, inject itself into the discussion, it's a good suspect to bring in for questioning. Not sure how I'm going to question him at this point, but it's a good next step.
DELL-Todd S
3 Apprentice
•
1.8K Posts
0
July 7th, 2015 13:00
Hi,
Two things, have you checked your connection speeds when this happens? (Especially wireless). Do you see the speeds being low prior to the host error message?
Second, have you updated the firmware on your router? Even though other systems aren't seeing the same problem it might be helpful to update that firmware just in case. If it's happening both wired and wireless then maybe resetting the tcp/ip stack would help? Not sure.
Let me know about the connection speed and when the last time is that you updated the firmware on the router. Doesn't seem like a driver issue since it's two different network cards showing the same issue.
Todd
harperjdav
8 Posts
0
July 7th, 2015 14:00
Hey Todd - Thanks for the reply. I haven't noticed any change in the connection speed prior to this occurring. The router is actually TWC property, so I may need to check with them. Next time it happens I'll try resetting the ip stack. I'll let you know. Thanks :)
DELL-Todd S
3 Apprentice
•
1.8K Posts
0
July 8th, 2015 08:00
Hi,
TWC usually gives you the admin account and password to you when they set up the router, and allow you to change the password. (I have a TWC modem myself). Updating the firmware is a pretty standard troubleshooting step and their techs can walk you through the process over the phone.
Todd
harperjdav
8 Posts
0
July 11th, 2015 18:00
Well, they wouldn't give me access to it, but they claim to have updated the firmware themselves. TWC always gives me the mysterious "I'm sending signals to your device."
Still having the same issue. I should put up a little context here, my background is in network design and management, so I've eliminated the basics. TWC always wants me to do the whole unplug this, unplug that thing. I'm fairly confident that the physical, data link, and network layers are fine. This has to be something at the transport layer or higher. It just has to be. If it isn't I'll eat my hat.
ejn63
9 Legend
•
87.5K Posts
0
July 11th, 2015 18:00
Try changing your local machine DNS settings -- point the DNS to OpenDNS (208.67.222.222 and 220.220).
harperjdav
8 Posts
0
July 11th, 2015 21:00
I'll give those a try. I did try static DNS before with the Google dns servers. I'll try these two and if that doesn't work I m ay just go static config all around.
dvchris
3 Posts
0
July 20th, 2015 13:00
3 or 4 users in our office are seeing this problem, which just started appearing within the last maybe 3 weeks. The only thing that seems fairly consistent is that those with the problems have relatively new Dells (<10 months old). Each of these is running Windows 7, though as our standard OS, all others (except for Mac users) are also running Windows 7.
Most in the office connect to the network wirelessly via the same SonicWall wireless router; others aren't seeing the problem.
Sorry I have no answers, but since it seems that these problems are related, I thought it was worth sharing this information.
Chris
harperjdav
8 Posts
0
July 20th, 2015 23:00
Excellent info, thanks for sharing. This is particularly relevant as I placed a wireshark on my network and found that DNS is resolving! I get DNS response packets back to both requests from the browser and from an nslookup. But both those apps hang up as if DNS is not resolving - web browsers say "look for host," and nslookup just hangs. Note that it does not time out - it hangs and the command window won't close, even by ending the process manually. So ... something else is happening here. Next I'm going to see if I can use some Winternals tools to see what's happening under the hood on Win7.
harperjdav
8 Posts
0
July 22nd, 2015 16:00
After further investigation, I have some additional information. What is happening is that the DNS request is not being sent immediately. Watching the packet trace live, I observed the following:
1. Web site requested via browser (or nslookup, or any other DNS triggering event)
2. Between 5 and 10 minutes later, the outbound DNS request is sent and the address resolved normally.
I tried flushing the DNS cache, but an empty DNS cache did not cause the issue to reoccur. Possibly something to do with the way Win7 handles TTLs of the DNS records?
This is happening sporadically, sometimes I'll go 24 hours between, other times it will happen twice within an hour. Since the DNS records will expire their TTL at different times, I'm considering it has something to do with that. But that is mere speculation.
Any ideas how I could test that theory?
dvchris
3 Posts
0
July 23rd, 2015 08:00
Do you have any access to DNS on two or three domains where you could adjust the TTL, with one set extremely short (say, <600), another set to 3600, and the third set to something larger? I haven't thought through exactly how you'd test, but it seems that at a minimum, you need to know how the TTLs are set, and they need to be set in such a way that it can be tested. Not much help, perhaps...
harperjdav
8 Posts
0
July 23rd, 2015 19:00
Unfortunately this is happening on my home network, so I don't have access to any of the DNS servers I'm accessing. And ... TWC doesn't allow DNS to non-TWC servers. At least not here, anyway.
dvchris
3 Posts
0
July 24th, 2015 11:00
Hmm.. Well, I was thinking about this last night and trying to come up with some plausible explanation for this, and why only some systems in our office are seeing it, when I thought to myself that we have a couple of different AntiVirus type products in use here. I'm using McAfee, and I think all of the machines with problems are as well. There are several other users using Norton. I'm seeing some potential in this simply because it's software that kind of injects itself into the discussion when it comes to requests in/out.
I really don't know that it's related, but do you have McAfee AV installed by any chance?
harperjdav
8 Posts
0
July 24th, 2015 15:00
I do. I'm running McAfee Live Safe on all computers here (there are 4 of them). Not experiencing this on any of the others, but they are also all different brands and this is the only one with Win7.
If McAfee is the culprit, it would explain the randomness of the events. Let me elaborate there. At different times I've been sure I've isolated it down to one thing. First it was Google Drive. The DNS issue was happening, I shut down Google Drive, an it immediately resolved itself. Uninstalled Google Drive Sync, problem solved. Until it wasn't. Then the same thing happened with the Trello web page. Close the tab, voila. Then it was SugarSync. Each time I would shut down whatever it was and the issue would immediately resolve. But then later something else would seem to be causing it.
Since McAfee does, as you say, inject itself into the discussion, it's a good suspect to bring in for questioning. Not sure how I'm going to question him at this point, but it's a good next step.