Tracert - Network problem

[Raekwon]

Question for all you network guru's out there.

We have a copier/scanner on our network in our Lethbridge office. When we try to connect to the web configuration from any other office via the IP it can't connect but If I remote a machine in Lethbridge it works no problems. I did a tracert to see the path and it hits 2 IP's that it bounces back and forth from.

18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5

Any idea's what is happening? Looks like its getting caught in a loop being redirected back and forth posibly between 2 switches.

PS: I didn't paste the hops prior to this loop its just shows leaving our office and through allstream to the Lethbridge IP, these are the hops immediately after it leaves allstream

[Rathji]

This is really a guess because I have no idea about your network config, but assuming both those IP addresses are managed switches or routers, then it is probably a switching loop or a buggered routing table.

Packet is hitting the router/firewall and then some other device which is routing it back to the router either because it's routing table does not contain the path to the IP address you have specified or because some idiot has plugged the switch back into the router (directly or indirectly). The machine you are remotely accessing to contact the printer there is probably not using this same switch/router.

Check the wiring and see if maybe your console on the router or switch has some sort of warnings or logs that tell you what is happening.

[Raekwon]

Yeah, sorry for being vauge. This is mostly handled by our head office IT dept. but they have been on this now for a few months.

One of the guys also said to me that he tried to connect before 7AM(Central) one day and was able to connect. They have access to the configuraton of the equipment and I don't so I'm only looking for idea's to speed up this process.

[Rathji]

Honestly, anyone who has access to the devices themselves and can't solve this in a couple hours isn't really trying very hard. Of course, I know that being in IT I have a lot of things on my plate that I don't even think about on a normal day that people are expecting me to be working on. Personally if someone told me they could print at our central office when they were sitting in a desk at our remote office, I would probably just tell them to email whatever they wanted printed.

Being able to access it at 7am is kind of interesting though. That might be the key to figuring this out.

What is different at 7am? Consider all factors

Maybe the person who has their desk above the switch is kicking the switch every day and sometimes it loses connection. That doesn't explain this particular problem, but it is the kind of thinking that might point you in the right direction

Are there machines off at that time? Maybe 2 devices are sharing a static IP address or the static IP is in the DHCP range,it shouldn't matter but who knows if printer is off when the lease for their IP expires maybe 2 devices end up with the same IP

Is there some sort of network access policy or filters on the router that restricts traffic to certain URL/IP ranges at certain times? Maybe something was set to deny when it should have been allow or maybe an old rule is causing problems when it shouldn't be. Again this probably shouldn't cause the problem you are talking about, but i have seen some pretty messed up rule sets for firewalls

[Azure]

There would probably be no network traffic at 7AM.

Which might indicate a hardware or wiring problem.

[Buff]

Probably have redundant and/or messed up routes setup in the routing tables. My first thought was that there was a loop, but you'd be seeing more problems than being able to access the printer.

This is a hard question to solve without having more knowledge of the network.

On a side note, how do you guys find Allstream? How is their support? We use them for one product and we are far from pleased with their tech support, product knowledge and communication.

[BloodFetish]

Any idea what devices have those IPs addresses? I'm with Rathji in that the tracert points to conflicting routing rules, but the 7AM observation kind of undermines this idea. I'd take with 7AM observation with a grain of salt until you can repeat the observation yourself.

Without having access to the routers, I would try to simplify the network path to the printer for testing purposes. For example, how many switches/hubs are between the printer and the router(s)? Can you plug the printer directly into a built-in switch on the router(s) and eliminate some hops? Failing that can you plug the printer into the same switch/hub that is used by the PC that you can remote control (since you know you have a valid path to that PC...)? If any of this works then it implies a wiring or hardware problem as Azure stated, but if you still can't reach the printer then I don't think any conclusion can be drawn without knowing more about the network within that location. And even if you can now reach the printer you may have caused issues for someone else...

In short, good luck. You're gonna need it.

[BloodFetish]

Originally Posted by Buff

On a side note, how do you guys find Allstream? How is their support? We use them for one product and we are far from pleased with their tech support, product knowledge and communication.

I have limited experience with them. We had a location with phones and long distance through Allstream. When it came time to move our store to another location and transition to Telus, Allstream suddenly didn't speak English anymore (okay, that's a metaphor, but you get the idea). Our employee responsible for transitioning the phone lines aged 5 years and started smoking.

[Raekwon]

I appreciate all the idea's. I have never been to this office and it only has about 30 employee's. I requested some pictures and for only having 2 switches it looks like a wiring mess. No UPS, hell it doesn't even look like there is a surge protector things are plugged right into the wall. I just got the go ahead to head down to that office and clean it up so I will have a chance to look at it closer and I will post what I find.

[mykalberta]

can you ping the network applicnce by name from the ho or branch office

Is the IP of the network appliance static and out of range of your DHCP router?

Wiring mess is common at a place where no IT has ever been, I wouldnt say that would be the first thing I check since you can get to it from the local branch and since someone can connect to it - it would be odd that wiring issue only exists for external communication to the network appliance and not internal communication.

The other thing, and this would be difficult to find, is that lets say your branch is configured like this

Lethbridge gateway: 10.11.3.5
Letbridge DHCP range: 10.11.3.50 - 10.11.3.254
IP of network appliance: 10.11.3.41

Branch X gateway: 10.11.33.5
Branch X network appliance: 10.11.3.41

If someone screwed up and gave the same IP to another device than local communication to each network appliance would be fine (depending on router configuration as you can set to block duplicate IP from coming in) but WAN communication (from head office for example) would be screwed).

I would turn off the device over lunch, have them power cycle the router and switches and do a constant ping to the network appliance (keep it off) and see if you get a reply.

[sclitheroe]

Originally Posted by Raekwon

Question for all you network guru's out there.

We have a copier/scanner on our network in our Lethbridge office. When we try to connect to the web configuration from any other office via the IP it can't connect but If I remote a machine in Lethbridge it works no problems. I did a tracert to see the path and it hits 2 IP's that it bounces back and forth from.

18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5
18 ms 18 ms 19ms 10.111.1.4
19 ms 20 ms 19ms 10.33.1.5

Any idea's what is happening? Looks like its getting caught in a loop being redirected back and forth posibly between 2 switches.

PS: I didn't paste the hops prior to this loop its just shows leaving our office and through allstream to the Lethbridge IP, these are the hops immediately after it leaves allstream

Maybe confirm the default gateway and netmask configuration on the copier - if the netmask or gateway is wrong, that could easily be why its only accessible from one particular subnet.

[mykalberta]

Good luck going out there.

FYI, bring a USB hard drive with install files of your common apps because once they see you, you will be there all day fixing these magical problems you didnt know existed but are there.

Also might be a good idea to bring a label maker, some colored network cables based on what your company standard is as well as some extra cables or a crimper and ends.

[mykalberta]

Any update? - inquiring minds want to now. Plus its a slow Tuesday.

[Raekwon]

No updates. It is weird that I can ping every workstation in that office from my office but can't ping the copier. At the same time I can ping the copier from a workstation in that office. Must be a something in the switch and seeing as I don't have access I have to leave it up to head office.

Considering this has been an issue for about 7 months now I'm guessing only another 3 or 4 before they actually resolve the problem

[sclitheroe]

Originally Posted by Raekwon

No updates. It is weird that I can ping every workstation in that office from my office but can't ping the copier. At the same time I can ping the copier from a workstation in that office. Must be a something in the switch and seeing as I don't have access I have to leave it up to head office.

Considering this has been an issue for about 7 months now I'm guessing only another 3 or 4 before they actually resolve the problem

It's the netmask on the printer, I'm tellin ya...

Incorrect Subnet Mask: If a network uses a subnet mask other than the default mask for its address class, and a client is still configured with the default subnet mask for the address class, communication will fail to some nearby networks but not to distant ones. As an example, if you create four subnets (such as in the subnetting example) but use the incorrect subnet mask of 255.255.255.0 in your TCP/IP configuration, hosts will not be able to determine that some computers are on different subnets than their own. When this happens, packets destined for hosts on different physical networks that are part of the same Class C address will not be sent to a default gateway for delivery. A common symptom of this is when a computer can communicate with hosts that are on its local network and can talk to all remote networks except those that are nearby and have the same class A, B, or C address. To fix this problem, just enter the correct subnet mask in the TCP/IP configuration for that host.

[Raekwon]

Originally Posted by sclitheroe

It's the netmask on the printer, I'm tellin ya...

Incorrect Subnet Mask: If a network uses a subnet mask other than the default mask for its address class, and a client is still configured with the default subnet mask for the address class, communication will fail to some nearby networks but not to distant ones. As an example, if you create four subnets (such as in the subnetting example) but use the incorrect subnet mask of 255.255.255.0 in your TCP/IP configuration, hosts will not be able to determine that some computers are on different subnets than their own. When this happens, packets destined for hosts on different physical networks that are part of the same Class C address will not be sent to a default gateway for delivery. A common symptom of this is when a computer can communicate with hosts that are on its local network and can talk to all remote networks except those that are nearby and have the same class A, B, or C address. To fix this problem, just enter the correct subnet mask in the TCP/IP configuration for that host.

I have little doubt in you. I can post the configuration if you want. I just want to believe that our IT dept. in our head office would be smart enough to figure this out.

PS: appreciate the response

[Azure]

Well post it here, so we know. You don't have to tell him what the problem is.

[Rathji]

Originally Posted by Raekwon

I have little doubt in you. I can post the configuration if you want. I just want to believe that our IT dept. in our head office would be smart enough to figure this out.

PS: appreciate the response

He has me convinced, and it only took him 2 posts!

Still doesn't explain the being able to access it at 7am though, but that was probably a red herring.

The thing about netmasks is that within an internal network you can get by without them being correct for a long time and not notice. This can be because you are attributing errors that come up to other factors, like the guesses in this thread, some made sense if you didn't put much thought into it, like my firewall rules guess or my switching loop theory.

The other thing is, the netmask is the least often culprit when typical connection problems occur. Of the 3 settings I would check on a device, IP address and Gateway would all come before me taking into account the netmask. After all, it is probably a difference of 255.255.255.0 to 255.255.0.0, something you might not notice unless you were looking for it.

I am almost willing to bet that your head office IT staff didn't put enough effort into it to rule out the other factors that they, like others in this thread, considered to be likely causes.

[Raekwon]

[Raekwon]