Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations strongm on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

My network connection keeps going down

Status
Not open for further replies.

reicher62

IS-IT--Management
Apr 23, 2008
10
US
I have 2 IPOffice Switches hooked up on my network in 2 buildings. They are connected over our network via bana circuits (copper lines that emulate ethernet). The one switch keeps dropping/locking up. I show no errors it justs drops and you can't see the switch. The only way to fix it is a reboot. It happened 3 times today already and 3 times yesterday. The system has been up almost a month. Outbound telephone still works however interoffice and voicemail do not because the voicemail is in the other site.
 
Is your IP office also the DHCP server?
Maybe it has a duplicate ip adress?

Greetzzz...Bas

y1pzZTEUdok1vrI5cLb3FdPX4PgTPlSONkb5WPjz0x50etSujaMSmhdRCbOx9vASnrRNzzXv0IxNQA

___________________________________________
It works! Now if only I could remember what I did...
___________________________________________
 
You could try using a better method of connecting the systems instead of copper lines. I cant image using copper lines for your ethernet connectivity between sites. Thats like using to dialup to surf the net. I would set up a vpn over the internet between sites and use that if it is an option. Using copper lines is ludicris.
 
First of all my ipoffice is not my dhcp server. I have 12 windows 2003 servers with dhcp running. Second Superjenks I would do some research before you condemn something you know little about. I get 19mb of bandwidth using hatteris network ethernet extenders with 4 lines. We have been using this technlogy for 8 years. It costs approximately $400 per month and is kept a secret by the local phone carriers for that reason. They are direct connections through the central office. There is absolutely no latency involved. I connect directly to my other office instantly.
 
I hope you don't have 12 DHCP servers on the same network!!!!

When say the "Swicth" goes down, do you mean the IPO??? I assume you do. When it goes down can you ping other devices on the other end of the link??

What is your IP Addressing at either end? What IP Routes do you have in both IPO's?

When it goes off, can it be pinged from a device on the same LAN as it??

Jamie Green

ACA:Implement - IP Office
ACS:Implement - IP Office


Football is not a matter of life and death-It is far more important!!!!
 
Sometimes an IP address conflict can cause this type of issue, we had something very similiar and that's what it turned out to be after weeks of headscratching. Also have had some 406's with LAN port problems that we ended up swapping out

ACS - IP Office Implement
 
Thanks for some insight. The network is split with one DHCP Server in each building each with a separate range for the users. When it goes down the network connection is totally lost. You can't ping it or connect to it. IP office system status and manager can not see it. The log on the sysmonitor says ********** contact lost with 216.187.16.68 at 08:23:23 23/4/2008 - reselect = 1 **********. I unplugged the ethernet port and rebooted the switch but the only thing that helps is a system reboot of the ipoffice500. The other system is 216.187.16.70 hosts the voicemail and has had no issues. We have had one other issue we keep getting no free channels were available on our T-1. It has happened 5 times today and at least 20 times yesterday. On friday it took me 5 minutes to get a line. They are all active and we have had no more than 7 calls max on them.
 
Both IPO's are in the same subnet, but on separate LANs. Is that right???

Jamie Green

ACA:Implement - IP Office
ACS:Implement - IP Office


Football is not a matter of life and death-It is far more important!!!!
 
I'm not condemning your copper line solution, especially if it works, but in my experiences that is not a pratical or realistic solution I would propose to a customer. Reguardless, it still could be the source of your problem...
Also, is this link used for anything else other than voice transmission such as data or internet access being used over the link to the main site? Just wondering if you're getting any lag from users streaming music or file sharing? Just a thought. Also, are you utilizing any QOS on your network for the voice traffic utilization?
 
What is your Business PArtner doing about this issue???

Jamie Green

ACA:Implement - IP Office
ACS:Implement - IP Office


Football is not a matter of life and death-It is far more important!!!!
 
They are both on the same lan. Basically it is just one big network. We are going to use MPLS with Paetec but have to configure our network properly. This is not a permanent solution. I think it is a hardware issue. We have not had any issues until this week. There is no qos on anything or anyway to put it in. It's basically setup like a direct cable between the offices. There is no outside interference. Our data network is fine when this happens as is the other ipoffice switch. I use remote desktop to access it and there is not outtage. I would agree with that assumption if both switches were locking up but they are not. I am waiting for the business partner to get me to an Avaya tech but figured someone else might have had similar problems. I checked dns and there are no duplicate names on the network showing.
 
I have to agree with others on this post that it sounds like an IP address conflict. as for the T-1 I would contact the carrier if it is a local T-1 to the switch.
 
On a side note, Avaya say not to use Remote Desktop as it can corrupt your voicemail modules and callflows. Just a heads up

ACS - IP Office Implement
 
Thanks everyone. I only use the remote desktop to check on the switches, I do everything else locally.
 
Man, we just implemented a 4 site SCN with a paetec mpls and it works awsome. The roundtrip response times are amazing. Its a great solution and highly recommend it.

As for the problem at hand, I think you will have to get some sysmon traces going to see whats happening and the cause. Avaya will ask for this anyways but I would start there. What version are you on, and are both systems on matching software levels? If you using 4.x or higher, I would also check the system status to see what it says about the actual IP line you have created. It will show any related alarms and may indicate the true issue.

If you would like to know if this is a software scn problem, you can turn off the scn and run a meshed network, but you will have to add shortcodes in both systems to route interoffice traffic from site to site. I have seen alot of scn issues lately since 4.0 came out and this test would determin if its a true scn problem. We can help you with that later if needed.
 
Ok, Hopefully considered as helpful.
You say you have one big network. What type of devices act as the bridge CO/Host.
We have a site that has £0000's of switches and routers and yet they route to their daughter site via £00 DSL technology bodge job router/bridge I could get from a PC World shop. In my personal opionion its the CAM/MAC tables on the bridges maxing out and killing the bridge. A reboot clears but only for a month and we get the exact same again.

I also have had some fantastic results with the types of circuits. Obviously its diffcult to guarentee any type of quality/service but for sites on a budget its a great way to introduce multi site connection.

However as others have said could be an IP Address issue, may be worth pinging the IP Offices from a PC then checking the ARP to ensure the mac matches that of the IP office and not a foreign device.

ACA - IPOffice implement
ACA - IP Telephony
CCNA - Passed at last
 
In regards to the network bridge we just use Hatteras ethernet extenders hooked into the main network switches, nothing else. It is a transparent 19mb pipe that connects the two offices. I sent you what the sysmon said from the phone system that went down. Hopefully this will help it was on the main machine that stayed up

2008231752mS CMLineTx: v=17
CMSetup
Line: type=IPLine 17 Call: lid=0 id=94418 in=0
Called[5155] Type=Unknown (0) Reason=CMDRdirect Calling[1983] Type=Unknown Plan=Unknown
BC: CMTC=Speech CMTM=Circuit CMTR=64 CMST=Default CMU1=ULaw
BChan: slot=9 chan=23
IE CMIESupplementaryService (3)
Network Facility Extension
sourceEntity: 0
destinationEntity: 0
Interpretation APDU
discardAnyUnrecognisedInvokePdu
CallingName.Invoke.CodePageISO8859-1
invokeId 31519
user 'MichaelVolpe' presentation Allowed
IE CMIECallSignalChannelData (8) unknown
Display [Michael Volpe>5155]
Timed: 23/04/08 08:24
Locale: enu
2008231754mS CD: CALL: 0.94416.0 BState=Idle Cut=0 Music=0.0 Aend="MichaelVolpe(1983)" (9.23) Bend="Line 17" [Line 17] (0.0) CalledNum=5155 () CallingNum=1983 (MichaelVolpe) Internal=0 Time=1345 AState=Dialling
2008231755mS H323Evt: v=0 stacknum=17 State, new=NullState, old=NullState id=-1
2008231757mS H323Evt: v=0 stacknum=17 State, new=Initiated, old=NullState id=94418
2008239756mS H323Evt: v=0 stacknum=17 State, new=NullState, old=Initiated id=94418
2008239758mS CMLineRx: v=17
CMReleaseComp
Line: type=IPLine 17 Call: lid=0 id=94418 in=0
Cause=38, Network out of order
2008239758mS CMTARGET: TargetOnProgress: res: 1
2008239758mS CMTARGET: 0.94416.0 26949 MichaelVolpe.0: CancelTimer CMTCNoAnswerTimeout
2008239761mS CMLOGGING: CALL:2008/04/2308:24,00:00:00,000,1983,O,5155,5155,MichaelVolpe,,,0,,""n/a,0
2008239761mS CD: CALL: 0.94416.0 BState=Disconnecting Cut=0 Music=0.0 Aend="MichaelVolpe(1983)" (9.23) Bend="Line 17" [Line 17] (0.0) CalledNum=5155 () CallingNum=1983 (MichaelVolpe) Internal=0 Time=9352 AState=Dialling
2008239779mS CMMap: a=21.7 b=0.0 DTMF::AllocateRAS allocated CMRTTonegen resource busy 5, total 8
2008239779mS CMMap: a=21.7 b=1.255 T
2008239779mS CMMap: a=21.7 b=9.23 M2
2008239780mS CMMap: a=6.1 b=0.0 Mapper::FreeCodec freed CMRTVocoder resource busy 0, total 8
2008239856mS RES: Wed 23/4/2008 08:24:46 FreeMem=76755312(508) CMMsg=7 (8) Buff=200 943 1000 7096 5 Links=12958
2008242839mS CMExtnRx: v=1983, p1=0
CMReleaseComp
Line: type=DigitalExtn 7 Call: lid=0 id=94416 in=0
2008242841mS CMTARGET: 0.94416.0 -1 BaseEP: ~CMTargetHandler
2008242842mS CMMap: a=21.7 b=9.23 M0
2008242842mS CMMap: a=21.7 b=0.0 T0
2008242842mS CMMap: a=21.7 b=0.0 DTMF::~DTMF freed CMRTTonegen resource busy 4, total 8
2008243078mS CMExtnRx: v=1983, p1=0
CMSetup
Line: type=DigitalExtn 7 Call: lid=0 id=94419 in=0
Called[] Type=Default (100) Reason=CMDRdirect Calling[1983] Type=Internal Plan=Default
BC: CMTC=Speech CMTM=Circuit CMTR=64 CMST=Default CMU1=ULaw
IE CMIECallingPartyName (110) MichaelVolpe
IE CMIECallingPartyKName (226) Michael Volpe
2008243079mS CMTARGET: 0.94419.0 26950 MichaelVolpe.0: LOOKUP CALL ROUTE: type=100 called_party= sub= calling=1983 in=0 complete=0 ses=0
2008243079mS CMTARGET: OUTGOING CALL ROUTE
2008243079mS CMTARGET: 0.94419.0 26950 MichaelVolpe.0: ADD TARGET (N): number= type=100 depth=1 nobar=1 setorig=1 ses=0
2008243080mS CMExtnTx: v=1983, p1=0
CMSetupAck
Line: type=DigitalExtn 7 Call: lid=0 id=94419 in=0
BChan: slot=9 chan=23
Timed: 23/04/08 08:24
2008243081mS CD: CALL: 0.94419.0 BState=Idle Cut=1 Music=3.0 Aend="MichaelVolpe(1983)" (9.23) Bend="" [] (0.0) CalledNum= () CallingNum=1983 (MichaelVolpe) Internal=1 Time=7 AState=DialInitiated
2008243082mS CMMap: a=9.23 b=0.0 D1
2008244375mS CMExtnRx: v=1983, p1=0
CMReleaseComp
Line: type=DigitalExtn 7 Call: lid=0 id=94419 in=0
2008244378mS CMLOGGING: CALL:2008/04/2308:24,00:00:00,000,1983,O,,,MichaelVolpe,,,1,,""n/a,0
2008244378mS CD: CALL: 0.94419.0 BState=Idle Cut=0 Music=0.0 Aend="MichaelVolpe(1983)" (9.23) Bend="" [] (0.0) CalledNum= () CallingNum=1983 (MichaelVolpe) Internal=1 Time=1304 AState=Idle
2008244383mS CMTARGET: 0.94419.0 -1 BaseEP: ~CMTargetHandler
2008244385mS CMMap: a=9.23 b=0.0 D0
2008244856mS RES: Wed 23/4/2008 08:24:51 FreeMem=76764088(510) CMMsg=7 (8) Buff=200 951 1000 7096 5 Links=12969
2008246379mS CMExtnTx: v=1983, p1=0
CMVoiceMailStatus
Line: type=DigitalExtn 7 Call: lid=0 id=-1 in=0
Called[MichaelVolpe Msgs=0 Old=0 Sav=0] Type=Default (100) Reason=CMDRdirect Calling[00000000] Type=Default Plan=Default
Display [MichaelVolpe Msgs=0]
Timed: 23/04/08 08:24
2008254902mS CMExtnRx: v=1983, p1=0
CMSetup
Line: type=DigitalExtn 7 Call: lid=0 id=94421 in=0
Called[] Type=Default (100) Reason=CMDRdirect Calling[1983] Type=Internal Plan=Default
BC: CMTC=Speech CMTM=Circuit CMTR=64 CMST=Default CMU1=ULaw
IE CMIECallingPartyName (110) MichaelVolpe
IE CMIECallingPartyKName (226) Michael Volpe
2008254903mS CMTARGET: 0.94421.0 26951 MichaelVolpe.0: LOOKUP CALL ROUTE: type=100 called_party= sub= calling=1983 in=0 complete=0 ses=0
2008254903mS CMTARGET: OUTGOING CALL ROUTE
2008254904mS CMTARGET: 0.94421.0 26951 MichaelVolpe.0: ADD TARGET (N): number= type=100 depth=1 nobar=1 setorig=1 ses=0
2008254904mS CMExtnTx: v=1983, p1=0
CMSetupAck
Line: type=DigitalExtn 7 Call: lid=0 id=94421 in=0
BChan: slot=9 chan=23
Timed: 23/04/08 08:25
2008254906mS CD: CA



 
>> Cause=38, Network out of order

Not IP Office rebooting.
You have a network issue.

From site A run a rolling ping to site B IP Office or Router address

From site B run a rolling ping to site A IP Office or Router address

What do you get.

ACA - IPOffice implement
ACA - IP Telephony
CCNA - Passed at last
 
The mac address is consistent on my pc and the routers. In regards to the network out of order error, If I can't see the switch or ping it therefore it makes sense that my other office would get a network down error since it can talk to it. The next time it goes down I will try to ping the address. If I get a response then I know it's a duplicate IP Address otherwise it's the lan card on the IPOffice. We are running the most current versions of software as of 03/27. These systems were installed 03/28/08
 
>>If I get a response then I know it's a duplicate IP Address otherwise it's the lan card on the IPOffice.

You can prove this. If you do get a response from a ping.
Then do an ARP /a from command prompt. It will show the mac address of the "Item" you have been involved in pinging.

If its the mac of the IP office fine.
If its the mac of some thing else start hunting.

ACA - IPOffice implement
ACA - IP Telephony
CCNA - Passed at last
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top