Running Exch2003. OWA has been working fine for months, however a couple of weeks ago OWA users from the internet complain of "page not found" when using the URL. OWA is working fine internally, so this isn't an Exchange/IIS service issue. Also applying SP2 did not effect anything. The NIC does not "complain" and other traffic between the Firewall (PIX) is fine. already swapped cables, and moved to another port on the switch. Packet captures have trunined up nothing ovcious, however the problem is so sporadic that to snapshot the problem at the time is difficult (as soon as the problem is reported, testing OWA is OK). The server is not stressed and the network util us very low (<1%). Traffic spikes off the T-1 do not correlate to the problem. Any ideas on what to try next??
Pathping the server from the PC with the problem. See if it is routing. "bk" wrote in message news:C1EE3B87-E039-4F4C-959D-CFEDC16863BA@microsoft.com... > Running Exch2003. OWA has been working fine for months, however a couple > of > weeks ago OWA users from the internet complain of "page not found" when > using > the URL. OWA is working fine internally, so this isn't an Exchange/IIS > service issue. > Also applying SP2 did not effect anything. The NIC does not "complain" and > other traffic between the Firewall (PIX) is fine. already swapped cables, > and > moved to another port on the switch. Packet captures have trunined up > nothing > ovcious, however the problem is so sporadic that to snapshot the problem > at > the time is difficult (as soon as the problem is reported, testing OWA is > OK). The server is not stressed and the network util us very low (<1%). > Traffic spikes off the T-1 do not correlate to the problem. Any ideas on > what > to try next??
Thanks Nick... I will try next time the issue pops up. The end-users would not be able to do this; also our ISP blocks icmp's at the CPE/perimeter - however I could at least see if the route gets that far. Odd thing is "user A" from "city A" will say they cant get OWA all weekend, however "user B" from "city B" will not have an issue, but may for a few minutes (either a refresh or trying again in a few minutes seems to work). This is becoming a big problem for our OWA-only users. "Nick Gillott [MVP]" wrote: > Pathping the server from the PC with the problem. See if it is routing. > > "bk" wrote in message > news:C1EE3B87-E039-4F4C-959D-CFEDC16863BA@microsoft.com... > > Running Exch2003. OWA has been working fine for months, however a couple > > of > > weeks ago OWA users from the internet complain of "page not found" when > > using > > the URL. OWA is working fine internally, so this isn't an Exchange/IIS > > service issue. > > Also applying SP2 did not effect anything. The NIC does not "complain" and > > other traffic between the Firewall (PIX) is fine. already swapped cables, > > and > > moved to another port on the switch. Packet captures have trunined up > > nothing > > ovcious, however the problem is so sporadic that to snapshot the problem > > at > > the time is difficult (as soon as the problem is reported, testing OWA is > > OK). The server is not stressed and the network util us very low (<1%). > > Traffic spikes off the T-1 do not correlate to the problem. Any ideas on > > what > > to try next?? > > >
results so far look fairly identical. also we put in a new nic (not same slot) - no diff. Here is a pathping test: error test: Tracing route to mail.mycorp.net [66.60.178.9] over a maximum of 30 hops: 0 LT10606.mycorp.org [10.202.46.3] 1 10.202.46.1 2 69.62.153.1 3 172.21.2.5 4 172.21.0.250 5 245.98-30-64.ftth.swbr.surewest.net [64.30.98.245] 6 249.152-60-66-empty.surewest.net [66.60.152.249] 7 66.60.130.126 8 162.6660131.CPE-T1.surewest.net [66.60.131.162] 9 * * * Computing statistics for 225 seconds... Source to Here This Node/Link Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address 0 LT10606.mycorp.org [10.202.46.3] 0/ 100 = 0% | 1 1ms 0/ 100 = 0% 0/ 100 = 0% 10.202.46.1 0/ 100 = 0% | 2 43ms 0/ 100 = 0% 0/ 100 = 0% 69.62.153.1 0/ 100 = 0% | 3 41ms 0/ 100 = 0% 0/ 100 = 0% 172.21.2.5 0/ 100 = 0% | 4 45ms 0/ 100 = 0% 0/ 100 = 0% 172.21.0.250 0/ 100 = 0% | 5 46ms 0/ 100 = 0% 0/ 100 = 0% 245.98-30-64.ftth.swbr.surewest.net [64.30.98.245] 0/ 100 = 0% | 6 44ms 0/ 100 = 0% 0/ 100 = 0% 249.152-60-66-empty.surewest.net [66.60.152.249] 0/ 100 = 0% | 7 46ms 0/ 100 = 0% 0/ 100 = 0% 66.60.130.126 0/ 100 = 0% | 8 53ms 0/ 100 = 0% 0/ 100 = 0% 162.6660131.CPE-T1.surewest.net [66.60.131.162] 100/ 100 =100% | 9 --- 100/ 100 =100% 0/ 100 = 0% LT10606.mycorp.org [0.0.0.0] Trace complete. ================ OK test: Tracing route to mail.mycorp.net [66.60.178.9] over a maximum of 30 hops: 0 LT10606.mycorp.org [10.202.46.3] 1 10.202.46.1 2 69.62.153.1 3 172.21.2.5 4 172.21.0.250 5 245.98-30-64.ftth.swbr.surewest.net [64.30.98.245] 6 249.152-60-66-empty.surewest.net [66.60.152.249] 7 66.60.130.126 8 162.6660131.CPE-T1.surewest.net [66.60.131.162] 9 * * * Computing statistics for 225 seconds... Source to Here This Node/Link Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address 0 LT10606.mycorp.org [10.202.46.3] 0/ 100 = 0% | 1 1ms 0/ 100 = 0% 0/ 100 = 0% 10.202.46.1 0/ 100 = 0% | 2 42ms 0/ 100 = 0% 0/ 100 = 0% 69.62.153.1 0/ 100 = 0% | 3 41ms 0/ 100 = 0% 0/ 100 = 0% 172.21.2.5 0/ 100 = 0% | 4 46ms 0/ 100 = 0% 0/ 100 = 0% 172.21.0.250 0/ 100 = 0% | 5 46ms 0/ 100 = 0% 0/ 100 = 0% 245.98-30-64.ftth.swbr.surewest.net [64.30.98.245] 0/ 100 = 0% | 6 50ms 0/ 100 = 0% 0/ 100 = 0% 249.152-60-66-empty.surewest.net [66.60.152.249] 0/ 100 = 0% | 7 49ms 0/ 100 = 0% 0/ 100 = 0% 66.60.130.126 0/ 100 = 0% | 8 52ms 0/ 100 = 0% 0/ 100 = 0% 162.6660131.CPE-T1.surewest.net [66.60.131.162] 100/ 100 =100% | 9 --- 100/ 100 =100% 0/ 100 = 0% LT10606.mycorp.org [0.0.0.0] Trace complete. "bk" wrote: > Thanks Nick... I will try next time the issue pops up. The end-users would > not be able to do this; also our ISP blocks icmp's at the CPE/perimeter - > however I could at least see if the route gets that far. Odd thing is "user > A" from "city A" will say they cant get OWA all weekend, however "user B" > from "city B" will not have an issue, but may for a few minutes (either a > refresh or trying again in a few minutes seems to work). This is becoming a > big problem for our OWA-only users. > > "Nick Gillott [MVP]" wrote: > > > Pathping the server from the PC with the problem. See if it is routing. > > > > "bk" wrote in message > > news:C1EE3B87-E039-4F4C-959D-CFEDC16863BA@microsoft.com... > > > Running Exch2003. OWA has been working fine for months, however a couple > > > of > > > weeks ago OWA users from the internet complain of "page not found" when > > > using > > > the URL. OWA is working fine internally, so this isn't an Exchange/IIS > > > service issue. > > > Also applying SP2 did not effect anything. The NIC does not "complain" and > > > other traffic between the Firewall (PIX) is fine. already swapped cables, > > > and > > > moved to another port on the switch. Packet captures have trunined up > > > nothing > > > ovcious, however the problem is so sporadic that to snapshot the problem > > > at > > > the time is difficult (as soon as the problem is reported, testing OWA is > > > OK). The server is not stressed and the network util us very low (<1%). > > > Traffic spikes off the T-1 do not correlate to the problem. Any ideas on > > > what > > > to try next?? > > > > > >
