From: "Sam Lewis" Received: from [192.168.100.201] (HELO mail.2rosenthals.com) by 2rosenthals.com (CommuniGate Pro SMTP 5.1.16) with ESMTP id 2372056 for os2-wireless_users@2rosenthals.com; Wed, 14 Apr 2010 23:56:22 -0400 Received: from secmgr-va.2rosenthals.com ([162.83.95.194] helo=mail2.2rosenthals.com) by secmgr-ny.randr with esmtps (TLSv1:AES256-SHA:256) (Exim 4.43) id 1O2GBg-0001OD-JA for os2-wireless_users@2rosenthals.com; Wed, 14 Apr 2010 23:56:21 -0400 Received: from mail-bw0-f223.google.com ([209.85.218.223]:62533) by mail2.2rosenthals.com with esmtp (Exim 4.69) (envelope-from ) id 1O2GBe-0000Dp-0N for os2-wireless_users@2rosenthals.com; Wed, 14 Apr 2010 23:56:18 -0400 Received: by bwz23 with SMTP id 23so875620bwz.26 for ; Wed, 14 Apr 2010 20:56:16 -0700 (PDT) X-CTCH-RefID: str=0001.0A020208.4BC68E62.00C8,ss=1,fgs=0 MIME-Version: 1.0 Received: by 10.204.56.135 with HTTP; Wed, 14 Apr 2010 20:56:15 -0700 (PDT) X-Originating-IP: [71.42.191.58] In-Reply-To: References: Date: Wed, 14 Apr 2010 22:56:15 -0500 Received: by 10.204.152.13 with SMTP id e13mr9714238bkw.30.1271303775895; Wed, 14 Apr 2010 20:56:15 -0700 (PDT) Message-ID: Subject: Re: [OS2Wireless] Tracerte, Delays and Unreachable Site To: "OS/2 Wireless Users Mailing List" Content-Type: text/plain; charset=ISO-8859-1 On Wed, Apr 14, 2010 at 8:45 PM, Hakan wrote: > Yes but in my e-mail I gave information for the eCS box /and/ a Win box > in the same location - both should use the same DNS server since it is > supplied by the Netgear/Westell. Also, did both of you read through the > other two cases which sheds additional light on the problem? Well we/I don't know that your different computers and even different Ethernet interfaces are using the same DNS servers. You can statically assign DNS servers to each interface, wired and wireless. So it may be possible that each computer/OS you are using as well as each interface has a different DNS server setup. I would at least verify what each interface and OS is using. Or your router could be getting bogged down for some reason that may not be clear which could cause the high latency. Now for your trace route problem I haven't a clue why the one in OS/2 doesn't work. All trace route does is send an ICMP packet to the destination starting with a TTL of 1 and then incrementing by one for each hop until it finally reaches the destination and displays information for each hop. So it is a mystery to me why trace route would work on one OS but not the other. If a piece of gear was configured to not respond to ICMP packets then you should just get the * and it should move on to the next hop after a timeout regardless of what OS it was run under. The evidence you provided seems to lead to DNS server's, or the router CPU being loaded down in one configuration IF it is handling the DNS for your network. I could be completely off base and be missing something. But I don't know what router setting would effect latency between pinging an IP address vs. domain name or discriminate between Wired and Wireless interfaces. Also if your router is handling DNS I would not allow that. I had recently had issues with the CPU being loaded down on a Linksys router at one location and when put in a different location doing the same thing it wasn't loaded down, so it is possible for consumer grade routers to get bogged down when you least expect it. My two cents worth, Sam P.S. I just reread your base case. Are you saying that you got the same results when using IP Address vs. domain name? At any rate I think the trace route and you latency/reliability issues are two separate things. Sam