<html>
<body>
<br>
ooooo.. To fit what Steve said in there. It is true that they
likely are<br>
classifying ICMP Echo requests (ping) down as well in addition to
the<br>
load balancing.<br><br>
You can easily check this yourself.<br><br>
Notice that when your pings are running at a "higher" rate, if
you<br>
run the game client, and enter a map, then take a look at the<br>
response time, you will likely find it up to 20ms lower. This is<br>
game dependant however in "the way" they measure that
response.<br><br>
ICMP is a link layer protocol, like UDP or TCP. Hence some more
advanced<br>
NIC cards today will respond to a ping even if the CPU on the box
is<br>
DOA. This is because at the lowest layer of the driver on the
NIC<br>
card, it only reviews as follows: "Am I in promiscuous mode?
If so,<br>
accept every packet which is not the norm. Normally your NIC
card<br>
is not in promiscuous mode. In this case the card looks at the<br>
MAC address and will push it up in to the card buffer if one of<br>
three things are true. The Destination MAC address matches<br>
the MAC of the card, the Destination MAC is Broadcast (All Bits<br>
set to 1), or a Multicast MAC. This is only at the Ethernet
layer.<br><br>
Once in the NIC cards buffer, the DMA based driver that interfaces<br>
your OS to the NIC card will pull packets from the card buffer and<br>
move them in to OS Driver space. At this low level point, it only<br>
knows if it is a ARP, IP, or ICMP (ping) packet. Some of the <br>
motherboards out there with integrated NICs and newer NICs<br>
never even pass the ICMP echo up to the OS! The lower level<br>
driver will swap the 8 bytes of IP address SA, DA, and copy<br>
the MAC address and respond to the ping. (Note that the<br>
entire OS could be crashed, and it will respond to ping!)<br>
The NIC card already learned it's IP address, as the very<br>
first thing a NIC card does when it initializes is send a ARP<br>
(Address resolution protocol) packet for it's own MAC address.<br>
(If someone else answers, that is bad, as this is how<br>
it knows it is unique and there is no duplicate MAC address<br>
on that network.). When your OS inits, the NIC gets it's IP<br>
address, either static, or most often by DHCP. THe NIC now<br>
knows it's MAC and IP address. If your OS crashed completely,<br>
that NIC card will still respond to pings.<br><br>
I went off on a bit of a tangent, because I wanted to let you,<br>
and others know how "unreliable" ping is at telling you
"That<br>
host is OK, I can ping it".<br><br>
Back to the original point, ICMP is a protocol on top of IP, at the<br>
same level as TCP and UDP. Hence it is easily identified by<br>
routers and switches, especially since PING is often used by<br>
"script kiddies" for DOS attacks. (With the newer NIC
cards<br>
on your PC, if you run a local software firewall this is a good<br>
thing, as your PC won't get bogged down by a ping DOS attack.<br>
(although it will still chew up your bandwidth, it is unlikely to<br>
have any noticeable impact on you. (That is the good part about<br>
the smarter NIC cards).<br><br>
Back to Steve's Point, many of the games will use a UDP based<br>
echo/response. When you check in your game, all of a sudden the<br>
response rates will be 20ms lower. So Steve's explanation is<br>
also a matter of fact. So in reality, it is good that theplanet
does<br>
prioritize the ICMP's lower. It allows for your clients
important<br>
game packets to have priority over someone's ping packet.<br>
The company is actually protecting your server and hence the<br>
clients gaming experience by giving this protocol a lower
priority.<br><br>
Ask him what he sees when he is playing in the game. That is<br>
more reliable. Many of todays games will measure the average<br>
response time based on the TCP ack's, rather than a ICMP<br>
echo ping once the client is connected to the gaming server.<br><br>
Sorry for the length, Jay will tell ya I disappear for a while, then
come<br>
back and they can't shut me up. ;-)) But the way some of this<br>
all works is pretty cool, and no one ever talks about what happens<br>
behind the scenes when you first power up your PC from the<br>
NIC cards perspective.<br><br>
So at least now if anyone ever asks you "What is the first
thing<br>
a NIC card does when PC is powered up, that you can actually<br>
see?", The answer is "It sends a Address Resolution
Protocol (ARP) packet<br>
to itself", to make sure it is the only one using that MAC
address;<br>
i.e. some other NIC card don't respond to the packet".<br><br>
They installed my ES4. Geez, I forgot about all the changes I made
over time<br>
on the darn thing. ;-(<br><br>
Dr. D<br><br>
<br>
At 11:39 PM 3/24/2005, you wrote:<br>
<blockquote type=cite class=cite cite><font face="arial" size=2>All I
have to say is WOW. That is one awesome explination bro.
</font><br>
<br>
<font face="arial" size=2>I just don't like the fact of my trace routes
from my customers being "blown off" because of packet priority
or whatever.</font><br>
<br>
<font face="arial" size=2>I'm probably gonna lose one of my customers
over this and it really hax me off. </font><br>
<br>
<font face="arial" size=2>Anyways thanx again for the explination
man. This explination makes me wanna open a ticket and paste it,
but it would probably blow their mind as much as it did mine
haha.</font><br>
<br>
<font face="arial" size=2>You will like RHEL4 - it's nice and
smooth.</font><br>
<br>
<br>
<font face="arial" size=2>Have a nice day Dr. D.</font><br>
<br>
<font face="arial" size=2>--</font><br>
<font face="arial" size=2>NateDog</font><br>
<dl>
<dd>----- Original Message ----- <br>
<dd>From:</b> <a href="mailto:defilm@acm.org">Mark J. DeFilippis</a>
<br>
<dd>To:</b> <a href="mailto:cod@icculus.org">cod@icculus.org</a> <br>
<dd>Sent:</b> Thursday, March 24, 2005 8:38 PM<br>
<dd>Subject:</b> Re: [cod] RHE4 new 2.6 Kernel<br><br>
<dd>Thanks Nate, and Jay as always! I have put the order in for the ES4 upgrade.<br>
<dd>I prey every time I put in a change ticket with support. Lately, it has been<br>
<dd>very poor. I hope they can get it right 100%. There is always something<br>
<dd>missing when they make a wholesale change...<br><br>
<dd>Nate, I noticed that for the period that they were migrating in the new data center<br>
<dd>they had major BGP convergence issues, and it would clear my servers.<br><br>
<dd>At first they blamed it on my location. Yea sure. Everyone around the USA, west coast,<br>
<dd>east, central, and it is the whole Internet not you.<br><br>
<dd>Glad I have access to certain info at a carrier that happens to feed the majority<br>
<dd>of servermatrix and theplanet. The peer showed loss of BGP adjacency. I understand<br>
<dd>they had to make changes and re groom fiber, but it was clear from the flapping that they<br>
<dd>don't have BGP dampening configured on their border routers.<br><br>
<dd>I have experience with CRisco 65xx, but on the carrier side for larger customers I design<br>
<dd>for I use Juniper M-20's and M-40's, as well as Laurel Networks 120's. These are pretty big boys.<br>
<dd>Smallest line card on a Laurel is the 13 port DS3 card. But these are big boys, with 8 port<br>
<dd>gig cards wire speed, etc. In their design, they are likely using the Crisco 6509's are layer<br>
<dd>two aggregation devices. While the design is a simple classic design, it relies on default<br>
<dd>load balancing provided by these switching/routers.<br><br>
<dd>I just have to wonder if they know what they are doing. I have noticed that<br>
<dd>my pings run anywhere from 60ms from NY direct Level3 to 92ms. I have the ability to test<br>
<dd>from many area's from around the US from work. From each location, the variation in delay<br>
<dd>is past the local loop in to servermatrix/theplanet. This means they are highly likely using<br>
<dd>load balancing on the Multi-link trunks that are inter-switch and/or switch/router uplinks.<br><br>
<dd>The above switches I have recently been working with in building Int-serv RSVP cores for service<br>
<dd>provider rfc2547bis MPBGP MPLS networks, where LSP's maintain MPLS based traffic engineering<br>
<dd>to provider QOS to IP based QOS CE devices. We found that that MLT load balancing is not<br>
<dd>a strong point in these switches. Adding in the additional distribution layers of inexpensive Crisco 650x<br>
<dd>switches for aggregation, and you have the poor network design they have.<br><br>
<dd>One of my guys designed something like this, with internal MLT load balancing in the core<br>
<dd>through the distribution layer, he would be unemployed. Maybe they have all CCIE's. there? ;-))<br>
<br>
<dd>You are not imagining it. I have a server over at EV1 as well, and my ping times to that<br>
<dd>server is rock solid at 52-66ms. This is reasonable considering much of the long haul is<br>
<dd>DWDM photonic, (plus local loops, and ISP), but note you are going ISP, to loop to loop<br>
<dd>to IXC to loop to LEC to loop to Datacenter. The bulk of the loops are phototonic SONET<br>
<dd>with little latency. If you are generous, add 10ms latency. NY to London on TAT14 I peaked<br>
<dd>at today is running steady at 74ms RTD (Round Trip Delay). (That is our Add/Drop Multiplexor directly<br>
<dd>off the light-wave mux driving the TAT cable. (It is not speed of light as this is<br>
<dd>not speed of light in a vacuum, it is speed of light through glass, which we estimate the<br>
<dd>refraction index (to be a liberal of 1.32), hence the higher delay... In a perfect vacuum we should<br>
<dd>get a RTD of about 53ms. So 74ms is pretty good).<br><br>
<dd>Their core is built differently, hence why their servers are more expensive. ;-) I have not made up my mind about Theplanet, but if my users keep getting dumped, saving $100/mo on a server that is nearly worthless, is throwing $225/mo a way, not saving $100/mo. I consider their networks vs EV1 as hamburger is to steak.<br><br>
<dd>Don't get me wrong. They are nice, growing, but their CEO had a "less expensive" model<br>
<dd>in place for their core than EV1 does, and it shows in ping times. I think I would<br>
<dd>rather a sustained 70ms ping time than a ping time that has 20+ms of jitter in it and jumps<br>
<dd>from 52ms to 70ms like a yo yo. I won't say I have never seen a network do this. I have been in the business for 18 years. I have. But we fixed it, as we consider that "badly broken".<br><br>
<br>
<dd>Dr D<br><br>
<br>
<dd>At 12:16 AM 3/21/2005, you wrote:<br>
<blockquote type=cite class=cite cite>
<dd><font face="arial" size=2>Yeah I've got RHEL 4 also. BTW, I was the one that posted about the teamspeak / mysql issue. You need the compat package for mysql - has the old libraries and such so that you can use them instead of the new ones that come with the new version of mysql on RHEL4:</font><br>
<dd><font face="arial" size=2>Snipet from the teamspeak server.ini:</font><br>
<dd><font face="arial" size=2>VendorLib=/usr/lib/libmysqlclient.so.10.0.0</font><br>
<dd> <br>
<dd><font face="arial" size=2>The package you need:</font><br>
<dd><font face="arial" size=2>MySQL-shared-compat-4.0.23-0.i386.rpm<br>
<dd>It may be a different set of numbers now but that's the name: MySQL-shared-compat</font><br>
<dd> <br>
<dd><font face="arial" size=2>After that teamspeak will work with mysql.</font><br>
<dd> <br>
<dd><font face="arial" size=2>I too ran into problems when I turned on journaling. The main reason is because on a heavy loaded system because of the journal update time being 5 seconds it can cause' some lag - there is a setting you can pass that lowers this - used to work great with the 2.4 kernel but really the 2.6 doesn't need the journal setting from what I've tested - hurts it more than helps as Jay mentioned.</font><br>
<dd> <br>
<dd><font face="arial" size=2>But you can do one performance trick that seems to have helped some. You can set this in your fstab:</font><br>
<dd><font face="arial" size=2>Example:</font><br>
<dd><font face="arial" size=2>ext3 defaults,noatime 1 2<br>
<dd>Add noatime after defaults on the hard drive your running your game servers off of. Basic explination:</font><br>
<dd><font face="arial" size=2>"The noatime setting eliminates the need by the system to make writes to the file system for files which are simply being read....."</font><br>
<font face="arial"><br>
</font>
<dd><font face="arial" size=2>Use at your own risk :) I just thought I'd mention it.</font><br>
<dd> <br>
<dd><font face="arial" size=2>I recently got upgraded to a dual 73gig SCSI server. ThePlanet stuck me in the new datacenter at infomart in Dallas. Has anyone been put in that datacenter yet? I've been receiving complaints of ping spikes from one of my customers that is from California. I gathered trace routes from everyone and it seems to be the link between datacenter 3 and 5 (the new one - infomart).</font><br>
<dd> <br>
<dd><font face="arial" size=2>Anyone else ran into anything yet? From what I can tell it seems to be only from the western side of the US because of the trace routes I've received - the ones that had the problem were in like Arizona and California etc:</font><br>
<dd> <br>
<dd><font face="arial" size=2>13 246 ms 200 ms 199 ms dist-vlan32.dsr3-2.dllstx3.theplanet.com [70.85.127.62]<br>
<dd>14 75 ms 57 ms 57 ms po32.dsr1-2.dllstx5.theplanet.com [70.85.127.110]<br>
<dd>15 58 ms 61 ms 58 ms po2.tp-car3.dllstx5.theplanet.com [70.84.160.165]<br>
<dd>16 57 ms 56 ms 56 ms 39.70-84-187.reverse.theplanet.com [70.84.187.39]</font><br>
<dd> <br>
<dd><font face="arial" size=2>And yes I've opened a support ticket but I'm not getting anywhere with that :( I realize that trace route packets are low priority blah blah but when it's consistent between a whole bunch of different people and the ping is that high - I don't buy into that.</font><br>
<dd> <br>
<dd><font face="arial" size=2>Anyways.......</font><br>
<dd> <br>
<dd><font face="arial" size=2>--</font><br>
<dd><font face="arial" size=2>NateDog</font><br>
<dd><font face="arial" size=2> </font>
<dd>----- Original Message -----
<dd>From: <a href="mailto:jayco1@charter.net">Jay Vasallo</a>
<dd>To: <a href="mailto:cod@icculus.org">cod@icculus.org</a>
<dd>Sent: Sunday, March 20, 2005 9:41 PM
<dd>Subject: Re: [cod] RHE4 new 2.6 Kernel<br><br>
<dd><font face="arial" size=2>Hey Mark,</font><br>
<dd>
<dd><font face="arial" size=2>I rented a few servers from the planet last month and have the same setup you do to the t. Works great. I also noticed that it deals with swap mem a little different than the rhe3. But other than that, runs fine. Did some research on the new file journaling but noticed a decrease in productivity and increase in ping when i set the journaling to on so that was a waste of time. But other than that, if you use it exactly the way the planet gives it to you, the server rocks.</font>
<dd>----- Original Message -----
<dd>From: <a href="mailto:defilm@acm.org">Mark J. DeFilippis</a>
<dd>To: <a href="mailto:cod@icculus.org">cod@icculus.org</a>
<dd>Sent: Sunday, March 20, 2005 9:23 PM
<dd>Subject: [cod] RHE4 new 2.6 Kernel<br>
<br>
<br>
<dd>Anyone had experience with running the COD binaries on the new RedHat Enterprise Server 4.0 with 2.6 SMP and threading enhancements?<br>
<dd>On theplanet.com, and servermatrix.com, there are a few quotes here and there about "nice performance increase". (Actually I would be happy if it is better than the existing ES3 SMP kernel which will often run a cpu up to 100% while the other sits idle at 0%, after the major lag, it kicks in. (yea! isn't that proactive!)<br>
<dd>I am hoping 2.6 enhancements to RHE4 does the trick.<br>
<dd>Anyone?<br>
<dd>I did see some issues with Teamspeak and issues with mysql. At the time of posting, the admins recommended solution was to rev back up2date for the mysql package to 4.0, and Teamspeak is a happy camper again.<br>
<dd>Any input from someone doing this already would be appreciated.<br>
<dd>Thanks<br>
<dd>Md<br><br>
<dd><tt>-------------------------------------------------------------------------------
<dd>S1,Mark J. DeFilippis, Ph. D EE defilm@acm.org
<dd> defilm@ieee.org<br>
</blockquote>
</dl>S1-------------------------------------------------------------------------------<br>
Mark J. DeFilippis, Ph. D EE defilm@acm.org<br>
defilm@ieee.org<br><br>
</blockquote>S4-------------------------------------------------------------------------------<br>
Mark J. DeFilippis defilm@acm.org<br>
defilm@ieee.org<br><br>
<br>
</body>
</html>