Quoting Shachar Shemesh, from the post of Sat, 08 Jan: > Assuming the latency is the same or better after the move. It may > actually mean that your server was bottlenecked before, and you have > managed to shift the bottleneck to the network latency. Good move if > that's true.
it feels that way... > The first significant change happens during the beginning of week 53. > Someone lowered the simultanious amount of Apache listening sockets. I can't remember what I did then with the old server, but it definitely was not anything to do with Apache. I must have shut down a service, but I can't remember now. > The second change happens torward the end of the same week. All of a > sudden, there are very little established connections, and lots and lots > of syn-recvd connections. At the same time, there is a sharp decrease in > the amount of "time wait" connections, but no change in the "closing" well, during Saturday just before noon, I switched from Apache 1.3 on a P3/500 machine on IDE at Actcom to Apache 2 on a P4/2800 machine on SATA at Nezeq. obviously this calls for some shifts, but nothing as major as this. the fact there are less connections open didn't bother me, I took it as an obvious sign that the machine's lower letancy causes sessions to be replied and closed sooner, the fact I had more yellow "syn recieved" than before worried me whether I was syn-attacked or something, that part of the graph was radically different than before, as you can see from the hourly graph, almost no "established" connections are marked http://jenna.scso.com/hotsanic/netstat/connections.html however the "time wait" number seems to stay oddly stable throughout the day, even during the graveyard shift of 2-6am. > One possible explanation is that someone changed the "persistent" option > for Apache. This option means that each TCP connection only handles one > HTTP request and then closes. I am not sure whether this means "on" or > "off". It depends on your site's usage characteristics. I'll have a look. I don't think I changed this behaviour from the default back in the Apach 1.3 days, and I definitely didn't change it in Apache2. YET. > 1. Was there any change in configuration besides the cpu switch? > Apache options etc? Apache version and ISP. > 2. What happened at the beginning of week 53? no idea, I am pretty sure I removed some useless services that were running for historical reasons (I think I took off ntalk that day, and was surprised to find cvs pserver was listening, although unavailable due to my IPTABLES) > 3. Was there some change in network latency that is unrelated to your > specific computer? I guess I have Nezek Beinleumi to blame, yet it seems odd that they would have more latency than, say, when I was hosted at Actcom. > The increase in SYN-RECV, however, is rather inexplicable to me. I can't > think of what would cause such a sharp rise, unless some horribly broken > firewall somewhere (have seen such things in the past, though). I guess. that is the only part I can't explain either. all my other guesses and explanations pretty much matched yours. -- Not Jesus Ira Abramov http://ira.abramov.org/email/ ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
