1 |
On 29/06/15 02:46, Neil Bothwick wrote: |
2 |
> On Sun, 28 Jun 2015 18:27:57 +0100, Mick wrote: |
3 |
> |
4 |
>>> Why did you stop using lighttpd? |
5 |
>> |
6 |
>> I avoided offering much explanation in my previous response because, |
7 |
>> well ... I would feel uncomfortable doing so without a pint in my |
8 |
>> hand. :-)) |
9 |
> |
10 |
> So this is turning into a pub argument about which web server is best? :) |
11 |
> |
12 |
>> All these are good servers for particular use cases. My use case for |
13 |
>> the lighttpd was an embedded system with a 266Mhz SoC and 32MB of RAM. |
14 |
>> I tried thttpd, lighttpd, apache and nginx on it. |
15 |
>> |
16 |
>> - lighttpd was heavier on memory usage, although not as bad as apache. |
17 |
>> |
18 |
>> - nginx was light, fast and full of features. |
19 |
>> |
20 |
>> - thttpd was very basic but got the job done with relatively low burden |
21 |
>> on resources. Slower than ligthttpd. |
22 |
>> |
23 |
>> - apache just about worked, but brought the little thing to its knees. |
24 |
>> |
25 |
>> Don't ask me for benchmarks please, because this was done some years |
26 |
>> ago. I went with nginx because it was faster and kept the CPU% and |
27 |
>> MEM% lowest among competitors. The task in hand was to serve some |
28 |
>> simple web pages with MRTG graphs on them. |
29 |
> |
30 |
> Thanks for the explanation, it appears I owe you a pint if you're ever in |
31 |
> my neck of the woods... |
32 |
> |
33 |
> |
34 |
|
35 |
same here! |
36 |
I decided to start with lighttpd and it seems to do the job. Will look |
37 |
at Nginx next. |
38 |
|
39 |
Thanks, |
40 |
BillK |