1 |
>> > +1 on Alan's hunch. I have not used Squid to comment on the specifics |
2 |
>> > and also Grant stated that another proxy gave him similar symptoms. |
3 |
>> > From my limited knowledge a proxy could be stalling because of cache |
4 |
>> > configuration problems, like running out fs space, or inodes and also |
5 |
>> > running out of memory if it has to process simultaneous requests from |
6 |
>> > too many clients at a time. If the problem also manifests when the |
7 |
>> > clients are within the same subnet, then this is unlikely to be a |
8 |
>> > network issue. |
9 |
>> |
10 |
>> Which hunch was that? I snipped a lot above but I couldn't find it in |
11 |
>> there. |
12 |
> |
13 |
> It was Alan's statement that this problem is not related to your AT&T router. |
14 |
> |
15 |
> I have to come back to this. I tried the www.google.com/nexus/ you mentioned |
16 |
> and noticed that the page eats up 1.3MB to load fully, before it starts |
17 |
> downloading a flash video. So seems to be a relatively large amount of data |
18 |
> that brings up this problem and this could point to tcp window scaling. |
19 |
|
20 |
It also happens on very lightweight sites, but never on |
21 |
squid-cache.org for some reason. |
22 |
|
23 |
>> echo 0 > /proc/sys/net/ipv4/tcp_window_scaling |
24 |
> |
25 |
> This is typically enabled, but if you notice that a connection stalls and then |
26 |
> later on it works fine again, it could be related to a firewall/router not |
27 |
> responding as it should to tcp_window_scaling. In this case disabling this |
28 |
> would fix the problem when traversing problematic nodes. |
29 |
> |
30 |
> If you saw no difference, this suggests that window scaling is not an issue. |
31 |
|
32 |
I just tested again and 'echo 0 > |
33 |
/proc/sys/net/ipv4/tcp_window_scaling' on both the client and server |
34 |
did not fix the stalls. |
35 |
|
36 |
> I would start with the simplest tests first, which involve isolating suspect |
37 |
> system components one at a time. Trying to use the same laptop-desktop |
38 |
> machines within the LAN, takes the router out the equation - full 1500 byte |
39 |
> MTU will be used by both laptop and desktop. |
40 |
|
41 |
OK I will try this as soon as I'm back in that location. |
42 |
|
43 |
Thanks a lot, |
44 |
Grant |