1 |
Am 20.09.2016 um 21:52 schrieb Grant: |
2 |
>>>>>>>> My web server's response time for http requests skyrockets every |
3 |
>>>>>>>> weekday between about 9am and 5pm. I've gone over my munin |
4 |
>>> graphs |
5 |
>>>>> and |
6 |
>>>>>>>> the only one that really correlates well with the slowdown is |
7 |
>>> "TCP |
8 |
>>>>>>>> Queuing". It looks like I normally have about 400 packets per |
9 |
>>>>> second |
10 |
>>>>>>>> graphed as "direct copy from queue" in munin throughout the day, |
11 |
>>>>> but 2 |
12 |
>>>>>>>> to 3.5 times that many are periodically graphed during work |
13 |
>>> hours. |
14 |
>>>>> I |
15 |
>>>>>>>> don't see the same pattern at all from the graph of all traffic |
16 |
>>> on |
17 |
>>>>> my |
18 |
>>>>>>>> network interface which actually peaks over the weekend. TCP |
19 |
>>>>> Queuing |
20 |
>>>>>>>> doesn't rise above 400 packets per second all weekend. This is |
21 |
>>>>>>>> consistent week after week. |
22 |
>>>>>>>> |
23 |
>>>>>>>> My two employees come into work during the hours in question, and |
24 |
>>>>> they |
25 |
>>>>>>>> certainly make frequent requests of the web server while at work, |
26 |
>>>>> but |
27 |
>>>>>>>> if their volume of requests were the cause of the problem then |
28 |
>>> that |
29 |
>>>>>>>> would be reflected in the graph of web server requests but it is |
30 |
>>>>> not. |
31 |
>>>>>>>> I do run a small MTU on the systems at work due to the config of |
32 |
>>>>> the |
33 |
>>>>>>>> modem/router we have there. |
34 |
>>>>>>>> |
35 |
>>>>>>>> Is this a recognizable problem to anyone? |
36 |
>>>>>>> |
37 |
>>>>>>> I'm in the midst of this. Are there certain attacks I should |
38 |
>>> check |
39 |
>>>>> for? |
40 |
>>>>>> |
41 |
>>>>>> It looks like the TCP Queuing spike itself was due to imapproxy |
42 |
>>> which |
43 |
>>>>>> I've now disabled. I'll post more info as I gather it. |
44 |
>>>>> |
45 |
>>>>> imapproxy was clearly affecting the TCP Queuing graph in munin but I |
46 |
>>>>> still ended up with a massive TCP Queuing spike today and |
47 |
>>>>> corresponding http response time issues long after I disabled |
48 |
>>>>> imapproxy. Graph attached. I'm puzzled. |
49 |
>>>>> |
50 |
>>>>> - Grant |
51 |
>>>> Things to check for: |
52 |
>>>> Torrent or other distributed downloads. |
53 |
>>>> Download program with multiple download threads |
54 |
>>> |
55 |
>>> There sure shouldn't be anything like that running either on the |
56 |
>>> server or in the office. Is there a good way to find out? Maybe |
57 |
>>> something that would clearly indicate it? |
58 |
>>> |
59 |
>>> |
60 |
>>>> Maybe another proxy running? Esp. as you saw this also with |
61 |
>>> imapproxy. |
62 |
>>> |
63 |
>>> |
64 |
>>> nginx acts as a reverse proxy to apache2 but that's a pretty common |
65 |
>>> config. Nothing else that I know of. |
66 |
>>> |
67 |
>>> - Grant |
68 |
>> Any way to find out between which hosts/servers those connections are for? |
69 |
>> That might help in locating the cause. |
70 |
>> |
71 |
>> Eg. which of your desktops/laptops inside your network and where they are trying to connect to. |
72 |
> |
73 |
> The spikes are taking place on my remote server but they seem to |
74 |
> roughly coincide with user activity within my own network. My |
75 |
> technical knowledge of networking internals is weak. Does anyone know |
76 |
> which tool will tell me more about the connections that are causing |
77 |
> the TCP Queuing spikes? |
78 |
> |
79 |
> - Grant |
80 |
> |
81 |
> |
82 |
|
83 |
wireshark or whatever it is called at the moment? |