Is there any problem on our svn server?
Tao Wang
dancefire at gmail.com
Sun May 23 15:07:38 EDT 2010
Hi,
I got mtr result, tested for 1500 packets, it looks like there is a
network problem/congestion at 'c-98-242-65-37.hsd1.ga.comcast.net',
it's about 10% packet loss rates, so it's about 150 packets lost
during the 1500 packets test. it's high.
============================================================
My traceroute [v0.75]
ubuntu1004 (0.0.0.0) Mon May 24 04:27:37 2010
Resolver: Received error response 2. (server failure)er of fields quit
Packets Pings
Host Loss% Snt Last Avg
Best Wrst StDev
1. mygateway1.ar7 0.0% 1543 5.5 1.7
0.4 112.8 5.0
2. syd-nxg-men-bras1-lo-20.tpgi.com.au 1.1% 1543 13.8 15.8
13.0 123.0 5.0
3. syd-nxg-men-csw-2-vlan-1.tpgi.com.au 1.0% 1543 24.7 20.4
13.4 222.7 24.6
4. syd-nxg-men-crt1-po-1.tpgi.com.au 1.0% 1543 15.2 16.0
13.2 122.6 5.6
5. if-3-0.core4.SQN-SanJose.as6453.net 1.2% 1543 165.3 165.6
163.6 318.4 5.4
6. Vlan45.icore1.SQN-SanJose.as6453.net 16.4% 1543 168.4 170.7
163.6 282.0 7.7
7. ix-2-0.icore1.SQN-SanJose.as6453.net 1.2% 1543 164.8 168.5
163.4 270.8 12.5
8. vlan79.csw2.SanJose1.Level3.net 1.0% 1543 165.6 170.9
164.3 295.5 6.5
9. ae-73-73.ebr3.SanJose1.Level3.net 1.2% 1542 176.4 171.6
164.6 256.0 6.0
10. ae-2-2.ebr3.LosAngeles1.Level3.net 0.9% 1542 176.4 179.7
171.9 398.6 9.1
11. ae-44-90.car4.LosAngeles1.Level3.net 0.5% 1533 174.7 184.7
171.6 391.4 34.8
12. BANDCON.car4.LosAngeles1.Level3.net 0.2% 1533 172.9 174.9
172.2 291.8 7.0
13. 199.232.45.157 0.5% 1532 273.0 244.2
240.7 373.0 7.7
14. so-7-0-0-001.br2.qcy1.ma.gnaps.net 0.4% 1532 248.3 250.0
247.3 395.2 6.3
15. ge-4-1-0-000.ar1.qcy1.ma.gnaps.net 0.3% 1531 249.2 265.6
247.2 550.8 45.9
16. link42-14.cent.net 0.3% 1531 248.0 256.9
246.9 465.1 27.8
17. c-98-242-65-37.hsd1.ga.comcast.net 10.8% 1531 280.9 282.6
278.0 372.0 6.3
18. svn.gnucash.org 10.1% 1529 301.5 310.6
299.9 1473. 62.6
============================================================
I also checked at an online mtr site:
http://www.websitepulse.com/help/tools.php [Network -> MTR -
traceroute & ping]
It also show the last 2 nodes often lost packets. The physical
location of the server running test is Seattle, WA. Different from
mine, Australia.
================================================================
Hop Hostname (IP) Country Loss Sent Rcvd Min (ms) Avg (ms) Max (ms)
1 66.249.6.105 US 0 % 5 5 0.333 0.398 0.490
2 po1-br0-tuk.wa.spry.com
(64.79.223.1) US 0 % 5 5 0.410 0.496 0.728
3 g2.8-br1-tuk.wa.spry.com
(64.79.223.26) US 0 % 5 5 0.292 0.418 0.730
4 66.162.128.21 US 0 % 5 5 0.799 0.953 1.063
5 66.192.253.170 US 0 % 5 5 29.092 29.278 29.718
6 199.232.45.157 US 0 % 5 5 101.994 102.159 102.502
7 so-7-0-0-001.br2.qcy1.ma.gnaps.net
(199.232.45.81) US 0 % 5 5 108.975 109.220 109.651
8 ge-4-1-0-000.ar1.qcy1.ma.gnaps.net
(199.232.44.134) US 0 % 5 5 108.731 113.601 131.499
9 link42-14.cent.net
(199.232.42.14) US 0 % 5 5 108.893 115.156 139.697
10 c-98-242-65-37.hsd1.ga.comcast.net
(98.242.65.37) US 0 % 5 5 136.684 138.118 141.109
11 code.gnucash.org
(204.107.200.65) US 20 % 5 4 161.923 162.851 163.888
================================================================
I don't know how to setup mtu in Windows, on my ubuntu, the result is
similiar. However, ubuntu is able to do the clean check out sometimes,
(failed twice today), but slow. I think linux tcp/ip stack maybe wait
longer before give up or it retried more times than Windows.
On Mon, May 24, 2010 at 2:29 AM, Derek Atkins <derek at ihtfp.com> wrote:
> Hi,
>
> Sorry for the top-posting here. Basically you've proven that your IP is
> not being blocked. However you might be having some congestion or other
> issue. Do you have access to mtr? Could you run an MTR to the server and
> see what it shows? It might show you where the congestion is happening.
>
> Another thing I noticed was on the server there are some error/dropped
> packets:
>
> RX packets:143012478 errors:11460 dropped:4623 overruns:0 frame:0
>
> I'm not exactly sure what would cause this.
>
> A final thing to try, just for kicks: Reduce your MTU:
>
> ifconfig eth0 mtu 1400
>
> And then see if the full checkout works. The network here should be able
> to provide 4-6mbps to you, so you should be able to pull at over
> 200KB/sec. I've certainly seen those speeds when I'm off-net. But there
> could be a problem somewhere between you and the server; an mtr output
> might help.
>
> -derek
>
> On Sun, May 23, 2010 12:13 pm, Tao Wang wrote:
>> On Mon, May 24, 2010 at 1:56 AM, Derek Atkins <derek at ihtfp.com> wrote:
>>> Hi,
>>>
>>> On Sun, May 23, 2010 8:33 am, Tao Wang wrote:
>>>> Hi, guys,
>>>>
>>>> I cannot check out the trunk from last night, but I don't think it's
>>>> my network's problem, since I can git clone gtkmm smoothly. So, is
>>>> there any problem on our svn server? Thanks.
>>>
>>> Nope, the server is running just fine. However if you've got a dynamic
>>> IP
>>> Address it's possible that you hit it from an address that's been
>>> blocked.
>>> What's your public-facing IP Address that you used? Can you ping the
>>> svn
>>> server?
>>>
>>
>> Thanks for reply. My ip is '60.241.113.71', I can ping and tracerouter
>> of the server.
>>
>> ==========================================
>> C:\Users\Tao>ping svn.gnucash.org
>>
>> Pinging svn.gnucash.org [204.107.200.65] with 32 bytes of data:
>> Reply from 204.107.200.65: bytes=32 time=307ms TTL=54
>> Reply from 204.107.200.65: bytes=32 time=312ms TTL=54
>> Reply from 204.107.200.65: bytes=32 time=302ms TTL=54
>> Reply from 204.107.200.65: bytes=32 time=304ms TTL=54
>>
>> Ping statistics for 204.107.200.65:
>> Packets: Sent = 4, Received = 4, Lost = 0 (0% loss),
>> Approximate round trip times in milli-seconds:
>> Minimum = 302ms, Maximum = 312ms, Average = 306ms
>>
>> C:\Users\Tao>tracert svn.gnucash.org
>>
>> Tracing route to svn.gnucash.org [204.107.200.65]
>> over a maximum of 30 hops:
>>
>> 1 3 ms 1 ms 1 ms mygateway1.ar7 [10.1.1.1]
>> 2 15 ms 15 ms 15 ms syd-nxg-men-bras1-lo-20.tpgi.com.au
>> [10.20.20.146]
>> 3 18 ms 15 ms 15 ms syd-nxg-men-csw-2-vlan-1.tpgi.com.au
>> [220.245.178.194]
>> 4 16 ms 14 ms 15 ms syd-nxg-men-crt1-po-1.tpgi.com.au
>> [202.7.171.1]
>> 5 165 ms 165 ms 165 ms if-3-0.core4.SQN-SanJose.as6453.net
>> [216.6.30.13]
>> 6 172 ms 165 ms 174 ms Vlan45.icore1.SQN-SanJose.as6453.net
>> [216.6.30.30]
>> 7 165 ms 166 ms 165 ms ix-2-0.icore1.SQN-SanJose.as6453.net
>> [209.58.116.14]
>> 8 174 ms 179 ms 180 ms vlan79.csw2.SanJose1.Level3.net
>> [4.68.18.126]
>> 9 173 ms 166 ms 175 ms ae-73-73.ebr3.SanJose1.Level3.net
>> [4.69.134.229]
>> 10 177 ms 178 ms 179 ms ae-2-2.ebr3.LosAngeles1.Level3.net
>> [4.69.132.10]
>> 11 176 ms 174 ms 173 ms ae-44-90.car4.LosAngeles1.Level3.net
>> [4.69.144.198]
>> 12 175 ms 174 ms 175 ms BANDCON.car4.LosAngeles1.Level3.net
>> [4.71.128.22]
>> 13 244 ms 243 ms 242 ms 199.232.45.157
>> 14 250 ms 249 ms 252 ms so-7-0-0-001.br2.qcy1.ma.gnaps.net
>> [199.232.45.81]
>> 15 250 ms 251 ms 250 ms ge-4-1-0-000.ar1.qcy1.ma.gnaps.net
>> [199.232.44.134]
>> 16 333 ms 318 ms 350 ms link42-14.cent.net [199.232.42.14]
>> 17 389 ms * 295 ms c-98-242-65-37.hsd1.ga.comcast.net
>> [98.242.65.37]
>> 18 303 ms 302 ms 310 ms wiki.gnucash.org [204.107.200.65]
>>
>> Trace complete.
>> ==========================================
>>
>> I feel the server is slow these days, but it was ok for me, since I
>> just run 'svn update'. However, I begin to do a clean check out
>> yesterday and found I failed. The check out speed is about 1-20KB/s,
>> it's mostly around 2-3KB/s, and sometimes it got error and interrupt
>> the svn checkout procedure. So, I can only update my previous checkout
>> since the traffic is smaller and more likely to be sucessed, but I
>> cannot do a clean check out, it always interrupt somewhere.
>>
>> I did check out of goffice and gtkmm several times today, it's roughly
>> 200KB/s or more, so I think it should not be my connection problem.
>>
>>>> Regards
>>>>
>>>> Tao Wang
>>>
>>> -derek
>>>
>>>
>>
>>
>>
>> --
>> Regards
>>
>> Tao Wang
>>
>
>
--
Regards
Tao Wang
More information about the gnucash-devel
mailing list