Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Really Weird Network issue
View unanswered posts
View posts from last 24 hours

Goto page 1, 2  Next  
Reply to topic    Gentoo Forums Forum Index Networking & Security
View previous topic :: View next topic  
Author Message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Wed Nov 11, 2015 11:43 am    Post subject: Really Weird Network issue Reply with quote

I have some domains hosted in an Aussie Web Hosting company.
My domains are on a specific virtual host using Web Hosting Mannager (WHM).

I boot my home Gentoo Linux machine, using either kernel 4.0.5-gentoo or 4.1.12-gentoo and run up my usual desktop (fluxbox)

I use Firefox generally but Chromium and google-chrome both exhibit the same behaviour.

Sometimes, I can access my domains, then they simply stop working. The SYN packets are sent out and nothing comes back.

I have tried rebooting my ADSL modem.
I have tried power cycling an 8 port switch into which my PC is connected.

BUT, If I boot to windows, or use a windows machine in my home network, they can access my domains.

Also, if I boot a windows Virtual Machine (VirtualBox) on my Gentoo machine, the Windows Guest can access my domains, but my Gentoo Host cannot.

I have no problems accessing other sites, such as this one.
I can do a tracepath all the way to the virtual server.
I even try something stupid like 'telnet thedomain.com 80' just to see if I can connect, which I can't.

I am really confused.

Any suggestions as to where I should even start on this one?

Edit: as a start, I am rebuilding the entire system, using emerge --emptytree, that will punish the computer for not working correctly.
_________________
...Lyall
Back to top
View user's profile Send private message
charles17
Advocate
Advocate


Joined: 02 Mar 2008
Posts: 2583

PostPosted: Wed Nov 11, 2015 12:34 pm    Post subject: Re: Really Weird Network issue Reply with quote

lyallp wrote:
Any suggestions as to where I should even start on this one?

Can you ping your domains from Gentoo?
Back to top
View user's profile Send private message
depontius
Advocate
Advocate


Joined: 05 May 2004
Posts: 3376

PostPosted: Wed Nov 11, 2015 4:16 pm    Post subject: Reply with quote

Have you told your Gentoo system or that it is on one of your hosted domains? Is there any sort of possible conflict between your LAN and your hosted domains?
_________________
.sigs waste space and bandwidth
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Wed Nov 11, 2015 9:10 pm    Post subject: Reply with quote

I can ping my domains.
I can traceroute to the domains.
Normally I can access my domains using firefox/chrome but recently, things seem to become unreliable.
I initially thought it was my hosting company but it's not because I can access the sites from other computers in my home LAN and from my mobile phone.
As stated above, I can even access them from a Virtual Machine hosted on the damn gentoo box.
At the time I am writing this reply, my emerge --emptytree is half way through and things seem to be working!
I wonder for how long....
_________________
...Lyall
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Thu Nov 12, 2015 11:34 am    Post subject: Reply with quote

Damn, stopped working again.
only 50 ebuilds to go for the entire system re-build.
Will see what happens after the rebuild is complete and I have done a reboot.
This is sooo annoying....
_________________
...Lyall
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 6968

PostPosted: Thu Nov 12, 2015 12:19 pm    Post subject: Reply with quote

You symptoms appears like someone using a bad MTU.

With bad MTU set on the interface, internet services may works, or fails, or get slow, some may works (ping...) while other may not (http...).

If it's badly set on Gentoo, any other devices with proper MTU set will works (windows, phone...)
And even the same interface with a different MTU set will works (VM working while its host is not)

Have a look here : https://forums.gentoo.org/viewtopic-t-935256-highlight-mtu.html
And see what user iceaway is saying, isn't what you are also seeing yourself?
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Thu Nov 12, 2015 9:13 pm    Post subject: Reply with quote

Awesome, will look into it and report back.
Thanks !
Edit Well, my MTU was at 1500 but my home ADSL modem, a FritzBox 7390, has a MTU of 1492 on ppp connections. I have tweaked my MTU down to 1492 and will monitor.
_________________
...Lyall
Back to top
View user's profile Send private message
dufeu
l33t
l33t


Joined: 30 Aug 2002
Posts: 896
Location: US-FL-EST

PostPosted: Fri Nov 13, 2015 12:42 am    Post subject: Reply with quote

lyallp wrote:
Awesome, will look into it and report back.
Thanks !
Edit Well, my MTU was at 1500 but my home ADSL modem, a FritzBox 7390, has a MTU of 1492 on ppp connections. I have tweaked my MTU down to 1492 and will monitor.

You always have to use same MTU as the device with the smallest maxium MTU. In your case, your ASDL device has the smallest maximum MTU.

A maximum MTU of 1492 for ADSL modems is a required design feature. It's part of the limitation of fitting an ADSL device into the Ethernet specs. Pings and traceroutes will almost always work because they don't {usually} fill a packet. VMs often work because the default MTU on their virtual network cards is often 1492. etc

You should probably check all your devices on you network and be sure the're all running with an MTU of 1492. Consistency in network connections is one of those things that make life a little less stressful.
_________________
People whom think M$ is mediocre, don't know the half of it.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Fri Nov 13, 2015 2:32 am    Post subject: Reply with quote

Thank you for taking the time to explain.

Now,I suspect, I really should be asking, "Why didn't Gentoo set the MTU to 1492 itself?"

My network setup is

My PC and 3 other PCs -> ASUS 8 port gigabit switch -> FritzBox


I figured I would keep as much local traffic out of the fritz box as possible by using switch, as my family complain about 'ping times' whilst gaming and anything that reduces the cpu load on the fritzbox is a good thing, I figured.
_________________
...Lyall
Back to top
View user's profile Send private message
dufeu
l33t
l33t


Joined: 30 Aug 2002
Posts: 896
Location: US-FL-EST

PostPosted: Fri Nov 13, 2015 4:28 am    Post subject: Reply with quote

lyallp wrote:
Thank you for taking the time to explain.

Now,I suspect, I really should be asking, "Why didn't Gentoo set the MTU to 1492 itself?"

You're welcome

Automatic MTU setting properly falls under things like network card drivers. Use of an ADSL modem removes visibilty of MTU size from your nic drivers (puts it in the ADSL modem instead) so it isn't really visible to any automated routine on your computer. In other words, your nic drivers need to set MTU size 'blind'. THat's why you sometimes have to do it manually.

And a lot of drivers are too stupid to do it right anyway.

It's just one of those things that you only find out by being burned one time. After that, you won't get fooled again. ;)
_________________
People whom think M$ is mediocre, don't know the half of it.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Fri Nov 13, 2015 1:57 pm    Post subject: Reply with quote

Problem is, it hasn't solved my problem.

:(

mtu is now 1492, and my sites have gone off-line again.
I have backed out my mtu setting change

I have done a full system rebuild (emptytree)
I rebooted.
I restarted the network.
I even powered the machine off and on again.

Damn sites still not available. (worst part, my email forms part of this!)
_________________
...Lyall
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 6968

PostPosted: Fri Nov 13, 2015 3:37 pm    Post subject: Reply with quote

If your modem is only a modem, your card should have 1492
If your modem is also a router, your card should have 1500. If the device only have a configuration option to connect to internet or none, it's a modem ; if it have more options (firewall, masquerading, bridging facility, port forwarding...) it's a modem/router and you should use 1500, it's when it will speak with internet that it will fragment them to the 1492 (something it will do alone).
card<--1492-->modem<-----------1492--->internet
card<--1500-->modem/router<---1492--->internet

Get back to 1500 if your modem is also a router then.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Fri Nov 13, 2015 10:12 pm    Post subject: Reply with quote

My adsl modem is a router (FritzBox 7390).
I am back to 1500
I left the computer off overnight and things worked when I powered on this morning.
It will be interesting to see how long before things go pear shaped.
_________________
...Lyall
Back to top
View user's profile Send private message
Logicien
Veteran
Veteran


Joined: 16 Sep 2005
Posts: 1361
Location: Montréal

PostPosted: Fri Nov 13, 2015 10:38 pm    Post subject: Reply with quote

I am following this thread. Did you try to boot a live Linux to see if you can access your site? Can it be a firewall problem and/or a kernel configuration and/or kernel parameter(s) issue(s)? Maybe not if virtual the Windows can access the site using Gentoo.
_________________
Paul
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 42596
Location: 56N 3W

PostPosted: Fri Nov 13, 2015 10:54 pm    Post subject: Reply with quote

lyallp,

In theory, you never need to set MTU manually. The route MTU (it can be different for different routes) should be auto detected and used.
However, there is a lot of badly configured hardware out there on the big bad internet and MTU does not always work.

1492 is the biggest MTU to fit over PPPoE. Its quite possible for the route MTU to be lower.
The route MTU can be discovered with ping by setting the do not fragment flag and testing with different size packets.
Any MTU less than 2048 can be discovered with no more than 11 pings, provided the target responds to ping requests.

Reading the ping man page and googling binary search, is left as an exercise for the reader :)
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Fri Nov 13, 2015 11:11 pm    Post subject: Reply with quote

Interesting.

Code:
root@lyalls-pc ~
# ping -M do -s 997 vmres08.auserver.com.au
PING vmres08.auserver.com.au (101.0.117.31) 997(1025) bytes of data.
^C
--- vmres08.auserver.com.au ping statistics ---
4 packets transmitted, 0 received, 100% packet loss, time 2999ms

root@lyalls-pc ~
# ping -M do -s 996 vmres08.auserver.com.au
PING vmres08.auserver.com.au (101.0.117.31) 996(1024) bytes of data.
1004 bytes from vmres08.auserver.com.au (101.0.117.31): icmp_seq=1 ttl=56 time=38.0 ms
1004 bytes from vmres08.auserver.com.au (101.0.117.31): icmp_seq=2 ttl=56 time=35.4 ms
1004 bytes from vmres08.auserver.com.au (101.0.117.31): icmp_seq=3 ttl=56 time=38.0 ms
1004 bytes from vmres08.auserver.com.au (101.0.117.31): icmp_seq=4 ttl=56 time=34.5 ms
^C
--- vmres08.auserver.com.au ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3003ms
rtt min/avg/max/mdev = 34.575/36.529/38.061/1.562 ms
root@lyalls-pc ~
#


996 is way less than I would have expected!
Does this mean I should do a tracepath and for each hop, do the same thing to try determine which point the problem lies, or is there not a problem?

EDIT: sites dropped off again, no longer accessible, although pingable.
EDIT: and back on-line again with me doing nothing.... sigh...
_________________
...Lyall


Last edited by lyallp on Sat Nov 14, 2015 12:01 am; edited 1 time in total
Back to top
View user's profile Send private message
dufeu
l33t
l33t


Joined: 30 Aug 2002
Posts: 896
Location: US-FL-EST

PostPosted: Fri Nov 13, 2015 11:54 pm    Post subject: Reply with quote

lyallp wrote:
996 is way less than I would have expected!
Does this mean I should do a tracepath and for each hop, do the same thing to try determine which point the problem lies, or is there not a problem?

There isn't a problem.

As Neddy said, in theory, you should never have to set MTU. This is true. What is supposed to happen when you have a PC with 1500 MTU passing through an ADSL modem (using PPPoE) which is at 1492 MTU is that the modem will fragment a full packet (1500 bytes of data) into two packats, one with 1492 bytes of data and a second with 8 bytes of data.

Completely transparent to you and your PC, your ADSL modem is sending and receiving 2 packets for every PC packet which has 1493~1500 bytes of data.

While theory is a wonderful thing, there are a number of real life issues with this situation.

The first three are performance, performance and performance. Perofmance in this situation is very complex. If you're really interested in discussing this in a lot more detail, you may want to go to www.dslreports.com and check out the forums there. They also have a number of ADSL related tools and tuning advice/FAQs.

In addition to the myriad performance issues, when you send out a lot of small, fragmented packets, some hops can potentially be running network monitoring software looking for bad players which make use of small, fragmented packets. While, in theory, this shouldn't effect you because you're not a 'bad player', any misconfiguration on their part will mess with your traffic. Since every good hosting service should be running network intrusion monitoring, you could potentially be creating an on again/off again problem for yourself at your hosting service with mismatched MTU's at your local network. I've seen this happen. {Thankfully, not a connection I was responsible for}.

Again, as Neddy said, there is a lot of misconfiguration out there in the big bad Internet. The best you can do is be sure that you don't create any problems at your end. Mismatched MTU between your ADSL modem and your local network PCs is just one of those problems you want to avoid.
_________________
People whom think M$ is mediocre, don't know the half of it.
Back to top
View user's profile Send private message
Fitzcarraldo
Veteran
Veteran


Joined: 30 Aug 2008
Posts: 1636
Location: United Kingdom

PostPosted: Sat Nov 14, 2015 2:18 am    Post subject: Reply with quote

lyallp, you could try enabling and disabling PLPMTUD (Packetization Layer Path MTU Discovery) to see what happens:

Code:
# echo 2 > /proc/sys/net/ipv4/tcp_mtu_probing


If "echo 2" does not solve the problem, try "echo 1" instead. The possible values are:

0 Do not perform PLPMTUD (Packetization Layer Path MTU Discovery).
1 Perform PLPMTUD only after detecting a 'blackhole'.
2 Always perform PLPMTUD.
_________________
Clevo W230SS: amd64 OpenRC elogind nvidia-drivers & xf86-video-intel.
Compal NBLB2: ~amd64 OpenRC elogind xf86-video-ati. Dual boot Win 7 Pro 64-bit.
KDE on both.

Fitzcarraldo's blog
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Sat Nov 14, 2015 5:43 am    Post subject: Reply with quote

An interesting thing to note.

Whilst the sites where off-line (from my machines perspective), I found I could connect using FileZilla BUT it could not retrieve a directory listing of the remote machine.
I will try this again, next time things go pear shaped.
Here is FileZillas output, it seems to connect but not....
EDIT: Updated to include more detailed FTP info, it's really weird, connects, but doesn't!

Code:
18:38:18   Status:   Disconnected from server
18:38:18   Status:   Resolving address of vmres08.auserver.com.au
18:38:18   Status:   Connecting to 101.0.117.31:21...
18:38:18   Status:   Connection established, waiting for welcome message...
18:38:18   Response:   220---------- Welcome to Pure-FTPd [privsep] [TLS] ----------
18:38:18   Response:   220-You are user number 2 of 60 allowed.
18:38:18   Response:   220-Local time is now 19:08. Server port: 21.
18:38:18   Response:   220-This is a private system - No anonymous login
18:38:18   Response:   220-IPv6 connections are also welcome on this server.
18:38:18   Response:   220 You will be disconnected after 15 minutes of inactivity.
18:38:18   Command:   AUTH TLS
18:38:18   Response:   234 AUTH TLS OK.
18:38:18   Status:   Initializing TLS...
18:38:18   Status:   Verifying certificate...
18:38:18   Status:   TLS connection established.
18:38:18   Command:   USER remotely
18:38:18   Response:   331 User xxxxxxx OK. Password required
18:38:18   Command:   PASS *********
18:38:18   Response:   230-Your bandwidth usage is restricted
18:38:18   Response:   230 OK. Current restricted directory is /
18:38:18   Command:   OPTS UTF8 ON
18:38:18   Response:   200 OK, UTF-8 enabled
18:38:18   Command:   PBSZ 0
18:38:18   Response:   200 PBSZ=0
18:38:18   Command:   PROT P
18:38:18   Response:   200 Data protection level set to "private"
18:38:18   Status:   Connected
18:38:18   Status:   Retrieving directory listing of "/public_html"...
18:38:18   Command:   CWD /public_html
18:38:18   Response:   250 OK. Current directory is /public_html
18:38:18   Command:   TYPE I
18:38:18   Response:   200 TYPE is now 8-bit binary
18:38:18   Command:   PASV
18:38:18   Response:   227 Entering Passive Mode (101,0,117,31,120,68)
18:38:18   Command:   MLSD
18:38:38   Error:   Connection timed out after 20 seconds of inactivity
18:38:38   Error:   Failed to retrieve directory listing

_________________
...Lyall
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Sun Nov 29, 2015 12:17 am    Post subject: Reply with quote

Still having this issue.
Only my Gentoo box is having problems.
I tried rebooting my adsl modem.
I tried re-plugging my 8 port switch, into which my PC is plugged.
I have tried turning off IPv6 in the adsl modem.
Wireshark does not show anything
It's interesting that tracepath on linux gives a different list of server names to tracert on windows 10.

Windows shows
Code:
Tracing route to the-pearces.com [101.0.117.31]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  fritz.box [192.168.1.254]
  2     7 ms     7 ms     7 ms  lns21.adl2.on.ii.net [203.16.215.199]
  3     8 ms     6 ms     7 ms  xe10-0-4.cr1.adl2.on.ii.net [150.101.33.203]
  4    27 ms    26 ms    28 ms  ae16.br1.syd4.on.ii.net [150.101.33.188]
  5    26 ms    26 ms    26 ms  55803.syd.equinix.com [202.167.228.136]
  6    27 ms    27 ms    27 ms  ae3-1000.cor02.eqsy3.digitalpacific.com.au [101.0.127.233]
  7    29 ms    27 ms    31 ms  ae11-10.dsw01-pod01.eqsy3.digitalpacific.com.au [101.0.127.54]
  8    26 ms    29 ms    26 ms  reth0-1073.fw02-pod01.eqsy3.digitalpacific.com.au [101.0.127.162]
  9    29 ms    29 ms    28 ms  vmres08.auserver.com.au [101.0.117.31]

Trace complete.
C:\WINDOWS\system32>route print
===========================================================================
Interface List
  3...00 15 e9 ad 72 06 ......D-Link AirPlus G DWL-G510 Wireless PCI Adapter(rev.C)
  8...00 16 e6 dc 18 8a ......Realtek PCIe GBE Family Controller
  9...08 00 27 00 d4 ea ......VirtualBox Host-Only Ethernet Adapter
  1...........................Software Loopback Interface 1
  6...00 00 00 00 00 00 00 e0 Microsoft Teredo Tunneling Adapter
  7...00 00 00 00 00 00 00 e0 Microsoft ISATAP Adapter
  4...00 00 00 00 00 00 00 e0 Microsoft ISATAP Adapter #2
===========================================================================

IPv4 Route Table
===========================================================================
Active Routes:
Network Destination        Netmask          Gateway       Interface  Metric
          0.0.0.0          0.0.0.0    192.168.1.254      192.168.1.5     10
        127.0.0.0        255.0.0.0         On-link         127.0.0.1    306
        127.0.0.1  255.255.255.255         On-link         127.0.0.1    306
  127.255.255.255  255.255.255.255         On-link         127.0.0.1    306
      169.254.0.0      255.255.0.0         On-link    169.254.54.107    276
   169.254.54.107  255.255.255.255         On-link    169.254.54.107    276
  169.254.255.255  255.255.255.255         On-link    169.254.54.107    276
      192.168.1.0    255.255.255.0         On-link       192.168.1.5    266
      192.168.1.5  255.255.255.255         On-link       192.168.1.5    266
    192.168.1.255  255.255.255.255         On-link       192.168.1.5    266
        224.0.0.0        240.0.0.0         On-link         127.0.0.1    306
        224.0.0.0        240.0.0.0         On-link       192.168.1.5    266
        224.0.0.0        240.0.0.0         On-link    169.254.54.107    276
  255.255.255.255  255.255.255.255         On-link         127.0.0.1    306
  255.255.255.255  255.255.255.255         On-link       192.168.1.5    266
  255.255.255.255  255.255.255.255         On-link    169.254.54.107    276
===========================================================================
Persistent Routes:
  None

IPv6 Route Table
===========================================================================
Active Routes:
 If Metric Network Destination      Gateway
  1    306 ::1/128                  On-link
  6    306 2001::/32                On-link
  6    306 2001:0:9d38:90d7:106e:48f:3485:14eb/128
                                    On-link
  8    266 2001:44b8:15c:6d00::/64  On-link
  8    266 2001:44b8:15c:6d00::/64  fe80::be05:43ff:fee6:f170
  8    266 2001:44b8:15c:6d00:703f:58d6:61d4:60f9/128
                                    On-link
  8    266 2001:44b8:15c:6d00:e570:9a64:4c16:40db/128
                                    On-link
  8    266 fe80::/64                On-link
  9    276 fe80::/64                On-link
  6    306 fe80::/64                On-link
  6    306 fe80::106e:48f:3485:14eb/128
                                    On-link
  8    266 fe80::703f:58d6:61d4:60f9/128
                                    On-link
  9    276 fe80::d5df:b8e6:7743:366b/128
                                    On-link
  1    306 ff00::/8                 On-link
  8    266 ff00::/8                 On-link
  9    276 ff00::/8                 On-link
  6    306 ff00::/8                 On-link
===========================================================================
Persistent Routes:
  None

C:\WINDOWS\system32>


Linux gives
Code:
# tracepath -b www.the-pearces.com
 1?: [LOCALHOST]                                         pmtu 1500
 1:  fritz.box (192.168.1.254)                             0.589ms
 1:  fritz.box (192.168.1.254)                             0.440ms
 2:  ppp235-20.static.internode.on.net (203.122.235.20)    0.486ms pmtu 1492
 2:  lns21.adl2.on.ii.net (203.16.215.199)                17.267ms
 3:  xe10-0-4.cr1.adl2.on.ii.net (150.101.33.203)         17.746ms
 4:  ae16.br1.syd4.on.ii.net (150.101.33.188)             75.386ms
 5:  55803.syd.equinix.com (202.167.228.136)              37.000ms
 6:  ae3-1000.cor02.eqsy3.digitalpacific.com.au (101.0.127.233)  37.570ms
 7:  ae11-10.dsw01-pod01.eqsy3.digitalpacific.com.au (101.0.127.54)  39.794ms
 8:  no reply
 9:  no reply
10:  no reply
  C-c C-c
root@lyalls-pc ~
# # route -nve
Kernel IP routing table
Destination     Gateway         Genmask         Flags   MSS Window  irtt Iface
0.0.0.0         192.168.1.254   0.0.0.0         UG        0 0          0 enp2s0
127.0.0.0       0.0.0.0         255.0.0.0       U         0 0          0 lo
127.0.0.0       127.0.0.1       255.0.0.0       UG        0 0          0 lo
192.168.1.0     0.0.0.0         255.255.255.0   U         0 0          0 enp2s0
root@lyalls-pc ~
#

_________________
...Lyall
Back to top
View user's profile Send private message
WWWW
Tux's lil' helper
Tux's lil' helper


Joined: 30 Nov 2014
Posts: 143

PostPosted: Sun Nov 29, 2015 12:39 am    Post subject: Reply with quote

I don't know whether this could help but relating to my recent post.

Have you checked your?:

Code:

/etc/resolv.conf


Make sure there are no shady nameservers inserted there. I recall long ago that a user had networking issues and he traced it back to this thing.

Try also a Gentoo LiveCD/SysrescueLiveCD and see if it's connects.

Another simple test is to temporarily rename your config to /etc/conf.d/net.bak then let the default openrc behavior do its thing getting a dhcp address for you. Then run some connection tests.

As puzzling as it may sound one time it happened to me like this that with my MANUAL configuration I was unable to connect to anything but letting default openrc do its automagic services for me worked.

cheers.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Sun Nov 29, 2015 1:04 am    Post subject: Reply with quote

What is really weird is everything else works, it's only the server on which my domains are hosted that is giving me grief.
The fact that it works on a windows 10 PC which is on the same table, or, even in a windows VirtualBox running ON the Gentoo host that is having the problem, is what gets me.
/etc/resolve.conf is auto generated by dhcpcd and only contains 2 entries, both pointing to my adsl modem, (a FritzBox 7390 running latest series 5 firmware).
My only setting in /etc/conf.d/net is
Code:
config_enp2s0="dhcp"

My /etc/dhcpcd.conf contains
Code:
# A sample configuration for dhcpcd.
# See dhcpcd.conf(5) for details.

# Inform the DHCP server of our hostname for DDNS.
hostname

# Use the hardware address of the interface for the Client ID.
# Enabling this may get a different lease than the kernel DHCP client.
# Some upstream DHCP servers may also require a ClientID, such as FRITZ!Box.
clientid
# or
# Use the same DUID + IAID as set in DHCPv6 for DHCPv4 ClientID as per RFC4361.
#duid

# Persist interface configuration when dhcpcd exits.
persistent

# Rapid commit support.
# Safe to enable by default because it requires the equivalent option set
# on the server to actually work.
option rapid_commit

# A list of options to request from the DHCP server.
option domain_name_servers, domain_name, domain_search, host_name
option classless_static_routes
# Most distributions have NTP support.
option ntp_servers
# Respect the network MTU.
# Some interface drivers reset when changing the MTU so disabled by default.
option interface_mtu

# A ServerID is required by RFC2131.
require dhcp_server_identifier

# Generate Stable Private IPv6 Addresses instead of hardware based ones
slaac private

# A hook script is provided to lookup the hostname if not set by the DHCP
# server, but it should not be run by default.
nohook lookup-hostname

# Allow users of this group to interact with dhcpcd via the control socket.
controlgroup wheel


_________________
...Lyall
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 6968

PostPosted: Sun Nov 29, 2015 5:30 am    Post subject: Reply with quote

lyallp wrote:

C:\WINDOWS\system32>route print
===========================================================================
Interface List
3...00 15 e9 ad 72 06 ......D-Link AirPlus G DWL-G510 Wireless PCI Adapter(rev.C)
8...00 16 e6 dc 18 8a ......Realtek PCIe GBE Family Controller
...
Network Destination Netmask Gateway Interface Metric
192.168.1.255 255.255.255.255 On-link 192.168.1.5 266
[/code]

You network configuration is more complex than what you said.
This traceroute from Windows tells us you have a realtek (typical pci card) and a wireless dlink.
And your 192.168.1.255 settings with a metric of 266 is a proof (assuming Windows metric works like linux ones) the wireless or a bounding is in use.

Metric is number of hops to reach another device, your metric should be set to 2 (1 to reach the asus switch + 1 to reach the fritzbox)
With wireless device (i think because of instability of packets sent and to reflect you will certainly resend more than once packets that were lost) metrics are set to 100+ ; this so when picking up a path, packets gets hint the best path is the easiest (lowest hops) and if packets have choice to be sent by the wired or wireless, wired will be use first.
Can you please provide route -n only (-nve hide the metric in use), i'm sure you'll see a 2 metric, proving in linux you are using only the realtek card.

Next to that: if Windows works fine using wireless and linux isn't but use the realtek.
The next device to look for trouble is then the asus switch that is between the realtek and the fritzbox (even if Windows is also connect to the asus, with bounding, wireless+realtek may be use, and when realtek fail, it can pickup the wireless path to succeed).
Is that a basic switch or does it have options? (basic switch have no ip and do the job, more advance switch or router have an ip and configuration options that may alter how a device connect to them will be). If it have options, you better check its configuration.
Or you can simply connect the realtek directly with the firtzbox and see if everything works with that configuration.
Back to top
View user's profile Send private message
lyallp
Veteran
Veteran


Joined: 15 Jul 2004
Posts: 1400
Location: Adelaide/Australia

PostPosted: Sun Nov 29, 2015 8:52 am    Post subject: Reply with quote

On the Windows PC, the wireless d-link was not connected, however, I have disabled that interface now.
Not that I expect any difference, that particular pc works just fine, it's the Gentoo Linux PC which is the problem.

Also, re-iterating the point that I can boot up a Virtualbox VM of Windows, ON the Gentoo Linux PC (which has the problem) and the Virtualbox Windows VM works just fine, it can access my domains.

The asus is a simple 8 port gigabit switch, nothing fancy at all. Been working fine for years, and still does, it seems.

From the Gentoo Linux box...
Code:
# route -n
Kernel IP routing table
Destination     Gateway         Genmask         Flags Metric Ref    Use Iface
0.0.0.0         192.168.1.254   0.0.0.0         UG    2      0        0 enp2s0
127.0.0.0       0.0.0.0         255.0.0.0       U     0      0        0 lo
127.0.0.0       127.0.0.1       255.0.0.0       UG    0      0        0 lo
192.168.1.0     0.0.0.0         255.255.255.0   U     2      0        0 enp2s0


Windows Traceroute
Code:
C:\Users\Jaxon>tracert www.the-pearces.com

Tracing route to the-pearces.com [101.0.117.31]
over a maximum of 30 hops:

  1    <1 ms    <1 ms    <1 ms  fritz.box [192.168.1.254]
  2     7 ms     7 ms     6 ms  lns21.adl2.on.ii.net [203.16.215.199]
  3     7 ms     6 ms     7 ms  xe10-0-4.cr1.adl2.on.ii.net [150.101.33.203]
  4    45 ms    27 ms    34 ms  ae16.br1.syd4.on.ii.net [150.101.33.188]
  5    26 ms    26 ms    26 ms  55803.syd.equinix.com [202.167.228.136]
  6    27 ms    27 ms    27 ms  ae3-1000.cor02.eqsy3.digitalpacific.com.au [101.0.127.233]
  7    27 ms    31 ms    27 ms  ae11-10.dsw01-pod01.eqsy3.digitalpacific.com.au [101.0.127.54]
  8    26 ms    26 ms    26 ms  reth0-1073.fw02-pod01.eqsy3.digitalpacific.com.au [101.0.127.162]
  9    29 ms    26 ms    26 ms  vmres08.auserver.com.au [101.0.117.31]

Trace complete.t


Gentoo Linux traceroute
Code:
root@lyalls-pc /etc
# tracepath www.the-pearces.com
 1?: [LOCALHOST]                                         pmtu 1500
 1:  fritz.box                                             0.489ms
 1:  fritz.box                                             0.417ms
 2:  ppp235-20.static.internode.on.net                     0.529ms pmtu 1492
 2:  lns21.adl2.on.ii.net                                 17.245ms
 3:  xe10-0-4.cr1.adl2.on.ii.net                          17.531ms
 4:  ae16.br1.syd4.on.ii.net                              47.766ms
 5:  55803.syd.equinix.com                                36.903ms
 6:  ae3-1000.cor02.eqsy3.digitalpacific.com.au           37.396ms
 7:  ae11-10.dsw01-pod01.eqsy3.digitalpacific.com.au      37.821ms
 8:  no reply
 9:  no reply
10:  no reply

_________________
...Lyall
Back to top
View user's profile Send private message
szatox
Veteran
Veteran


Joined: 27 Aug 2013
Posts: 1717

PostPosted: Sun Nov 29, 2015 11:13 am    Post subject: Reply with quote

:lol: my tracepath exceeded maximum number of 30 hops.
Really have a look at DNS configuration. You might be using different DNS on windows and on linux, and since VM guest has it's own DNS configuration it will work even if your host doesn't use any DNS at all.

Your tracers are not that much different. Linux reports 2 names for step 2 and that's it.
Also, windows and linux tracers use different protocols. One uses ICMP ping and the other UDP sent to some high port, and some hosts filter traffic so they won't send ICMP TTL exceeded message and then you see "no reply", but tracer keeps going because it may still get some answer from next hop if it does not filter traffic. Or you may be pointed to a wrong destination IP. (DNS again)
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Networking & Security All times are GMT
Goto page 1, 2  Next
Page 1 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum