Network Conections Problems

Home » CentOS » Network Conections Problems
CentOS 6 Comments

Hi;
I have a CentOS 6.8 x86_64 server where just run Oracle 11g 64 bit data base server. Is happen that sometimes it loose all connections (no ping, I can not access via ssh, no TNSping of Oracle server have success). When this happend I make ping from this server to some IP address of my network and then everything work fine again. I check logs looking for some mistake or problem in the network device, but I do not find anything or maybe I not looking where must be. I check that from a workstation (Windows XP or Windows 7 or Windows 10) when I make ping to the server, after ~ 8 seconds in get answer and always the first one is loose. If you try again ping, everything work fine (no answers are loose)

C:\Users\administrator>ping pgtm.gtm.gob.cu

Haciendo ping a pgtm.gtm.gob.cu [192.168.41.4] con 32 bytes de datos:
Tiempo de espera agotado para esta solicitud. Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64 Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64 Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64 Estadísticas de ping para 192.168.41.4: Paquetes: enviados = 4, recibidos = 3, perdidos = 1 (25% perdidos), Tiempos aproximados de ida y vuelta en milisegundos: Mínimo = 0ms, Máximo = 0ms, Media = 0ms This are some of the configs related with networks [root@pgtm ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0
TYPE=Ethernet UUID=11dcddd4-6530-457a-8d3e-01a8339fb113
ONBOOT=yes NM_CONTROLLED=yes BOOTPROTO=none HWADDR=6C:92:BF:26:C7:02
IPADDR=192.168.41.4
PREFIX=24
GATEWAY=192.168.41.1
DNS1=192.168.41.17
DOMAIN=gtm.gob.cu DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME=”System eth0″

cat /etc/hosts
127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.41.4 pgtm pgtm.gtm.gob.cu

[root@pgtm ~]# cat /etc/resolv.conf
# Generated by NetworkManager search gtm.gob.cu nameserver 192.168.41.17

[root@pgtm mail]# cat /etc/sysconfig/network NETWORKING=yes HOSTNAME=pgtm.gtm.gob.cu GATEWAY=192.168.41.1

Someone that happend something like this, or some place for where keep looking to try to fix this loose of conection?
Thank and sorry for my horryble English.

Rommel Rodriguez Toirac rommelrt@nauta.cu

6 thoughts on - Network Conections Problems

  • First guess is that you may have two devices on the network with the same ip address.

    Next time this happens, try doing

    1. ‘arp -n’ from a machine other than the db server
    2. ping the other machine from the db server, then
    3. ‘arp -n’ from the other machine

    Compare the outputs of the two invocations of arp. If the outputs show different MAC addresses for 192.168.41.4 then you have two different devices with the same IP address.

  • El 21 de febrero de 2017 7:00:03 GMT-05:00, CentOS-request@CentOS.org escribió:

    I resolve the problem of network conection loose, but still a dude of how it happend and how it can be fixe. When I check with arping the MAC of sever change, for example

    rommel@p6:~$ arping 192.168.41.4
    ARPING 192.168.41.4 from 192.168.41.6 enp3s0
    Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.653ms Unicast reply from 192.168.41.4 [6C:92:BF:26:C7:03] 0.683ms Unicast reply from 192.168.41.4 [6C:92:BF:26:C7:03] 0.622ms Unicast reply from 192.168.41.4 [6C:92:BF:26:C7:03] 0.631ms
    ^CSent 3 probes (1 broadcast(s))
    Received 4 response(s)

    The first answer is with a MAC diferent to the others one. But when I arping from the server inseft look the MAC associate to de IP address:

    [root@pgtm ] arping 192.168.41.4 -I eth1
    ARPING 192.168.41.4 from 192.168.41.4 eth1
    Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.658ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.654ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.654ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.662ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.655ms Sent 5 probes (1 broadcast(s))
    Received 5 response(s)

    Looking in the config of network device I can not find the MAC 00:1D:09:FF:44:4B

    [root@pgtm ] ifconfig eth0 Link encap:Ethernet HWaddr 6C:92:BF:26:C7:02
    UP BROADCAST MULTICAST MTU:1500 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
    Memory:c7220000-c723ffff eth1 Link encap:Ethernet HWaddr 6C:92:BF:26:C7:03
    inet addr:192.168.41.4 Bcast:192.168.41.255 Mask:255.255.255.0
    inet6 addr: fe80::6e92:bfff:fe26:c703/64 Scope:Link
    UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
    RX packets:95819 errors:0 dropped:0 overruns:0 frame:0
    TX packets:1924 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:11728605 (11.1 MiB) TX bytes:263674 (257.4 KiB)
    Memory:c7200000-c721ffff eth2 Link encap:Ethernet HWaddr 00:E0:ED:33:4E:9C
    UP BROADCAST MULTICAST MTU:1500 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
    Memory:c7120000-c713ffff eth3 Link encap:Ethernet HWaddr 00:E0:ED:33:4E:9D
    UP BROADCAST MULTICAST MTU:1500 Metric:1
    RX packets:0 errors:0 dropped:0 overruns:0 frame:0
    TX packets:0 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:1000
    RX bytes:0 (0.0 b) TX bytes:0 (0.0 b)
    Memory:c7100000-c711ffff lo Link encap:Local Loopback
    inet addr:127.0.0.1 Mask:255.0.0.0
    inet6 addr: ::1/128 Scope:Host
    UP LOOPBACK RUNNING MTU:65536 Metric:1
    RX packets:249609 errors:0 dropped:0 overruns:0 frame:0
    TX packets:249609 errors:0 dropped:0 overruns:0 carrier:0
    collisions:0 txqueuelen:0
    RX bytes:52090343 (49.6 MiB) TX bytes:52090343 (49.6 MiB)

    The solution was another IP address to this network device and then everything work fine. Why this happend? How I can erase the link beteewn MAC 00:1D:09:FF:44:4B and IP 192.168.41.4? Where can be stored this link?
    Right now in the network is not assignet the IP address 192.168.41.4 to no one device (printserver, switch, router, workstation or server) and still whe I make arping have the answer:

    rommel@p6:~$ arping 192.168.41.4
    ARPING 192.168.41.4 from 192.168.41.6 enp3s0
    Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.631ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.623ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.623ms Unicast reply from 192.168.41.4 [00:1D:09:FF:44:4B] 0.691ms
    ^CSent 4 probes (1 broadcast(s))
    Received 4 response(s)

    arping to the new server IP address look like this (this is the MAC of eth0, where a plug the network cable):

    rommel@p6:~$ arping 192.168.41.7
    ARPING 192.168.41.7 from 192.168.41.6 enp3s0
    Unicast reply from 192.168.41.7 [6C:92:BF:26:C7:02] 0.580ms Unicast reply from 192.168.41.7 [6C:92:BF:26:C7:02] 0.607ms Unicast reply from 192.168.41.7 [6C:92:BF:26:C7:02] 0.613ms Unicast reply from 192.168.41.7 [6C:92:BF:26:C7:02] 0.594ms
    ^CSent 4 probes (1 broadcast(s))
    Received 4 response(s)

    Rommel Rodriguez Toirac rommelrt@nauta.cu

  • Yup, my guess is that someone has plugged in another device into your network and that device has a ip address 192.168.41.4 statically assigned. Or someone has reconfigured a device and set that address by accident. I had a UPS do this to me once.

    A quick lookup on the https://macvendors.com/ shows the offending mac address belongs to a device from Dell Inc. That may help you track down the offender.

    Hope this helps.


    Kahlil (Kal) Hodgson GPG: C9A02289
    Chief Technology Officer (m) +61 (0) 4 2573 0382
    Direct Pricing Exchange Pty Ltd

    Suite 1415
    401 Docklands Drive Docklands VIC 3008 Australia

    “All parts should go together without forcing. You must remember that the parts you are reassembling were disassembled by you. Therefore, if you can’t get them together again, there must be a reason. By all means, do not use a hammer.” — IBM maintenance manual, 1925

  • Hi;
    I have a CentOS 6.8 x86_64 server where just run Oracle 11g 64 bit data base server. Is happen that sometimes it loose all connections (no ping, I
    can not access via ssh, no TNSping of Oracle server have success). When this happend I make ping from this server to some IP address of my network and then everything work fine again. I check logs looking for some mistake or problem in the network device, but I do not find anything or maybe I not looking where must be. I check that from a workstation (Windows XP or Windows 7 or Windows 10)
    when I make ping to the server, after ~ 8 seconds in get answer and always the first one is loose. If you try again ping, everything work fine (no answers are loose)

    C:\Users\administrator>ping pgtm.gtm.gob.cu

    Haciendo ping a pgtm.gtm.gob.cu [192.168.41.4] con 32 bytes de datos:
    Tiempo de espera agotado para esta solicitud. Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64
    Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64
    Respuesta desde 192.168.41.4: bytes=32 tiempo<1m TTL=64

    Estadísticas de ping para 192.168.41.4:
    Paquetes: enviados = 4, recibidos = 3, perdidos = 1
    (25% perdidos), Tiempos aproximados de ida y vuelta en milisegundos:
    Mínimo = 0ms, Máximo = 0ms, Media = 0ms

    This are some of the configs related with networks

    [root@pgtm ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0
    DEVICE=eth0
    TYPE=Ethernet UUID=11dcddd4-6530-457a-8d3e-01a8339fb113
    ONBOOT=yes NM_CONTROLLED=yes BOOTPROTO=none HWADDR=6C:92:BF:26:C7:02
    IPADDR=192.168.41.4
    PREFIX=24
    GATEWAY=192.168.41.1
    DNS1=192.168.41.17
    DOMAIN=gtm.gob.cu DEFROUTE=yes IPV4_FAILURE_FATAL=yes IPV6INIT=no NAME="System eth0"
    L

    I noticed from your ipcfg-eth0 file, the address 192.168.41.1 is assigned to eth0. When you typed the command "ifconfig", that same address was assigned to eth1. Try changing the name of this file to ipcfg-eth1 and restart networkmanager.

    As was stated earlier, check your gateway and see what other devices could be using that address. Put a block on that unknown MAC address if you can.

    cat /etc/hosts
    127.0.0.1 localhost localhost.localdomain localhost4
    localhost4.localdomain4
    ::1 localhost localhost.localdomain localhost6
    localhost6.localdomain6
    192.168.41.4 pgtm pgtm.gtm.gob.cu

    [root@pgtm ~]# cat /etc/resolv.conf
    # Generated by NetworkManager search gtm.gob.cu nameserver 192.168.41.17

    [root@pgtm mail]# cat /etc/sysconfig/network NETWORKING=yes HOSTNAME=pgtm.gtm.gob.cu GATEWAY=192.168.41.1

    Someone that happend something like this, or some place for where keep looking to try to fix this loose of conection?
    Thank and sorry for my horryble English.

    Rommel Rodriguez Toirac rommelrt@nauta.cu

  • One thing to try if you have access to the switch where all the devices are plugged in:
    *$ sudo ping -f 192.168.41.4*

    Then watch the switch for a lot of activity on 2 ports – the one you are plugged into and another. The other busy port will be the one that’s causing you grief. If you have daisy chained switches, then it might take a little longer to track it down.

    Happy hunting – :).

  • Another solution, if you are the admin with access to the router, is to null route the packets from that IP. The person using that device will let you know shortly after (lack of Internet access is the quickest way to get the attention of users on the network):

    $ sudo ip r a blackhole 192.168.41.4