Ping within same vlan two hops?? (Now it's ->Destination Host Unreachable) HELP!



  • Hi all,

    Anyone know why ping within a same vlan on windows showed two hops?

    e.g.
    The gateway of vlan is 192.168.5.1
    PC A: 192.168.5.10
    PC B: 192.168.5.11

    when ping from 192.168.5.10 to 192.168.5.11
    there are two hops instead of one
    1. 192.168.5.1
    2. 192.168.5.11

    How do I reduce it to only one hop??

    Thanks



  • ping doesnt show hops (not on any of the OS' i use anyways)

    post some more details about your setup (schematic, screenshots of config, ….)


  • LAYER 8 Global Moderator

    Do you mean traceroute?  Show the actual output of your traceroute then..

    Only reason you would hit your gateway before talking to a box you think should be on same network is your mask is wrong..



  • @heper:

    ping doesnt show hops (not on any of the OS' i use anyways)

    post some more details about your setup (schematic, screenshots of config, ….)

    My fault, it's tracert, not ping.
    After rebooting the switch and pfsense, the situation is changed a bit, instead of 2 hops, now two clients within same vlan can't even ping each other.
    Here's images.

    Ping between those two PCs always ending with "Destination Host Unreachable"
    Other than that, internet/ ping PCs in other vlans/lan all works.














  • @johnpoz:

    Do you mean traceroute?  Show the actual output of your traceroute then..

    Only reason you would hit your gateway before talking to a box you think should be on same network is your mask is wrong..

    Yep, tracert that is.
    After rebooting, the gateway problem disappeared, now those two PCs can't ping each other, but pinging PCs in other vlan/lan works, internet works for those PCs too.
    No idea what's going on….


  • LAYER 8 Global Moderator

    Wireless can not ping each other? Then you prob have isolation on your AP.. This is very common setting, where you do not want wireless clients to talk to each other.  Depending on the maker of your AP it could be called AP Isolation, Station Isolation, Client Isolation or in pfsense if wifi is on it called "Allow intra-BSS communication"



  • @johnpoz:

    Wireless can not ping each other? Then you prob have isolation on your AP.. This is very common setting, where you do not want wireless clients to talk to each other.  Depending on the maker of your AP it could be called AP Isolation, Station Isolation, Client Isolation or in pfsense if wifi is on it called "Allow intra-BSS communication"

    Interesting things is, I just migrated the WLAN vlans from ue1 to ue0, and wireless clients was able to ping each other.
    Nothing changed at AP side.

    :-\ :-\


  • LAYER 8 Global Moderator

    Sorry but pfsense has ZERO to do with your wireless clients talking to each other when connected to same AP.  Pfsense doesn't even need to be connected for clients of an AP to talk to each other..  Nor does your switch.. Other than getting dhcp, a gateway or using dns the rest of your network has zero to do with 2 clients connected to the same AP talking to each other.

    Client A pinging B connected to the same wifi with IPs that are in the same network has zero to do with the rest of your network in the drawing.  you sure you didn't mess up and 1 client is vlan 10 and the other is in vlan 12?



  • @johnpoz:

    Sorry but pfsense has ZERO to do with your wireless clients talking to each other when connected to same AP.  Pfsense doesn't even need to be connected for clients of an AP to talk to each other..

    Yea, probably I should try reboot the AP, I didn't do that after migrated the vlan to new interface.
    I'll try reboot, see how it goes, also the inter-vlan smb transfer speed is slow as well  :-\



  • [Solved]

    After changing of settings, vlan, interface etc.
    Reboot Switch, pfsense, AP.

    For ping issue, turn off "Client ARP Caching" in AP, so that ARP request will not be blocked by AP thus passed on to switch/pfsense etc. then ping will work


  • LAYER 8 Global Moderator

    " AP thus passed on to switch/pfsense etc. then ping will work"

    Huh.. Sorry that is not how it works..

    Its not forwarding on to your switch/pfsense

    Your talking a Cisco AP right??

    When ARP caching is disabled, the access point forwards all ARP requests through the radio port to associated clients, and the client to which the ARP request is directed responds. When ARP caching is enabled, the access point responds to ARP requests for associated clients and does not forward requests to clients. When the access point receives an ARP request for an IP address not in the cache, the access point drops the request and does not forward it.



  • @johnpoz:

    " AP thus passed on to switch/pfsense etc. then ping will work"

    Huh.. Sorry that is not how it works..

    Its not forwarding on to your switch/pfsense

    Your talking a Cisco AP right??

    When ARP caching is disabled, the access point forwards all ARP requests through the radio port to associated clients, and the client to which the ARP request is directed responds. When ARP caching is enabled, the access point responds to ARP requests for associated clients and does not forward requests to clients. When the access point receives an ARP request for an IP address not in the cache, the access point drops the request and does not forward it.

    Yep, your quote is right.
    Here's a more visual way to interpret ENABLED/DISABLED/OPTIONAL
    http://www.my80211.com/cisco-auton-cli-commands/2010/3/12/autonomous-understanding-cisco-ap-arp-caching-disabled-enabl.html

    And I suspect it's this part "When ARP caching is enabled, the access point responds to ARP requests for associated clients and does not forward requests to clients. When the access point receives an ARP request for an IP address not in the cache, the access point drops the request and does not forward it." was causing the trouble.

    Don't know why there was no problem at all before migrating the vlan. :o Any ideas?


  • LAYER 8 Global Moderator

    Your clients changed IP?  Cache was wrong.. your AP should be able to cache and work just fine..



  • @johnpoz:

    Your clients changed IP?  Cache was wrong.. your AP should be able to cache and work just fine..

    Um, it's weird, those two are both assigned with static IP from DHCP/pfsense, and "ARP Table Static Entry" ticked



  • @johnpoz:

    Your clients changed IP?  Cache was wrong.. your AP should be able to cache and work just fine..

    Do you know why when copying file from one lan to another vlan (two separate physical nic) speed is around 5-16MB/s
    But copying from lan to vlan (same single one nic) speed is around 26MB/s

    I can't figure out why…. :'(


  • LAYER 8 Global Moderator

    Is this wifi or wired?

    26MB is like watching paint dry for any sort of local copy..  So I take it this is wireless?

    If anything your hairpin should be slower then when using 2 different nics.. But what other traffic is flowing at time of copy on these nics?



  • @johnpoz:

    Is this wifi or wired?

    26MB is like watching paint dry for any sort of local copy..  So I take it this is wireless?

    If anything your hairpin should be slower then when using 2 different nics.. But what other traffic is flowing at time of copy on these nics?

    Yep, both situation is server on 1Gbps cable, client on 300Mbps wifi.
    For cable to cable it will be better.
    I can accept 26Mbps when on wifi, but 5 - 16Mbps is definitely too slow  :o

    I mean, for both conditions, gateway is involved, only difference is the first one is on same nic/port, second one involves two nic/ports.
    But all of them are 1Gbps, and I don't see high CPU usage or high memory usage  :-\ :-\


Log in to reply