Multicast issue - IGMP on switches (HP)



  • Server
    • FOG Version: 1.4
    • OS: Debian 8
    Client
    • Service Version:
    • OS:
    Description

    Hi, multicast deployment doesn’t work on my server, as clients hangs at the blue screen.
    I 've read carefully the wiki and performed the 1 client test with commands udp-receiver and sender. This test failed :
    On server side, i see a new connection with the correct ip, on client side, the clients seems to be ready, but whent i hit a key on the client, it just prompt " sending go signal" and nothing appears …
    The multicas.log shows the task, but when the transfer should begin, it shows " TimeOut"

    I’ve checked the network switches ( IGMP and SPT is enabled ) , on the server iptables is enable but without rules, ping 224.0.0.1 is ok, …
    If anyone has an idea, I would be very grateful .

    Thanks for your amazing work on FOG



  • @Sebastian-Roth You’re welcome …
    For info, my switches are HP Procurve and I used these links to configure them !

    https://community.spiceworks.com/how_to/26481-enable-igmp-on-hp-procurve

    http://whp-hou9.cold.extweb.hp.com/pub/networking/software/AdvTraff-Oct2005-59908853-Chap04-IGMP.pdf

    Now, Multicast is working properly : I imaged 3 clients at 1.4 Gb/min, next I will try 10 or more !


  • Developer

    @Flavalf Thanks for keeping us posted on this. This is very helpful information for others having issues with multicast!



  • Hi, feedback on my issue : it was ( as usual ) a network cause.
    IGMP was enable on switches but no switch was querier and no multicast group has been set.

    Now , the 1 client test is ok, I can go further and make it works !
    Thanks for support.



  • I’ve replaced the files in /var/www/html/fog/service/ipxe ( no need to reboot I guess ? )
    I’ve rebooted the client and … get the same issue :disappointed_relieved:
    I don’t know if it’s normal or not but I can’ t ping 224.0.0.1 from the client :
    0_1495567311380_Capture.PNG


  • Moderator

    @Flavalf I just chatted with one of the devs and he recommended that you test the 1.4.1RC0 inits. There was a fix put into those inits to address a multicasting issue. Your issue sounds similar.

    https://fogproject.org/inits/init.xz
    https://fogproject.org/inits/init_32.xz
    

    Download those and replace the ones on the fog server in /var/www/html/fog/service/ipxe directory.



  • Test with 1 client :

    multicast.log

    [05-23-17 8:01:32 pm] * No tasks found!
    [05-23-17 8:01:22 pm] * No tasks found!
    20:01:12.912924 
    20:01:12.912855 Transfer complete.
    20:01:12.909402 Disconnecting #0 (172.16.13.207)
    20:01:12.598314 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    [05-23-17 8:01:12 pm] * No tasks found!
    20:01:06.531724 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    [05-23-17 8:01:02 pm] * No tasks found!
    20:01:00.465126 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    20:00:54.398543 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    [05-23-17 8:00:52 pm] * No tasks found!
    20:00:48.331956 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    20:00:42.265365 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    [05-23-17 8:00:42 pm] * No tasks found!
    20:00:40.198868 Timeout notAnswered=[0] notReady=[0] nrAns=0 nrRead=0 nrPart=1 avg=10000
    20:00:40.108834 Starting transfer: 00000009
    [05-23-17 8:00:32 pm] * No tasks found!
    20:00:24.781281 Broadcasting control to 172.16.255.255
    20:00:24.781257 UDP sender for /opt/fog/.fogsettings at 172.16.1.111 on eth0 
    20:00:24.781134 Using mcast address 236.16.1.111
    
    

    On server and client :

    0_1495562681483_Capturemulti.PNG



  • On server :

    root@fogserver:~# ifconfig
    eth0      Link encap:Ethernet  HWaddr 44:37:e6:62:7c:8f
              inet adr:172.16.1.111  Bcast:172.16.255.255  Masque:255.255.0.0
              adr inet6: fe80::4637:e6ff:fe62:7c8f/64 Scope:Lien
              UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
              RX packets:45004809 errors:0 dropped:4 overruns:0 frame:0
              TX packets:3727057 errors:0 dropped:0 overruns:0 carrier:0
              collisions:0 lg file transmission:1000
              RX bytes:3524256236 (3.2 GiB)  TX bytes:126333435337 (117.6 GiB)
              Interruption:17
    
    lo        Link encap:Boucle locale
              inet adr:127.0.0.1  Masque:255.0.0.0
              adr inet6: ::1/128 Scope:Hôte
              UP LOOPBACK RUNNING  MTU:16436  Metric:1
              RX packets:122881 errors:0 dropped:0 overruns:0 frame:0
              TX packets:122881 errors:0 dropped:0 overruns:0 carrier:0
              collisions:0 lg file transmission:0
              RX bytes:14737940 (14.0 MiB)  TX bytes:14737940 (14.0 MiB)
    

    On client ( in debug task) :
    0_1495561745421_Capture.PNG


  • Moderator

    @Flavalf What is your subnet mask for both the fog server and the dhcp range. The reason why I ask is that there is a potential for the FOG server to be on another subnet (vlan) than your dhcp clients.

    FOG server
    172.16.1.111

    dhcp range
    172.16.13.0 to 172.16.16.255

    If your subnet mask is something like 255.255.0.0 then what you’ve posted is correct. What I’m trying to understand is if there is a router between the clients and the FOG server. FOG-casting on the same subnet shouldn’t be a problem between subnets then you might have an issue.



  • As I’m not a specialist, yes they are : fogserver is 172.16.1.111 and the dhcp range is
    172.16.13.0 to 172.16.16.255. the dhcp server is a windows 2012 server


  • Moderator

    Is the fog server and the target computers on the same IP subnet?


Log in to reply
 

424
Online

39.3k
Users

11.0k
Topics

104.6k
Posts

Looks like your connection to FOG Project was lost, please wait while we try to reconnect.