Am Freitag, den 09.07.2010, 11:40 +0200 schrieb Schaefer, Dirk
Alexander:
> Hello,
> 
>  
> 
> I'm trying to setup a two nodes ha loadbalancer for two webservers and i've
> got two curious problems with heartbeat/ldirectord.
> 
>  
> 
> The first is, that I cannot active the broadcast mechanism. If i uncomment
> 'bcast ethXY' in the ha.cf file then the following error is reported in the
> logs during the startup:
> 
>  
> 
>                 heartbeat[31484]: 2010/07/09_11:10:04 ERROR: glib: Error
> setting socket option SO_BINDTODEVICE: Protocol not available
> 
> heartbeat[31484]: 2010/07/09_11:10:04 ERROR: cannot open bcast eth0
> 
>  
> 
> and the service is getting stopped.
> 
>  
> 
> the second problem is that although everything seems to work fine I cannot
> access the webServers through the vip. Ipvsadmin shows that the vip and the
> real servers are registered correctly. It also shows that requests to the
> vip are getting received. Even more, the failover to the second
> heartbeat/ldirectord machine works once I shutdown the current muster. But
> what ever I do the requests to the vip seems not to get routed/nated/. (I've
> tried gate and masq definitions in ldirectord conf) to the real servers.
> 
>  
> 
> My setup:
> 
>  
> 
> Os: gentoo 10.1 (amd64) (XEN DomU)
> 
> Heartbeat: 2.0.8
> 
>  
> 
> ha.cf:
> 
> debugfile /var/log/ha-debug
> 
> logfile       /var/log/ha-log
> 
> logfacility   local0
> 
> mcast eth0 225.0.0.1 694 1 0
> 
> auto_failback off
> 
> node   lb1 lb2
> 
> respawn root /usr/lib64/heartbeat/ipfail
> 
> apiauth ipfail gid=cluster uid=cluster
> 
>  
> 
> 
> 
> harecourses:
> 
> lb1 \
> 
>         ldirectord::/etc/ha.d/ldirectord.www.cf \
> 
>         LVSSyncDaemonSwap::master \
> 
>         IPaddr2::172.22.40.50/20/eth1/172.22.47.255
> 
>  
> 
> 
> 
> ldirector.www.cf:
> 
> checktimeout=10
> 
> checkinterval=2
> 
> autoreload=yes
> 
> logfile="local0"
> 
> logfile="/var/log/ldirectord.log"
> 
> quiescent=yes
> 
>  
> 
> virtual=172.22.40.50:80
> 
>         real=172.22.40.110:80 masq
> 
>         real=172.22.40.111:80 masq
> 
>         service=http
> 
>         request="lbProbe.html"
> 
>         receive="OK"
> 
>         scheduler=rr
> 
>  
> 
> 
> 
> kernel config:
> 
> CONFIG_IP_VS=m
> 
> # CONFIG_IP_VS_IPV6 is not set
> 
> CONFIG_IP_VS_DEBUG=y
> 
> CONFIG_IP_VS_TAB_BITS=12
> 
> # IPVS transport protocol load balancing support
> 
> CONFIG_IP_VS_PROTO_TCP=y
> 
> CONFIG_IP_VS_PROTO_UDP=y
> 
> CONFIG_IP_VS_PROTO_AH_ESP=y
> 
> CONFIG_IP_VS_PROTO_ESP=y
> 
> CONFIG_IP_VS_PROTO_AH=y
> 
> CONFIG_IP_VS_PROTO_SCTP=y
> 
> # IPVS scheduler
> 
> CONFIG_IP_VS_RR=m
> 
> CONFIG_IP_VS_WRR=m
> 
> CONFIG_IP_VS_LC=m
> 
> CONFIG_IP_VS_WLC=m
> 
> CONFIG_IP_VS_LBLC=m
> 
> CONFIG_IP_VS_LBLCR=m
> 
> CONFIG_IP_VS_DH=m
> 
> CONFIG_IP_VS_SH=m
> 
> CONFIG_IP_VS_SED=m
> 
> CONFIG_IP_VS_NQ=m
> 
> # IPVS application helper
> 
> CONFIG_IP_VS_FTP=m
> 
> # CONFIG_SCSI_MVSAS is not set
> 
>  
> 
> I have no idea what to check else. Can someone imagine what the problem
> could be?
> 
>  
> 
> Thanks a lot and .
> 
>  
> 
> Mit freundlichen Gruessen / With kind regards
> 
>  
> 
> Dirk Alexander Schaefer
> 
>  
> 
> _______________________________________________
> Linux-HA mailing list
> [email protected]
> http://lists.linux-ha.org/mailman/listinfo/linux-ha
> See also: http://linux-ha.org/ReportingProblems


First of all, do NOT use 2.0.8 any more. It is for about 5 years out of
date, buggy and, well, bad.


-- 
Dr. Michael Schwartzkopff
Guardinistr. 63
81375 München

mob: 0163 172 50 98

_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to