Am Sun, 29 May 2016 03:00:56 -0700 "Ngie Cooper (yaneurabeya)" <yaneurabeya_at_gmail.com> schrieb: > > On May 29, 2016, at 00:32, O. Hartmann <ohartman_at_zedat.fu-berlin.de> wrote: > > > > After updating sources and build- and installworld, I realize that all rpcbind related > > services, so far NFS, are not working. On a client I check the start of rpcbind by > > setting option -d and receive the output shown below. > > > > Well, prior to r300949 rpcbind started without complains - as it did with r300901. > > > > [...] > > rpcbind not running? > > Starting rpcbind. > > rpcbind debugging enabled. > > can't get local ip6 address: hostname nor servname provided, or not known > > couldn't create ip6 socket/etc/rc.d/rpcbind: WARNING: failed to start rpcbind > > Hi, > Could you please try this patch with -d (it’ll continue on instead of exiting… > I’m curious as to why it was failing before). Does IPv6 work in your environment? > Thanks! > -Ngie Well, funny is: staring rpcbind with options "-h 192.168.xxx.xxx -l -s -W" does not make the rpcbind trying to create an IPv6 socket - which it fails due to absence of IPv6 in kernel, but only "-h 192.168.xxx.xxx" does! I can reproduce this weird behaviour on several CURRENT systems on which IPv6 is disabled by default.
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:05 UTC