Re: [vserver] multiple pormap daemons?

From: Petar Hitij <petahi_at_gmail.com>
Date: Mon 17 May 2010 - 13:48:21 BST
Message-ID: <AANLkTinKo8Cae8_4z8jjDGCYYO2tXUqpfiXYwgCjFSmK@mail.gmail.com>

On Mon, May 17, 2010 at 2:16 PM, JF Straeten <jfstraeten@scarlet.be> wrote:
> On Mon, May 17, 2010 at 01:10:08PM +0200, Petar Hitij wrote:
>> The -i option of portmapper was my first attempt. It failed
>> because nfs on host did'nt work.
>
> I think it's the opposite : NFS can't start because portmap isn't
> running...

portmap is running. It is not listening on 127.0.0.1 so other daemons
cannot register.

I cannot get portmap to listen just to 127.0.0.1 and to eth0 interface
ip. It is one or the other.

>
>
>>    rpcinfo -p
>>    rpcinfo: can't contact portmapper: RPC: Remote system error -
>> Connection refused
>>
>>    /etc/init.d/nfs-common start
>>    Starting NFS common utilities: statd failed!
>>
>> Maybe I should try to fix this somehow.
>
> If I remember correctly, you're under Debian *Lenny* ?
>
> If it's right, you should try to fix the boot script order. Portmap
> should start first, and the services which need it only after, but OOB
> it doesn't come in the right order under Lenny.
>
> I've a host here running portmap, with some guests running it too :
>
> kelsen:~# vps -auxw|grep portmap
> Warning: bad ps syntax, perhaps a bogus '-'? See http://procps.sf.net/faq.html
> daemon    2171     0 MAIN           0.0  0.0   8020   516 ?        Ss   Apr07   0:00 /sbin/portmap -i 127.0.0.1
> root     22916     0 MAIN           0.0  0.0   7236   840 ?        S+   13:44   0:00 grep portmap
> daemon   24747  5009 aghora         0.0  0.0   8020   544 ?        Ss   May04   0:00 /sbin/portmap
> daemon   26395  5070 fulci          0.0  0.0   8020   576 ?        Ss   May04   0:00 /sbin/portmap
> daemon   27694  5014 oda            0.0  0.0   8020   584 ?        Ss   May04   0:01 /sbin/portmap
> daemon   28106  5079 othon          0.0  0.0   8020   572 ?        Ss   May04   0:00 /sbin/portmap
>
>
> At least 4 guests are running portmap on the same host without a
> problem to report.
>
> NB 1) out of memory : it you set "-i 127.0.0.1" for portmap on the
>   host, IIRC you should specify the 'port' and 'mountport' options on
>   the clients for the NFS mount to succeed (they can't contact the
>   host portmap since it's not running on the network IP, just on the
>   lo, so a hint is necessary)

True, but not maintanable.

>
> NB 2) to reorder the boot sequence, I use "file-rc" in the guests : it
>   has a single conf file (/etc/runlevel.conf) and is more easy than
>   the habitual mess of symlink (ok, it's a personal taste ; nothing
>   objective...).
>
> Hih,
>
> --
>
> JFS.
>

There is one more option - use one portmapper for all vservers.
It doesn't work because portmap is filtering register requests
from "non-local" ip numbers.

Looks like I have been bitten by some old bugs:
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=280537

Regards
Petar
Received on Mon May 17 13:49:12 2010

[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Mon 17 May 2010 - 13:49:15 BST by hypermail 2.1.8