About this list Date view Thread view Subject view Author view Attachment view

From: Herbert Poetzl (herbert_at_13thfloor.at)
Date: Wed 07 Sep 2005 - 02:05:51 BST


On Tue, Sep 06, 2005 at 10:02:18AM +0200, Nicolas Costes wrote:
> Le Lundi 5 Septembre 2005 17:49, Nicolas Costes a écrit :
> > I noticed a few bugs when using util-vserver.208 (The Mandrake rpms):
>
> I forgot two other issues :
>
> > - /var/lib/run/vservers.rev is not created, so the vservers do not
> > start - /etc/init.d/vserver-default cannot be added by "chkconfig
> > --add" because the corresponding line in the script is contains "-"
> > instead of "345".
>
> *) As said in a previous mail, /var/lib/run/vservers.rev seems to never be
> cleaned of its symlinks. Multiple ctxID point to a given vserver... I
> worried about some symlinks not having a trailing "/", but it's only
> because i've deleted the corresponding vserver ;-)
> # ls -l /var/lib/run/vservers.rev/
> total 0
> 49152 -> /etc/vservers/filesv
> 49153 -> /etc/vservers/distcc1/
> 49154 -> /etc/vservers/filesv
> 49155 -> /etc/vservers/filesv

as I always say, thou shalt not use dynamic
context ids :)

anyway, thanks for the report, I guess Enrico
will look at it sooner or later ...

> *) When I stop a vserver on some hosts, I get this message :
> # vserver foo stop
> /usr/lib/util-vserver/vserver.stop: line 85: 9287 Killed
> "${NICE_CMD[@]}" ${USE_VNAMESPACE:+$_VNAMESPACE --enter "$S_CONTEXT" -- }
> $_VCONTEXT $SILENT_OPT --migrate --chroot --xid "$S_CONTEXT" --
> "${INITCMD_STOP[@]}"
>
> And by the way, stopping a vserver on any host is faaaaaarrrr longer than
> starting one.

probably because it has to wait until the
timeout (30 secs IIRC) is reached and the guest
is finally force killed .. why? because the
scripts inside the guest do not shut down the
processes properly ... (we might work around
that soon, for now, you can reduce the timeout)

> *) Ok, the last is kernel related (iptables, I think) : I run Samba in a
> vserver, but I installed it without Cups (I don't need it). Samba
> complains in its own logs that it can't join Cupsd on port 631, etc...
> And in the kernel logs (/var/log/kernel/warnings):
> ...
> Sep 6 09:57:01 hostmachine kernel: PROTO=6 127.0.0.1:59338 127.0.0.1:631
> L=60 S=0x00 I=19015 F=0x4000T=64
> Sep 6 09:57:01 hostmachine kernel: ip_local_deliver: bad skb: PRE_ROUTING
> LOCAL_IN LOCAL_OUT POST_ROUTING
> Sep 6 09:57:01 hostmachine kernel: skb: pf=2 (unowned) dev=lo len=40
> Sep 6 09:57:01 hostmachine kernel: PROTO=6 127.0.0.1:631 127.0.0.1:59338
> L=40 S=0x00 I=2817 F=0x4000 T=64
> Sep 6 09:57:01 hostmachine kernel: ip_local_deliver: bad skb: PRE_ROUTING
> LOCAL_IN LOCAL_OUT POST_ROUTING
> Sep 6 09:57:01 hostmachine kernel: skb: pf=2 (unowned) dev=lo len=60
> Sep 6 09:57:01 hostmachine kernel: PROTO=6 127.0.0.1:59339 127.0.0.1:631
> L=60 S=0x00 I=28353 F=0x4000T=64
> Sep 6 09:57:01 hostmachine kernel: ip_local_deliver: bad skb: PRE_ROUTING
> LOCAL_IN LOCAL_OUT POST_ROUTING
> Sep 6 09:57:01 hostmachine kernel: skb: pf=2 (unowned) dev=lo len=40
> ...
>
> This is the only vserver host who does that here... I made something
> wrong ?

connections over 127.0.0.1 are somewhat tricky,
so that might change if you use real ips ...
(but maybe it's just a samab/cups issue)

best,
Herbert

> --
> ,,
> (°> Nicolas Costes
> /|\ IUT de La Roche / Yon
> ( ^ ) Clé publique: http://www.keyserver.net/
> ^ ^ Musique libre: http://musique-legale.info/ -
> http://www.jamendo.com/?s=concept

> _______________________________________________
> Vserver mailing list
> Vserver_at_list.linux-vserver.org
> http://list.linux-vserver.org/mailman/listinfo/vserver

_______________________________________________
Vserver mailing list
Vserver_at_list.linux-vserver.org
http://list.linux-vserver.org/mailman/listinfo/vserver


About this list Date view Thread view Subject view Author view Attachment view
[Next/Previous Months] [Main vserver Project Homepage] [Howto Subscribe/Unsubscribe] [Paul Sladen's vserver stuff]
Generated on Wed 07 Sep 2005 - 02:06:19 BST by hypermail 2.1.3