[Xorp-hackers] XORP Testing

[Vikram]

The system freezes when I try to pick up the default-system-config in
any case ie either through the config file or when I commit the
changes using xorpsh.
When any XORP process is starting, there is no freeze but I do receive
a few warnings.

Here is a list of the processes that were running before the freeze occured:-

  PID  Uid     VmSize Stat Command
    1 root       1160 S   /bin/bash
    2 root            SWN [ksoftirqd/0]
    3 root            SW  [watchdog/0]
    4 root            SW< [events/0]
    5 root            SW< [khelper]
    6 root            SW< [kthread]
   10 root            SW< [kblockd/0]
   31 root            SW  [pdflush]
   32 root            SW  [pdflush]
   34 root            SW< [aio/0]
   33 root            SW  [kswapd0]
  104 root            SW< [kseriod]
  133 root            SW< [rpciod/0]
  135 root       1104 S   sh
  140 root        428 S   /usr/sbin/telnetd
  141 root       1080 S   -sh
  142 root       1072 S   -sh
  183 root        924 S   /sbin/syslogd -n
  193 root       1016 S   -sh
  196 root       9760 S   ./xorp_rtrmgr -- -v
  197 root        808 S   tail -f /var/log/messages
  200 root       5308 S   /usr/local/xorp/bin/xorpsh
  201 root       7244 S   xorp_fea
  212 root       5200 S   xorp_rib
  213 root       3136 S   xorp_policy
  214 root       4232 S   xorp_static_routes
  215 root       2892 S   xorp_rl_interfaces
  243 root        984 S   -sh
  248 root       2924 S   xorp_rl_system

I am using a slow system [400Mhz] and the mounted root volume is on
nfs, through eth0 , and have specified default-system-config for eth0
and the problem doesnt go away.
What do you think is the problem?


On 12/19/06, Pavlin Radoslavov <pavlin at icir.org> wrote:
> > I have made no changes with eth0. I just tell it to pick up the
> > default-system-config. Even then the same problem occurs. There are no
> > routes defined in my configuration file. There is only a defination of
> > eth0 which is supposed to pick up the default-system-config.
>
> Interesting.
> When exactly do you see the freeze and how long time does it take
> for it to go away (if it ever goes away)?
> E.g., do you see it when the XORP processes are starting?
> If yes, do you see it if you have an empty "interfaces" section or
> even an empty configuration file.
> I am trying to find whether just starting the XORP processes (or any
> large processes) themselves triggers the freeze.
>
> Pavlin
>
> > On 12/18/06, Pavlin Radoslavov <pavlin at icir.org> wrote:
> > > > Here is my test environment. My XORP router is running on a slow
> > > > machine whose root volume is mounted via nfs on eth0. I telnet into
> > > > the XORP router and perform all configurations. Whenever I try to
> > > > anything with eth0, the telnet sessions just freeze up. The telnet
> > > > sessions are also coming through eth0. I would like to know if this is
> > > > normal for the telnet sessions to freeze or is there something wrong
> > > > with my setup or anything else?
> > >
> > > What kind of changes do you do with eth0?
> > > Obviously, if you change the IP address or take the interface down,
> > > this will affect all network traffic via eth0 (including telnet and
> > > NFS).
> > > Also, note that other operations like changing the MTU or the MAC
> > > address will trigger (internally in the FEA) taking the interface
> > > down, performing the change and then setting it up again.
> > > Also, have in mind that if there are NFS issues, the interactive
> > > sessions might also be affected.
> > >
> > > Regards,
> > > Pavlin
> > >
> >
> > _______________________________________________
> > Xorp-hackers mailing list
> > Xorp-hackers at icir.org
> > http://mailman.ICSI.Berkeley.EDU/mailman/listinfo/xorp-hackers
>