Mailing List Archive

(no subject)
(no subject) [ In reply to ]
I retried this, starting up just the second machine....all I get is a
message in syslog "no such interface". what is it talking about? i've got
an eth0, which is what i'm trying to assign the vip to. Is there a flag to
increase the debugging?
(no subject) [ In reply to ]
Wednesday, April 2, 2003, 10:33:20 AM, John Jolet wrote:

JJ> I retried this, starting up just the second machine....all I get is a
JJ> message in syslog "no such interface". what is it talking about? i've got
JJ> an eth0, which is what i'm trying to assign the vip to. Is there a flag to
JJ> increase the debugging?

JJ> _______________________________________________
JJ> wackamole-users mailing list
JJ> wackamole-users@lists.backhand.org
JJ> http://lists.backhand.org/mailman/listinfo/wackamole-users


Hi John,

If you run wackamole -d, it prints debugging info to screen and not
to syslog, besides giving out more information.

Ashima mailto:munjal@cnds.jhu.edu
-----------------------------------
(no subject) [ In reply to ]
I don't know of any flags to increase debugging output... afaik its an
all-or-nothing thing. Please send your wackamole.conf files (or just
one, if they're identical), your spread.conf file, and the output from
wackamole run with -d. I'll take a look, and see if I can help.

--Ryan

John Jolet wrote:
> I retried this, starting up just the second machine....all I get is a
> message in syslog "no such interface". what is it talking about? i've got
> an eth0, which is what i'm trying to assign the vip to. Is there a flag to
> increase the debugging?
>
> _______________________________________________
> wackamole-users mailing list
> wackamole-users@lists.backhand.org
> http://lists.backhand.org/mailman/listinfo/wackamole-users
>
(no subject) [ In reply to ]
This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_000_01C2F938.C6FE6200
Content-Type: text/plain;
charset="iso-8859-1"

Here are the conf files, but when i try wackamole with the -d flag, it gives
a segmentation fault, on both machines. without that flag, it works on one
machine (who's physically on the same subnet with the vip subnet), but not
on the machine that's on a different subnet.

-----Original Message-----
From: Ryan Caudy [mailto:caudy@jhu.edu]
Sent: Wednesday, April 02, 2003 10:09 AM
To: wackamole-users@lists.backhand.org
Subject: Re: [Wackamole-users] (no subject)


I don't know of any flags to increase debugging output... afaik its an
all-or-nothing thing. Please send your wackamole.conf files (or just
one, if they're identical), your spread.conf file, and the output from
wackamole run with -d. I'll take a look, and see if I can help.

--Ryan

John Jolet wrote:
> I retried this, starting up just the second machine....all I get is a
> message in syslog "no such interface". what is it talking about? i've
got
> an eth0, which is what i'm trying to assign the vip to. Is there a flag
to
> increase the debugging?
>
> _______________________________________________
> wackamole-users mailing list
> wackamole-users@lists.backhand.org
> http://lists.backhand.org/mailman/listinfo/wackamole-users
>


_______________________________________________
wackamole-users mailing list
wackamole-users@lists.backhand.org
http://lists.backhand.org/mailman/listinfo/wackamole-users


------_=_NextPart_000_01C2F938.C6FE6200
Content-Type: application/octet-stream;
name="spread.conf"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="spread.conf"

# Blank lines are permitted in this file.=0A=
# spread.conf sample file=0A=
# =0A=
# questions to spread@spread.org=0A=
#=0A=
=0A=
#MINIMAL REQUIRED FILE=0A=
#=0A=
# Spread should work fine on one machine with just the uncommented =0A=
# lines below. The rest of the file documents all the options and=0A=
# more complex network setups.=0A=
#=0A=
# This configures one spread daemon running on port 4803 on =
localhost.=0A=
=0A=
Spread_Segment 239.0.0.1:4803 {=0A=
wiz 10.16.0.54=0A=
panda 10.32.0.236=0A=
rage 207.8.16.100=0A=
}=0A=
=0A=
=0A=
# Spread options=0A=
#-----------------------------------------------------------------------=
----=0A=
#-----------------------------------------------------------------------=
----=0A=
#Set what internal Spread events are logged to the screen or file =0A=
# (see EventLogFile).=0A=
# Default setting is to enable PRINT and EXIT events only. =0A=
#The PRINT and EXIT types should always be enabled. The names of others =
are:=0A=
# EXIT PRINT DEBUG DATA_LINK NETWORK PROTOCOL SESSION =0A=
# CONFIGURATION MEMBERSHIP FLOW_CONTROL STATUS EVENTS =0A=
# GROUPS MEMORY SKIPLIST ALL NONE =0A=
# ALL and NONE are special and represent either enabling every type =
=0A=
# or enabling none of =
them.=0A=
# You can also use a "!" sign to negate a type, =0A=
# so { ALL !DATA_LINK } means log all events except data_link =
ones.=0A=
=0A=
#DebugFlags =3D { PRINT EXIT }=0A=
=0A=
#Set whether to log to a file as opposed to stdout/stderr and what =0A=
# file to log to.=0A=
# Default is to log to stdout.=0A=
#=0A=
#If option is not set then logging is to stdout.=0A=
#If option is set then logging is to the filename specified.=0A=
# The filename can include a %h or %H escape that will be replaced at =
runtime=0A=
# by the hostname of the machine upon which the daemon is running.=0A=
# For example "EventLogFile =3D spreadlog_%h.log" with 2 machines =0A=
# running Spread (machine1.mydomain.com and machine2.mydomain.com) =
will=0A=
# cause the daemons to log to "spreadlog_machine1.mydomain.com.log" =
and=0A=
# "spreadlog_machine2.mydomain.com.log" respectively.=0A=
=0A=
#EventLogFile =3D testlog.out=0A=
=0A=
#Set whether to add a timestamp in front of all logged events or =
not.=0A=
# Default is no timestamps. Default format is "[%a %d %b %Y =
%H:%M:%S]".=0A=
#If option is commented out then no timestamp is added.=0A=
#If option is enabled then a timestamp is added with the default =
format=0A=
#If option is enabled and set equal to a string, then that string is =
used=0A=
# as the format string for the timestamp. The string must be a valid =
time=0A=
# format string as used by the strftime() function.=0A=
=0A=
#EventTimeStamp=0A=
# or=0A=
EventTimeStamp =3D "[%a %d %b %Y %H:%M:%S]"=0A=
=0A=
#Set whether to allow dangerous monitor commands =0A=
# like "partition, flow_control, or kill"=0A=
# Default setting is FALSE.=0A=
#If option is set to false then only "safe" monitor commands are =
allowed =0A=
# (such as requesting a status update).=0A=
#If option is set to true then all monitor commands are enabled. =0A=
# THIS IS A SECURTIY RISK IF YOUR NETWORK IS NOT PROTECTED!=0A=
=0A=
#DangerousMonitor =3D false=0A=
=0A=
#Set handling of SO_REUSEADDR socket option for the daemon's TCP=0A=
# listener. This is useful for facilitating quick daemon restarts =
(OSes=0A=
# often hold onto the interface/port combination for a short period of =
time=0A=
# after daemon shut down).=0A=
#=0A=
# AUTO - Active when bound to specific interfaces (default).=0A=
# ON - Always active, regardless of interface.=0A=
# SECURITY RISK FOR ANY OS WHICH ALLOW DOUBLE BINDS BY DIFFERENT =
USERS=0A=
# OFF - Always off.=0A=
=0A=
#SocketPortReuse =3D ON=0A=
=0A=
#Sets the runtime directory used when the Spread daemon is run as =
root=0A=
# as the directory to chroot to. Defaults to the value of the=0A=
# compile-time preprocessor define SP_RUNTIME_DIR, which is =
generally=0A=
# "/var/run/spread".=0A=
=0A=
#RuntimeDir =3D /var/run/spread=0A=
=0A=
#Sets the unix user that the Spread daemon runs as (when launched as=0A=
# the "root" user). Not effective on a Windows system. Defaults to=0A=
# the user and group "spread".=0A=
=0A=
#DaemonUser =3D spread=0A=
#DaemonGroup =3D spread=0A=
=0A=
=0A=
#Set the list of authentication methods that the daemon will allow=0A=
# and those which are required in all cases.=0A=
# All of the methods listed in "RequiredAuthMethods" will be =
checked,=0A=
# irregardless of what methods the client chooses.=0A=
# Of the methods listed is "AllowedAuthMethods" the client is=0A=
# permitted to choose one or more, and all the ones the client =
chooses=0A=
# will also be checked.=0A=
#=0A=
# To support older clients, if NULL is enabled, then older clients =
can=0A=
# connect without any authentication. Any methods which do not =
require=0A=
# any interaction with the client (such as IP) can also be enabled=0A=
# for older clients. If you enable methods that require interaction,=0A=
# then essentially all older clients will be locked out.=0A=
#=0A=
#The current choices are:=0A=
# NULL for default, allow anyone authentication=0A=
# IP for IP based checks using the spread.access_ip file=0A=
=0A=
#RequiredAuthMethods =3D " "=0A=
#AllowedAuthMethods =3D "NULL"=0A=
=0A=
#Set the current access control policy.=0A=
# This is only needed if you want to establish a customized policy.=0A=
# The default policy is to allow any actions by authenticated =
clients.=0A=
#AccessControlPolicy =3D "PERMIT"=0A=
=0A=
=0A=
# network description line.=0A=
# Spread_Segment <multicast address for subnet> <port> {=0A=
# port is optional, if not specified the default 4803 port is used.=0A=
=0A=
#Spread_Segment 127.0.0.255:4803 {=0A=
=0A=
# either a name or IP address. If both are given, than the name is =
taken =0A=
# as-is, and the IP address is used for that name.=0A=
=0A=
# localhost 127.0.0.1=0A=
#}=0A=
# repeat for next sub-network=0A=
=0A=
#Spread_Segment x.2.2.255 {=0A=
=0A=
# other1 128.2.2.10=0A=
# 128.2.2.11=0A=
# other3.my.com=0A=
#}=0A=
# Spread will feel free to use broadcast messages within a =
sub-network.=0A=
# if you do not want this to happen, you should specify your machines =
on=0A=
# different logical sub-networks.=0A=
=0A=
# IP-Multicast addresses can also be used as the multicast address =
for=0A=
# the logical sub-network as in this example. If IP-multicast is =
supported=0A=
# by the operating system, then the messages will only be received=0A=
# by those machines who are in the group and not by all others in the =
same=0A=
# sub-network as happens with broadcast addresses=0A=
=0A=
#Spread_Segment 225.0.1.1:3333 {=0A=
# mcast1 1.2.3.4=0A=
# mcast2 1.2.3.6=0A=
#}=0A=
=0A=
# Multi-homed host setup=0A=
#=0A=
# If you run Spread on hosts with multiple interfaces you may want to =
=0A=
# control which interfaces Spread uses for client connections and =
for=0A=
# the daemon-to-daemon (and monitor control) messages. This can be =
done=0A=
# by adding an extra stanza to each configured machine. =0A=
#=0A=
#Sample:=0A=
#=0A=
#Spread_Segment 225.0.1.1 {=0A=
# multihomed1 1.2.3.4 {=0A=
# D 192.168.0.4=0A=
# C 1.2.3.4 }=0A=
# multihomed2 1.2.3.5 {=0A=
# D 192.168.0.5=0A=
# C 1.2.3.5=0A=
# C 127.0.0.1 }=0A=
# multihomed3 1.2.3.6 {=0A=
# 192.168.0.6=0A=
# 1.2.3.6 }=0A=
#}=0A=
# This configuration sets up three multihomed machines into a Spread =
segment.=0A=
# The first host has a 'main' IP address of 1.2.3.4 and listens for =
client=0A=
# connections only on that interface. All daemon-to-daemon UDP =
multicasts and=0A=
# the tokens and any monitor messages must use the 192.168.0.4 =
interface.=0A=
# The second host multihomed2 has a similar setup, except it also =
listens for=0A=
# client connections on the localhost interface as well as the 1.2.3.5 =
interface.=0A=
# If you make any use of the extra interface stanza ( a { } block ) =
then you must=0A=
# explicitly configure ALL interfaces you want as Spread removes all =
defaults when=0A=
# you use the explicit notation.=0A=
# The third multihomed3 host uses a shorthand form of omitting the D or =
C option and=0A=
# just listening for all types of traffic and events on both the =
192.168.0 and 1.2.3 =0A=
# networks. If no letter is listed before the interface address then =
ALL types of =0A=
# events are handled on that interface.=0A=

------_=_NextPart_000_01C2F938.C6FE6200
Content-Type: application/octet-stream;
name="wackamole.conf"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: attachment;
filename="wackamole.conf"

# The Spread daemon we are going to connect to. It should be on the =
local box=0A=
Spread =3D 4803=0A=
SpreadRetryInterval =3D 5s=0A=
# The group name=0A=
Group =3D wack1=0A=
# Named socket for online control=0A=
Control =3D /var/run/wack.it=0A=
=0A=
# Denote the interface we prefer to have=0A=
#prefer eth0:10.3.4.5/8=0A=
#prefer eth0:0.0.0.0=0A=
#prefer { eth0:10.2.3.4/8 eth1:192.168.10.23/24 }=0A=
=0A=
# In most cases, I just don't care. Let wackamole decide.=0A=
Prefer None=0A=
=0A=
# List all the virtual interfaces (ALL of them)=0A=
VirtualInterfaces {=0A=
# The following two lines have the same effect=0A=
# en0:192.168.1.2/24=0A=
{ eth0:10.32.0.235/23 }=0A=
# This is how you say 2 or more IPs are to be treated as a single=0A=
# "set" or "virtual interface". If wackamole decides that this=0A=
# machine will manage it, you are ensured to get ALL the ips in the=0A=
# set.=0A=
#{ en1:10.0.0.1/8 en0:192.168.35.64/26 }=0A=
}=0A=
=0A=
# Collect and broadcast the IPs in our ARP table every so often=0A=
Arp-Cache =3D 90s=0A=
=0A=
# List who we will notify=0A=
# Here the netblock (/24 or /28) can be deceptive. It is NOT a =
netmask=0A=
# for a single IP. It is how one will describe that they want to=0A=
# notify ALL IPs in a segment.=0A=
Notify {=0A=
# Let's notify our router:=0A=
eth0:10.32.1.254/32=0A=
# Notify out DNS servers=0A=
eth0:10.12.0.1/32 throttle 128=0A=
# 10.0.0.0 -> 10.0.0.255, but only 128 notifications/sec=0A=
en0:10.32.0.0/23 throttle 128=0A=
# Wackamole shares arp-cache across machines, this says to=0A=
# notify every IP address in the aggregate shared arp-cache.=0A=
arp-cache=0A=
}=0A=
balance {=0A=
# This field is the maximum number of IP addresses that will move=0A=
# from one wackamole to another during a round of balancing.=0A=
AcquisitionsPerRound =3D all=0A=
# Time interval in each balancing round.=0A=
interval =3D 4s=0A=
}=0A=
# How long it takes us to mature=0A=
mature =3D 5s=0A=

------_=_NextPart_000_01C2F938.C6FE6200--
(no subject) [ In reply to ]
John Jolet wrote:
> Here are the conf files, but when i try wackamole with the -d flag, it gives
> a segmentation fault, on both machines. without that flag, it works on one
> machine (who's physically on the same subnet with the vip subnet), but not
> on the machine that's on a different subnet.

-d will set the debug flags (all of them)

There is a stack smash problem in 2.0.0. You can try the latest CVS version
to see if that fixes your segmentation fault.

Alternatively you can set your stack size to be _big_ (like 4MB) and see if
that helps.

--
Theo Schlossnagle
Principal Consultant
OmniTI Computer Consulting, Inc. -- http://www.omniti.com/
Phone: +1 410 872 4910 x201 Fax: +1 410 872 4911
1024D/82844984/95FD 30F1 489E 4613 F22E 491A 7E88 364C 8284 4984
2047R/33131B65/71 F7 95 64 49 76 5D BA 3D 90 B9 9F BE 27 24 E7
(no subject) [ In reply to ]
> John Jolet wrote:
> > I retried this, starting up just the second machine....all I get is a
> > message in syslog "no such interface". what is it talking about? i've
> got
> > an eth0, which is what i'm trying to assign the vip to. Is there a flag
> to
> > increase the debugging?

I took a peak at your wackamole.conf

Do you have an en0 interface that is listed under Notify section:

en0:10.32.0.0/23

Cheers,
-.mag
(no subject) [ In reply to ]
oops, no, it's eth0. I got it to work, and it was failing over, then both
boxes started saying invalid session on spread connect. I don't think i
changed anything before it started working, but maybe so...i'll fix those
statements and see what it does.

-----Original Message-----
From: Mark A. Garcia [mailto:wackamole-users@wavecentral.com]
Sent: Wednesday, April 02, 2003 12:47 PM
To: wackamole-users@lists.backhand.org
Subject: Re: [Wackamole-users] (no subject)


> John Jolet wrote:
> > I retried this, starting up just the second machine....all I get is a
> > message in syslog "no such interface". what is it talking about? i've
> got
> > an eth0, which is what i'm trying to assign the vip to. Is there a flag
> to
> > increase the debugging?

I took a peak at your wackamole.conf

Do you have an en0 interface that is listed under Notify section:

en0:10.32.0.0/23

Cheers,
-.mag


_______________________________________________
wackamole-users mailing list
wackamole-users@lists.backhand.org
http://lists.backhand.org/mailman/listinfo/wackamole-users