Mailing List Archive: I start backhand, but I can only see myself on the status page. Why?

All,

I apologize if this subject has been dealt with before...I looked through
the archives for the last several months and couldn't find anything
applicable.

At the FAQ, there is a question (stated in the subject) that is
representative of my problem [1]. I think I have the cluster set up
correctly (including the recommendations at [1]); however, when I "hit" the
backhand status/diagnostic page, only the local host is shown (in red).

I assume that this is a problem with my configuration instead of a problem
with the diagnostics page itself. I looked at the tcpdump utility output
and it appears that both application servers are broadcasting information
(as they should) - tcpdump snippets below. However, how can I tell if the
local application server (i.e. mod_backhand module) is "accepting"
broadcasts from the the other application servers?

Any help you guys could give me regarding a fix or a way to further diagnose
the problem would be much appreciated.

[1] http://www.backhand.org/mod_backhand/FAQ.shtml#question12

Cheers
--
Here is the cluster information:
--
Application server 1 (AS1): 163.1.13.16
Application server 2 (AS2): 163.1.13.17
--

Here are the relevant httpd.conf snippets:
--
On AS1:

<IfModule mod_backhand.c>
UnixSocketDir /usr/local/apache/backhand
MulticastStats 163.1.13.16 163.1.13.255:4445
AcceptStats 163.1.13.0/24

<Location "/backhand/">
SetHandler backhand-handler
</Location>
</IfModule>

On AS2:

<IfModule mod_backhand.c>
UnixSocketDir /usr/local/apache/backhand
MulticastStats 163.1.13.17 163.1.13.255:4445
AcceptStats 163.1.13.0/24

<Location "/backhand/">
SetHandler backhand-handler
</Location>
</IfModule>
--

Here are the tcpdump snippets:
--

From AS1:

[root@climateapps1:/usr/local/apache/bin]# tcpdump |grep 163.1.13.255
tcpdump: listening on eth0
07:46:12.231417 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:12.397260 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:13.231375 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:13.397257 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:14.231383 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:14.397252 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:15.231402 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:15.397269 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:16.231495 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:46:16.397252 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)

212 packets received by filter
0 packets dropped by kernel

From AS2:

[root@climateapps2:/usr/local/apache/logs]# tcpdump |grep 163.1.13.255
tcpdump: listening on eth0
07:44:22.494652 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:23.328584 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:23.494604 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:24.328587 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:24.494613 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:25.328602 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:25.494597 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:26.328591 climateapps1.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)
07:44:26.494595 climateapps2.oucs.ox.ac.uk.4445 > 163.1.13.255.4445: udp 100
(DF)

162 packets received by filter
0 packets dropped by kernel
--

--
Steve Bannerman
steve.bannerman@comlab.ox.ac.uk
44.(0)1865.273866

All,

I found out why my mod_backhand modules were not "accepting" traffic. We're
using the Linux operating system and our firewall definitions on them
(iptables) were not allowing UDP traffic for the appropriate subnetwork.
Once I added this rule, the /backhand/ diagnostics page showed both servers
in the cluster (on each machine).

Hopefully this information will help someone else at some point in time.

Cheers
--
Steve Bannerman
steve.bannerman@comlab.ox.ac.uk
44.(0)1865.273866

> -----Original Message-----
> From: Steve Bannerman [mailto:steve.bannerman@comlab.ox.ac.uk]
> Sent: 21 August 2003 08:19
> To: backhand-users@lists.backhand.org
> Subject: I start backhand, but I can only see myself on the status page.
> Why?
>
>
> All,
>
> I apologize if this subject has been dealt with before...I looked
> through the archives for the last several months and couldn't
> find anything applicable.
>
> At the FAQ, there is a question (stated in the subject) that is
> representative of my problem [1]. I think I have the cluster set
> up correctly (including the recommendations at [1]); however,
> when I "hit" the backhand status/diagnostic page, only the local
> host is shown (in red).
>
> I assume that this is a problem with my configuration instead of
> a problem with the diagnostics page itself. I looked at the
> tcpdump utility output and it appears that both application
> servers are broadcasting information (as they should) - tcpdump
> snippets below. However, how can I tell if the local application
> server (i.e. mod_backhand module) is "accepting" broadcasts from
> the the other application servers?
>
> Any help you guys could give me regarding a fix or a way to
> further diagnose the problem would be much appreciated.
>
> [1] http://www.backhand.org/mod_backhand/FAQ.shtml#question12
>
> Cheers
> --
> Here is the cluster information:
> --
> Application server 1 (AS1): 163.1.13.16
> Application server 2 (AS2): 163.1.13.17
> --
>
> Here are the relevant httpd.conf snippets:
> --
> On AS1:
>
> <IfModule mod_backhand.c>
> UnixSocketDir /usr/local/apache/backhand
> MulticastStats 163.1.13.16 163.1.13.255:4445
> AcceptStats 163.1.13.0/24
>
> <Location "/backhand/">
> SetHandler backhand-handler
> </Location>
> </IfModule>
>
> On AS2:
>
> <IfModule mod_backhand.c>
> UnixSocketDir /usr/local/apache/backhand
> MulticastStats 163.1.13.17 163.1.13.255:4445
> AcceptStats 163.1.13.0/24
>
> <Location "/backhand/">
> SetHandler backhand-handler
> </Location>
> </IfModule>
> --
>
> Here are the tcpdump snippets:
> --
>
> From AS1:
>
> [root@climateapps1:/usr/local/apache/bin]# tcpdump |grep 163.1.13.255
> tcpdump: listening on eth0
> 07:46:12.231417 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:12.397260 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:13.231375 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:13.397257 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:14.231383 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:14.397252 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:15.231402 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:15.397269 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:16.231495 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:46:16.397252 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
>
> 212 packets received by filter
> 0 packets dropped by kernel
>
> From AS2:
>
> [root@climateapps2:/usr/local/apache/logs]# tcpdump |grep 163.1.13.255
> tcpdump: listening on eth0
> 07:44:22.494652 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:23.328584 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:23.494604 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:24.328587 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:24.494613 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:25.328602 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:25.494597 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:26.328591 climateapps1.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
> 07:44:26.494595 climateapps2.oucs.ox.ac.uk.4445 >
> 163.1.13.255.4445: udp 100 (DF)
>
> 162 packets received by filter
> 0 packets dropped by kernel
> --
>
> --
> Steve Bannerman
> steve.bannerman@comlab.ox.ac.uk
> 44.(0)1865.273866
>