I have a host 192.168.32.20 that I monitor with Munin, and everything is working fine.
However, I also use it for monitoring routers via SNMP.
So, on 192.168.32.20 in /etc/munin/munin.conf, I have a stanza like this:
which has always worked in the past.
However, since an upgrade to Munin 1.3.4, the HTML output for router1 now includes the graphs for the host 192.168.32.20:
# ls -1 /var/www/localhost/htdocs/munin/router1/router1/*.html
This is wrong. router1 is a Cisco router, and definitely isn't running Apache :)
(I also monitor other routers via SNMP, and they're all showing the same problems)
Listing the plugins available for the node though shows just the snmp ones, as it should.
# telnet 192.168.32.20 4949
Connected to 192.168.32.20.
Escape character is '^]'.
# munin node at foo
snmp_router1_if_8 snmp_router1_if_2 snmp_router1_if_11 snmp_router1_if_10 snmp_router1_if_15 snmp_router1_if_12 snmp_router1_if_13 snmp_router1_if_7 snmp_router1_if_4 snmp_router1_if_1 snmp_router1_if_14 snmp_router1_if_5 snmp_router1_if_9 snmp_router1_if_3 snmp_router1_if_6
Connection closed by foreign host.
But regressing back to net-analyzer/munin-1.3.3-r2 makes it all work again.
So I'd suggest not closing this until this issue is fixed in a newer version than 1.3.4.
Upstream issue, please report it there.