[omd-users] Fwd: [Check_mk (english)] Fwd: multisite from Main to Remote doesn't work! Help!

JB Segal jailbait at gmail.com
Wed Mar 28 22:24:17 CEST 2012


I'm forwarding this mail I sent to check_mk-en here in case there happens
to be anyone reading this who isn't reading that... It _might_ be an omd
problem, though I doubt it.

I apologize for the not-really-spam, but sort of OffTopic-ness (probably,
but maybe not) of this.
 = = = =
I've been banging my head against this for days and I don't know what
the HELL is going on.

Everything is this scenario is running OMD 0.52 with check_mk 1.1.12p7.

>From my desktop, in Boston, I go to my main multisite in DC. That
travels over an IPSec tunnel.

multisite/DC is configured to talk to multisite/Seattle and
multisite/London.
Those are also each a tunnel away. Seattle is the SAME isp, London isn't.

DC is the hub of all the tunneling. There's no direct route from
Boston<->London (or Seattle)

Seattle works fine. I can see host and service status. I can click on
a pnp icon and go to a page of graphs. The 'Hosts' snapin renders
'Seattle', then 'DC' (which has a "> Main Directory" subitem) in no
time.
(though Seattle SHOULD, I think, have a similar item, while the apache
logs say something's wrong:
10.1.22.10 - - [26/Mar/2012:11:07:44 -0500] "GET
/seattle/check_mk/ajax_wato_files.py HTTP/1.1" 401 401
"http://{dcmachine}/prod_dc/check_mk/side.py"
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:11.0) Gecko/20100101
Firefox/11.0"
)

------> Note to omd list - I realize now that the Hosts snapin _never_
worked with the remote sites, even when I was on 0.52's default 1.1.12p6
check_mk/multisite. Does anyone else see this?

London is messed up... but it's not ALL the way messed up. It's just
PARTS of it.

 - I can view host and service statuses Just Fine.
 - I can ssh to the DC host and use elinks to talk to London. I can
use netcat to talk to the livestatus socket.

-- But I can't view pnp graphs. The 'Hosts' snapin draws the 'London'
button and then hangs for 75 seconds and then times out (according to
firebug) before rendering the close of the snapin box and the rest of
the side bar.

 - For a host in Seattle, I can hit the WATO button and get an edit page.
(10.1.22.10 - omdadmin [26/Mar/2012:17:13:10 -0500] "GET
/seattle/check_mk/images/contextlink_hi.png HTTP/1.1" 200 617 "
http://10.20.129.149/seattle/check_mk/wato.py?mode=edithost&host={seattleHost}"
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:11.0) Gecko/20100101
Firefox/11.0" )
 - For a host in London, I wait another 75 seconds and get a firefox
timeout page in the main frame.

In firebug, I see this URL:
http://10.1.250.125/eu01/check_mk/wato.py?mode=edithost&host={london
machine} with a status of "Aborted".

DC's apache logs say
10.1.22.10 - omdadmin [26/Mar/2012:18:14:29 -0400] "GET
/prod_dc/check_mk/view.py?view_name=host&host={londonmachine}t&site=london&_display_options=htbfcoderw&selected_rows=&_do_actions=&_ajaxid=1332800068000
HTTP/1.1" 200 5620
"http://{dcmachine}/prod_dc/check_mk/view.py?view_name=host&host={londonmachine}&site=london"
"Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6;rv:11.0) Gecko/20100101
Firefox/11.0"
is fine, while edithost is never logged.

WHAT IS GOING ON??

I'm totally confused. Does this make any sense to anyone else?
Please? Someone?

Thanks,
JB, feeling pathetic.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mathias-kettner.de/pipermail/omd-users/attachments/20120328/c2e2f358/attachment.html>


More information about the omd-users mailing list