Hi Tim,
Thanks very much for your email, I have been advised to check the server status module before, which was a very good point because I found out that despite the status_module was activated on both nodes I still had to tweak some configuration in httpd.conf to get the status page working on http://machine/server-status : Apache Server Status for 192.168.0.103 Server Version: Apache/2.2.17 (Ubuntu) PHP/5.3.5-1ubuntu7.2 with Suhosin-Patch Server Built: Sep 1 2011 09:31:14 Current Time: Saturday, 17-Sep-2011 13:03:11 BST Restart Time: Saturday, 17-Sep-2011 13:03:08 BST Parent Server Generation: 0 Server uptime: 2 seconds Total accesses: 0 - Total Traffic: 0 kB CPU Usage: u0 s0 cu0 cs0 0 requests/sec - 0 B/second - 1 requests currently being processed, 4 idle workers ____W........................................................... ................................................................ ................................................................ ................................................................ Scoreboard Key: "_" Waiting for Connection, "S" Starting up, "R" Reading Request, "W" Sending Reply, "K" Keepalive (read), "D" DNS Lookup, "C" Closing connection, "L" Logging, "G" Gracefully finishing, "I" Idle cleanup of worker, "." Open slot with no current process Srv PID Acc M CPU SS Req Conn Child Slot Client VHost Request 4-0 29147 0/0/0 W 0.00 0 0 0.0 0.00 0.00 192.168.0.2 node1 GET /server-status HTTP/1.1 My knowledge of corosync isn't deep enough, so bare with me if I am wrong, but my guess is that the error you are talking about on line 174 : + wget -O- -q -L --no-proxy --bind-address=127.0.0.1 http://localhost: happens because wget expects to see the status of the server at the URL http://localhost, but the right URL would be http://localhost/server-status As for Apache (apologies to the list for more apache log spam...) access.log only shows lines like 127.0.0.1 - - [17/Sep/2011:13:19:33 +0100] "GET / HTTP/1.0" 200 322 "-" "Wget/1.12 (linux-gnu)" error.log only shows lines like : [Sat Sep 17 13:19:32 2011] [notice] Apache/2.2.17 (Ubuntu) PHP/5.3.5-1ubuntu7.2 with Suhosin-Patch configured -- resuming normal operations [Sat Sep 17 13:19:33 2011] [notice] caught SIGTERM, shutting down Looks like I can only see the consequence of the Apache crash but not the cause. Since I am running corosync in a test VM I am going to wipe it out and restart from the top, because I really can't see why both nodes worked like a charm for 2 weeks and then this happened all of a sudden after I have rebooted the nodes. I remember playing around with crm configure edit before it all happened, so maybe I have broken something by accident. Thanks for your help G On 20 September 2011 11:10, Tim Serong <[email protected]> wrote: > On 20/09/11 19:49, Guillaume Bettayeb wrote: > > Hi Tim, > > > > Thanks for the debugging link. > > > > I have tried what the page (and yourself) suggested but I can't make any > > sense of the debugging output: > > > > http://pastebin.com/pwzReW5c > > > > > > it shows when apache stops but still doesn't say why it stops...unless I > > haven't properly figured out how to read that ? > > That ouput is the sequence of every command the RA executed from start > to finish. Apache is started between lines 93-99. The RA notices it's > running around about line 147. Then it tries to request the server > status page around line 174, using wget. This fails: > > + wget -O- -q -L --no-proxy --bind-address=127.0.0.1 http://localhost: > + rc=1 > > Presumably your apache server-status configuration is broken. It needs > to work with requests from/to localhost. Your apache error log may be > helpful at this point. > > Regards, > > Tim > -- > Tim Serong > Senior Clustering Engineer > SUSE > [email protected] > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
