Hi there I was able to narrow the problem down to a 'Huawei' router by running spine manually on the CLI... # ./spine --first=644 --last=807 -R --verbosity=5 --snmponly --option=max_threads:1 Disabling the host solves the problem, but I need this host graphed... Host description: System:Huawei Versa...

Moin Sorry for the late response. The spine version I'm using is 'SPINE 0.8.7g'. Cacti is also on 0.8.7g. I've just checked out 'SPINE 0.8.8-alpha' from svn and compiled it. Unfortunately it does not work due to changes in the db schema for 0.8.8. FATAL: MySQL Error:'1146', Table 'cacti.device' does...

G'day Last Friday morning spine decided to segfault with exit status 11. Running spine with strace ends with that: futex(0x2b674893994c, FUTEX_WAKE_PRIVATE, 1) = 0 futex(0x42e02810, FUTEX_WAKE_PRIVATE, 1) = 1 rt_sigreturn(0x42e02810) = 0 munmap(0x2aaaae643000, 409602/16/2011 01:44:35 PM - SPINE: Pol...

Hi Thanks, but doesn't help me on CentOS... I've actually just read, that thread_concurrency only applies on Solaris.. That would explain to me why the poller got much faster, but the GUI build-up (besides rendering all the rrd stuff) has not improved a lot.. Guess the thread stuff goes out of my.cn...

Hi guys.. A couple of mysql tweaks reduced the already quite fast polling to half of the time. New: 08/27/2010 05:45:43 PM - SYSTEM STATS: Time:41.4858 Method:spine Processes:5 Threads:10 Hosts:2341 HostsPerProcess:469 DataSources:25950 RRDsProcessed:18202 Before: 05/06/2010 04:56:33 PM - SYSTEM STA...

Hi there I can only think of two possible reasons... 1. You're running net-snmp (not snmp trough php). --> That would trigger Cacti to issue system calls from php to the snmp binaries containing all the information required for the command to be executed... 2. You're running script queries which exp...

G'day Our server specs... HP DL385G2 2 x Dual-Core AMD Opteron(tm) Processor 2214 HE (4 Cores) 4GB RAM 4 x 146GB Disk-Drives (2 x RAID 1+0) CentOS 5.4 Stats: 05/06/2010 04:56:33 PM - SYSTEM STATS: Time:92.0414 Method:spine Processes:5 Threads:10 Hosts:2199 HostsPerProcess:440 DataSources:23010 RRDsP...

G'day Howie, Today I was changing the header content-type to 'text/plain' in your 'viewimage' function. This allowed me to store the the output as a local .png file for comparison with the one created by the poller. Result was, that the original .png file in the 'output' folder has been created perf...

Was trying with 0.95a, 0.95b and 0.96test1, but still no luck.. Gives me the same problem on all of them. I'll restore my old 0.941 for now and dig some more tomorrow... :-) I'm sure it must be something silly, as the whole map building process works like a charm... Just displaying them is not worki...

nope.. The change from dir to chdir did not solve it... When I check the .png file that has been created, everything looks fine.. Custom fonts are being used and all the images have been found. I'm now going trough the code and add some cacti_log calls for debugging.. Maybe the file gets corrupt in ...

I'm running it on Red Hat Enterprise Linux 5 with Apache and GD support in php.
Your check.php states, that all is ok when called from the webbrowser.
Calling it trough the cli gives me a notice about the missing PEAR module. But this can be ignored, as I use it with Cacti. Right?

Hello, Thanks for the quick response! :-) My output directory looks as follows.... [[email protected] output]# ll total 860 -rw-r--r-- 1 apache apache 13502 Sep 3 13:07 03ff50add5afb60cfb4d.html -rw-r--r-- 1 apache apache 18156 Sep 3 13:07 03ff50add5afb60cfb4d.png -rw-r--r-- 1 apache apache 5741 Sep 3 13:0...

G'day! Is there already a solution for this particular problem? My installation has got the same problem. Images are created with every poller cycle, but they're not shown (thumbs & maps). I'm currently running v0.941... Same issue on upgrading from 0.941 to 0.95a or 0.95b... Removing maps and re-ad...

Good day! Import the attached templates. This one gives you a graph showing the amount of DSL sessions: -> cacti_graph_template_erx_dsl_sessions.xml -> cacti_data_template_erx_router_dsl_sessions.xml And the following one is displaying the amount of L2TP sessions on the specified host: -> cacti_grap...

Hi Jay! Thanks for your reply. I think the problem is solved. The gaps appeared after traffic has reached 100000000bytes --> 762.939453125Mbit... This was the max value that has been set for the .rrd files and the cacti datasource by the "Interface Traffic" Data template. What I did was the followin...