Help! Keep getting WARNING: Result from SERVER not valid

[DazedandConfused]

On several W2k3 servers that I'm trying to monitor I keep getting gaps, well actually its more like I get a hit once and a while and the rest of the graph is blank. I believe that the error below is the reason, thought I'm not sure what exactly is causing it. I'm running cacti 0.8.6g, on CentOS 4.1 and I've already replaced the 4 files mentioned from http://forums.cacti.net/about9352.html, yet I continue to get no data on these graphs.

10/10/2005 09:55:09 AM - CMDPHP: Poller[0] Host[31] DS[127] WARNING: Result from SERVER not valid. Partial Result: U
10/10/2005 09:55:09 AM - CMDPHP: Poller[0] Host[31] DS[127] SERVER: /var/www/cacti/scripts/ss_host_disk.php ss_host_disk 10.100.1.4 public 2 31 161 600 get used 8, output: U

Thanks for the assistance.
Dazed

[TheWitness]

You need to "clear your poller cache". This option is located in "System Utilities".

Once your are done, look for "Action: 2" in the cache. Your syntax for the Script Server stuff should change. You should then see a series of text separated with the colon character ":" instead of spaces in the calling parameters.

TheWitness

[TheWitness]

Make sure you apply the files in the attached link.

Larry

http://www.cacti.net/downloads/patches/ ... e-patched/

[DazedandConfused]

I cleared the poller cache, or at elast I think I did, it doesn't seem to do anything but display the poller cache when I click on clear poller cache, this is an excerpt from "Action: 2"

Data Source: MHSNOOSD002 - Used Space - Virtual Memory
RRD: /var/www/cacti/rra/mhsnoosd002_hdd_total_90.rrd
Action: 2, Script: /var/www/cacti/scripts/ss_host_disk.php ss_host_disk 10.1.1.4 public 2 28 161 600 get total 5

As for the the files at http://www.cacti.net/downloads/patches/ ... e-patched/, I assume all I do is download them and replace the existing files with those? if that is the case I've already done that as well, and appear to be getting the same error still.

Thanks
Dazed

[TheWitness]

Your XML file under <path_cacti>/resource/script_server has not been updated or your clear poller cache is not working. Please double check those XML files.

TheWitness

[TheWitness]

Please look at this post. It will solve your problem. Remember to clear your poller cache.

http://forums.cacti.net/viewtopic.php?p=44322#44322

TheWitness

[DazedandConfused]

Thank you, that solved the one error, but more follow, sometimes I wonder why I chose this career path. Seriously though, as far as comminicating with my Cisco Switches and PIX, it seems to be functioning just fine. The handful of W2k3 servers that I'm trying to monitor are causing problems. I've attached the output of doing >php poller.php, it appears that when it times out when trying to connect to my W2k3 servers, and then after it finishes running the poller.php, it throws back two errors.

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.1

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.18

Also, in the cacti.log file I'm now seeing some new warnings;

10/11/2005 07:20:17 AM - CMDPHP: Poller[0] ASSERT: '26875714<' failed. Recaching host '10.100.1.2', data query #1
10/11/2005 07:20:29 AM - SYSTEM STATS: Time:28.2634 Method:cmd.php Processes:1 Threads:N/A Hosts:26 HostsPerProcess:26 DataSources:176 RRDsProcessed:101
10/11/2005 07:20:30 AM - POLLER: Poller[0] Host[24] WARNING: Recache Event Detected for Host
10/11/2005 07:20:30 AM - POLLER: Poller[0] Host[29] WARNING: Recache Event Detected for Host
10/11/2005 07:20:31 AM - POLLER: Poller[0] Host[30] WARNING: Recache Event Detected for Host
10/11/2005 07:20:32 AM - POLLER: Poller[0] Host[31] WARNING: Recache Event Detected for Host
10/11/2005 07:20:33 AM - RECACHE STATS: RecacheTime:3.5169 HostsRecached:4
10/11/2005 07:25:22 AM - CMDPHP: Poller[0] ASSERT: '26528009<' failed. Recaching host '10.100.1.4', data query #1
10/11/2005 07:25:23 AM - SYSTEM STATS: Time:21.2472 Method:cmd.php Processes:1 Threads:N/A Hosts:26 HostsPerProcess:26 DataSources:176 RRDsProcessed:48
10/11/2005 07:25:23 AM - POLLER: Poller[0] Host[31] WARNING: Recache Event Detected for Host
10/11/2005 07:25:26 AM - RECACHE STATS: RecacheTime:2.7621 HostsRecached:1
10/11/2005 07:29:44 AM - CMDPHP: Poller[0] ASSERT: '26935919<' failed. Recaching host '10.100.1.2', data query #1
10/11/2005 07:29:51 AM - CMDPHP: Poller[0] ASSERT: '26755142<' failed. Recaching host '10.100.1.3', data query #1
10/11/2005 07:29:58 AM - CMDPHP: Poller[0] ASSERT: '26557270<' failed. Recaching host '10.100.1.4', data query #1
10/11/2005 07:30:02 AM - SYSTEM STATS: Time:1.0814 Method:cmd.php Processes:1 Threads:N/A Hosts:26 HostsPerProcess:26 DataSources:176 RRDsProcessed:33
10/11/2005 07:30:03 AM - POLLER: Poller[0] Host[29] WARNING: Recache Event Detected for Host
10/11/2005 07:30:03 AM - SYSTEM STATS: Time:30.2274 Method:cmd.php Processes:1 Threads:N/A Hosts:26 HostsPerProcess:26 DataSources:176 RRDsProcessed:122
10/11/2005 07:30:03 AM - POLLER: Poller[0] Host[29] WARNING: Recache Event Detected for Host
10/11/2005 07:30:05 AM - POLLER: Poller[0] Host[30] WARNING: Recache Event Detected for Host
10/11/2005 07:30:05 AM - POLLER: Poller[0] Host[30] WARNING: Recache Event Detected for Host
10/11/2005 07:30:06 AM - POLLER: Poller[0] Host[31] WARNING: Recache Event Detected for Host
10/11/2005 07:30:07 AM - POLLER: Poller[0] Host[31] WARNING: Recache Event Detected for Host
10/11/2005 07:30:08 AM - RECACHE STATS: RecacheTime:5.6634 HostsRecached:3
10/11/2005 07:30:09 AM - RECACHE STATS: RecacheTime:5.3075 HostsRecached:3
10/11/2005 07:35:02 AM - CMDPHP: Poller[0] ASSERT: '59501571<59501571' failed. Recaching host '10.100.0.1', data query #1
10/11/2005 07:35:20 AM - CMDPHP: Poller[0] Host[31] NOTICE: HOST EVENT: Host Returned from DOWN State:
10/11/2005 07:39:59 AM - POLLER: Poller[0] Maximum runtime of 296 seconds exceeded. Exiting.
10/11/2005 07:40:17 AM - CMDPHP: Poller[0] ASSERT: '26996053<' failed. Recaching host '10.100.1.2', data query #1
10/11/2005 07:40:30 AM - SYSTEM STATS: Time:28.3427 Method:cmd.php Processes:1 Threads:N/A Hosts:26 HostsPerProcess:26 DataSources:176 RRDsProcessed:96
10/11/2005 07:40:30 AM - POLLER: Poller[0] Host[24] WARNING: Recache Event Detected for Host
10/11/2005 07:40:31 AM - POLLER: Poller[0] Host[29] WARNING: Recache Event Detected for Host
10/11/2005 07:40:32 AM - POLLER: Poller[0] Host[2] WARNING: Recache Event Detected for Host
10/11/2005 07:40:34 AM - POLLER: Poller[0] Host[30] WARNING: Recache Event Detected for Host
10/11/2005 07:40:35 AM - POLLER: Poller[0] Host[31] WARNING: Recache Event Detected for Host
10/11/2005 07:40:36 AM - RECACHE STATS: RecacheTime:5.7303 HostsRecached:5


Let me know if you have any further thoughts on this.

[TheWitness]

This is not good:

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.1

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.18

This means that your sysUptime is not increasing. I suspect an SNMP Agent issue:

10/11/2005 07:35:02 AM - CMDPHP: Poller[0] ASSERT: '59501571<59501571' failed. Recaching host '10.100.0.1', data query #1

I would definately check with your hardware vendor for problems with sysUptime and general agent issues.

TheWitness

[DazedandConfused]

How do I determin which devices are causing these errors?

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.1

Error in packet.
Reason: (genError) A general failure occured
Failed object: .1.3.6.1.2.1.31.1.1.1.18

As for these three lines
10/11/2005 07:29:44 AM - CMDPHP: Poller[0] ASSERT: '26935919<' failed. Recaching host '10.100.1.2', data query #1
10/11/2005 07:29:51 AM - CMDPHP: Poller[0] ASSERT: '26755142<' failed. Recaching host '10.100.1.3', data query #1
10/11/2005 07:29:58 AM - CMDPHP: Poller[0] ASSERT: '26557270<' failed. Recaching host '10.100.1.4', data query #1

These are W2k3 servers, running both the standard W2k3 SNMP service as well as SNMP Informant Std. The graphs for these three servers seem to update sporadically, and varies depending on the resource being graphed. Disks are almost never graphed, CPU's absolutely nothing at all, and the host mib traffic, prcesseses, and users hits about 50% of the time. As for gaps in my graphs, I'm seeing that on almost every device I'm monitoring, I'm assuming the errors I'm getting about the "CMDPHP: Poller[0] ASSERT: '26557270<' failed." is whats causing my servers graphs to show up with no data, but I'm also seeing this on all my Cisco devices as well.

Roland

[TheWitness]

Pick any remote host (better if > T1 WAN) and run the following command:

ping -l 1400 -t xx.xx.xx.xx

Where xx.xx.xx.xx is the remote IP address of the host. If you are loosing packets, you know that:

1) Duplex mismatch on your server
2) Poor Cabling/Termination Somewhere (Start at the Polling Server)
3) Defective LAN Card

These problems are prioritized from most likely to least likely with 1 and 2 being pretty much interchangable.

TheWitness

[TheWitness]

Or better yet, ping one of the problem servers. Also, there is an excellent tool for diagnosing bottlenecks/cable issues in the layer3 world:

http://www.pingplotter.com

Cheap and dependable.

TheWitness

[DazedandConfused]

Ok, I ran pingplotter from my desktop (10.100.101.52/24), testing both the cacti server(10.100.1.40/24), and one of the W2k3 servers (10.100.1.4/24) that I'm having problems monitoring. I ran it for 112 minutes, looping it's test every 15 seconds, and didn't have a single failure, error, etc. Also while running pingplotter from my desktop, i was pinging both 10.100.1.4, and 10.100.0.1 from the cacti server, there again, no lost packets, and on average, the response times were roughly .300 ms.

As far as

Pick any remote host (better if > T1 WAN) and run the following command:

I haven't even begun to monitor stuff across my WAN yet, everything I'm trying to monitor is at worst on my local LAN here, or in the case of the servers, PIX, and one of the switches, they're all on the same switch with the cacti server. My W2k3 servers, and the cacti server are all connected on gigabit ports, so there really are no slow links at this point that I'm trying to monitor, and I'm still showing quite a few gaps in my graphs. Also, I no longer appear to be receiving any errors in either the cacti.log, or when i run >php poller.php.

To be completely honest, I'm at a total loss as to what the problem could be. If you don't mind me emailing, or pming you, I'll be glad to send you a link & login to my cacti server so that you can take a look for yourself.

Thanks once again for all the assistance you've provided me, I must say, thats not something I'm used to. Thanks again.

Roland