I monitor lots or access points over WAN and LAN connections and mostly its OK. There is a problem with one of our public access controllers (I'll call it a PAC now on, but its a router if you don't know what a PAC really does.) that we use in hotels. There are portmaps to the APs in the hotels to port 80 and I use the advanced ping to monitor the APs. The PAC has bandwidth limiting for users and when I turn it off, everything is pretty much fine. I attached what the graph looks like.
Since we use other PACs and this doesn't happen with them, then I'm guessing its just the PAC and nothing to really do with cacti, but maybe someone has an idea. I attached what the graph looks like. When the peak is reached the threshold goes off because its like it stopped responding. The high threshold is 1000 and low is 0.
The PAC is a Zyxel G4100 with latest firmware and it doesn't matter what AP model it is. Oh, and nothing ever actually goes down.
Weird up/down problem monitoring WAN portmaps
Moderators: Developers, Moderators
Weird up/down problem monitoring WAN portmaps
- Attachments
-
- sleepap.png (33.65 KiB) Viewed 2899 times
[size=75]Solaris 10 x86 [color=red][b]||[/b][/color] Cacti - 0.8.7 [color=red][b]||[/b][/color] MySQL - 5.0.45 [color=red][b]||[/b][/color] PHP - 5.2.6 [color=red][b]||[/b][/color] RRDTool - 1.2.23 [color=red][b]||[/b][/color] NET-SNMP - 5.4.1 [color=red][b]||[/b][/color] Syslog-ng 2.0.5
[url=http://www.indigo-networks.com]indigo-networks.com[/url][/size]
[url=http://www.indigo-networks.com]indigo-networks.com[/url][/size]
Without a drawing this is harder, but here are my thoughts / questions?
Are you also monitoring the ping time to the PAC?
Does the PING response to the PAC go up as well?
Does the PAC have SNMP capability and are you / can you monitor CPU time on the PAC?
Is it always the same PAC and the same AP?
Are you monitoring interface statistics on the PAC and if so do you see any errors on the interfaces in use that could indicate a bad cable or speed or duplex mismatch?
This is tough without a drawing...
I take it the PAC provides the port 80 mapping to your AP?There are portmaps to the APs in the hotels to port 80
Are you also monitoring the ping time to the PAC?
Does the PING response to the PAC go up as well?
Does the PAC have SNMP capability and are you / can you monitor CPU time on the PAC?
Is it always the same PAC and the same AP?
Are you monitoring interface statistics on the PAC and if so do you see any errors on the interfaces in use that could indicate a bad cable or speed or duplex mismatch?
This is tough without a drawing...
Thank you,
Ernie
http://www.NMSWorld.com
[b]Dual Zeon Dual Core 2.6Ghz / 8GB RAM / 4x15k RPM SATA RAID5[/b]
[b]Cacti Version[/b] - 0.8.7b
[b]Poller Type[/b] - cactid 0.8.7 with Boost v1.7
[b]Server Info[/b] - FreeBSD 7.0-RELEASE
[b]Web Server[/b] - Apache/2.2.6 (Unix) mod_ssl/2.2.6 OpenSSL/0.9.8g DAV/2 PHP/5.2.5 mod_perl/2.0.3 Perl/v5.8.8
[b]PHP[/b] - 5.2.6
[b]MySQL[/b] - 5.0.51b Mod: poller_output ENGINE = MEMORY
[b]RRDTool[/b] - 1.3.0
[b]SNMP[/b] - 5.4.1
[b]Plugins[/b] - Host Info (hostinfo - v0.2), Update Checker (update - v0.3), Network Tools (tools - v0.2), FlowView (flowview - v0.3), Read-only Devices Tab (devices - v0.4), Network Discovery (discovery - v0.8.3), Syslog Monitoring (syslog - v0.5.2), Thresholds (thold - v0.3.9), Device Monitoring (monitor - v0.8.2), PHP Network Weathermap (weathermap - v0.941), SuperLinks (superlinks - v0.72), Report Creator (reports - v0.1b)
Ernie
http://www.NMSWorld.com
[b]Dual Zeon Dual Core 2.6Ghz / 8GB RAM / 4x15k RPM SATA RAID5[/b]
[b]Cacti Version[/b] - 0.8.7b
[b]Poller Type[/b] - cactid 0.8.7 with Boost v1.7
[b]Server Info[/b] - FreeBSD 7.0-RELEASE
[b]Web Server[/b] - Apache/2.2.6 (Unix) mod_ssl/2.2.6 OpenSSL/0.9.8g DAV/2 PHP/5.2.5 mod_perl/2.0.3 Perl/v5.8.8
[b]PHP[/b] - 5.2.6
[b]MySQL[/b] - 5.0.51b Mod: poller_output ENGINE = MEMORY
[b]RRDTool[/b] - 1.3.0
[b]SNMP[/b] - 5.4.1
[b]Plugins[/b] - Host Info (hostinfo - v0.2), Update Checker (update - v0.3), Network Tools (tools - v0.2), FlowView (flowview - v0.3), Read-only Devices Tab (devices - v0.4), Network Discovery (discovery - v0.8.3), Syslog Monitoring (syslog - v0.5.2), Thresholds (thold - v0.3.9), Device Monitoring (monitor - v0.8.2), PHP Network Weathermap (weathermap - v0.941), SuperLinks (superlinks - v0.72), Report Creator (reports - v0.1b)
Yes, the PAC doesN3NCY wrote: I take it the PAC provides the port 80 mapping to your AP?
Yes, I am monitoring that too, but it doesn't go up and down like the APs.N3NCY wrote: Are you also monitoring the ping time to the PAC?
Does the PING response to the PAC go up as well?
No, the PAC doesn't have any SNMP, I really wish it did. We are fading these PACs out because of this up/down problem and no SNMP. It is every PAC of this model and any AP behind the PAC, doesn't matter what model. I have like 20 graphs like the one I posted above.N3NCY wrote: Does the PAC have SNMP capability and are you / can you monitor CPU time on the PAC?
Is it always the same PAC and the same AP?
Are you monitoring interface statistics on the PAC and if so do you see any errors on the interfaces in use that could indicate a bad cable or speed or duplex mismatch?
My boss is currently talking to some engineers at Zyxel to see if they can maybe release some new firmware. But we'll see.
[size=75]Solaris 10 x86 [color=red][b]||[/b][/color] Cacti - 0.8.7 [color=red][b]||[/b][/color] MySQL - 5.0.45 [color=red][b]||[/b][/color] PHP - 5.2.6 [color=red][b]||[/b][/color] RRDTool - 1.2.23 [color=red][b]||[/b][/color] NET-SNMP - 5.4.1 [color=red][b]||[/b][/color] Syslog-ng 2.0.5
[url=http://www.indigo-networks.com]indigo-networks.com[/url][/size]
[url=http://www.indigo-networks.com]indigo-networks.com[/url][/size]
Who is online
Users browsing this forum: No registered users and 1 guest