problem in graphics of more than 1000 devices
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
It seems that your server is probably not fast enough.
You may be able to fix this somewhat by upgrading to the latest version and using php7.2 on Ubuntu 18.04 or something else very new.
Adam.
Sent from BlueMail
On 13 Sep 2018, 20:52, at 20:52, "Wilbert Martínez López" wilbert@inetmexico.mx wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Yes, I/O can also be an issue.
This can be partially mitigated by using rrdcached.
You can also disable modules that you don't really need, like FDB collection.
Thanks, Adam.
Sent from BlueMail
On 13 Sep 2018, 21:52, at 21:52, "Gray, Thomas" tgray@infonline.de wrote:
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi,
check your polling times. Most polling issues are sourced there…
Von: observium [mailto:observium-bounces@observium.org] Im Auftrag von Adam Armstrong Gesendet: Donnerstag, 13. September 2018 23:46 An: Observium observium@observium.org Betreff: Re: [Observium] problem in graphics of more than 1000 devices
Yes, I/O can also be an issue. This can be partially mitigated by using rrdcached. You can also disable modules that you don't really need, like FDB collection. Thanks, Adam. Sent from BlueMailhttp://www.bluemail.me/r?b=13569 On 13 Sep 2018, at 21:52, "Gray, Thomas" <tgray@infonline.demailto:tgray@infonline.de> wrote:
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
________________________________
observium mailing list observium@observium.orgmailto:observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
________________________________
observium mailing list observium@observium.orgmailto:observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi,
Derived from the first recommendation, update to ubuntu 18.04.01, also update to the latest version of Observium 18.9.9420, also update my version of php to 7.2 and after all that the behavior is the same.
2018-09-14 4:20 GMT-05:00 Klimek, Denis DKlimek@stadtwerke-norderstedt.de:
Hi,
check your polling times. Most polling issues are sourced there…
*Von:* observium [mailto:observium-bounces@observium.org] *Im Auftrag von *Adam Armstrong *Gesendet:* Donnerstag, 13. September 2018 23:46 *An:* Observium observium@observium.org *Betreff:* Re: [Observium] problem in graphics of more than 1000 devices
Yes, I/O can also be an issue.
This can be partially mitigated by using rrdcached.
You can also disable modules that you don't really need, like FDB collection.
Thanks,
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569
On 13 Sep 2018, at 21:52, "Gray, Thomas" tgray@infonline.de wrote:
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
This is probably I/O related then.
Your system isn't fast enough, you need either more CPU, more storage throughout or both.
Adam.
Sent from BlueMail
On 14 Sep 2018, 15:56, at 15:56, "Wilbert Martínez López" wilbert@inetmexico.mx wrote:
Hi,
Derived from the first recommendation, update to ubuntu 18.04.01, also update to the latest version of Observium 18.9.9420, also update my version of php to 7.2 and after all that the behavior is the same.
2018-09-14 4:20 GMT-05:00 Klimek, Denis DKlimek@stadtwerke-norderstedt.de:
Hi,
check your polling times. Most polling issues are sourced there…
*Von:* observium [mailto:observium-bounces@observium.org] *Im Auftrag
von
*Adam Armstrong *Gesendet:* Donnerstag, 13. September 2018 23:46 *An:* Observium observium@observium.org *Betreff:* Re: [Observium] problem in graphics of more than 1000
devices
Yes, I/O can also be an issue.
This can be partially mitigated by using rrdcached.
You can also disable modules that you don't really need, like FDB collection.
Thanks,
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569
On 13 Sep 2018, at 21:52, "Gray, Thomas" tgray@infonline.de wrote:
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with
SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard
drive
of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti
and
mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
What would be your recommendation the cpu, memory, hard disk since it is estimated to have up to 1500 devices?
Wilbert.
2018-09-14 10:19 GMT-05:00 Adam Armstrong adama@memetic.org:
This is probably I/O related then.
Your system isn't fast enough, you need either more CPU, more storage throughout or both.
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569 On 14 Sep 2018, at 15:56, "Wilbert Martínez López" wilbert@inetmexico.mx wrote:
Hi,
Derived from the first recommendation, update to ubuntu 18.04.01, also update to the latest version of Observium 18.9.9420, also update my version of php to 7.2 and after all that the behavior is the same.
2018-09-14 4:20 GMT-05:00 Klimek, Denis <DKlimek@stadtwerke- norderstedt.de>:
Hi,
check your polling times. Most polling issues are sourced there…
*Von:* observium [mailto:observium-bounces@observium.org] *Im Auftrag von *Adam Armstrong *Gesendet:* Donnerstag, 13. September 2018 23:46 *An:* Observium observium@observium.org *Betreff:* Re: [Observium] problem in graphics of more than 1000 devices
Yes, I/O can also be an issue.
This can be partially mitigated by using rrdcached.
You can also disable modules that you don't really need, like FDB collection.
Thanks,
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569
On 13 Sep 2018, at 21:52, "Gray, Thomas" tgray@infonline.de wrote:
Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs.
Thomas
On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:
hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics
graph a month from a port
graph to 6 hours a port
the same situation occurs with the graphics of memory, processor, memory or any ports of all devices
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hi Wilbert
I have a 4 Core with HT Xeon E3 1270 CPU @ 3.5Ghz 32GB Ram
Then 3 Raid 1 Arrays on Fast SSD Drives 1 for System, 1 for MySQL and 1 for RRD
The RRD one has to be on a Enterprise SSD which can handle big amounts of writes (several 10GB/5min) else it will get broken very fast. To reduce writes on the SSD I also recommend using rrdcached.
Regards
Matthias
On 14/09/2018 17:26, Wilbert Martínez López wrote:
What would be your recommendation the cpu, memory, hard disk since it is estimated to have up to 1500 devices?
Wilbert.
2018-09-14 10:19 GMT-05:00 Adam Armstrong <adama@memetic.org mailto:adama@memetic.org>:
This is probably I/O related then. Your system isn't fast enough, you need either more CPU, more storage throughout or both. Adam. Sent from BlueMail <http://www.bluemail.me/r?b=13569> On 14 Sep 2018, at 15:56, "Wilbert Martínez López" <wilbert@inetmexico.mx <mailto:wilbert@inetmexico.mx>> wrote: Hi, Derived from the first recommendation, update to ubuntu 18.04.01, also update to the latest version of Observium 18.9.9420, also update my version of php to 7.2 and after all that the behavior is the same. 2018-09-14 4:20 GMT-05:00 Klimek, Denis <DKlimek@stadtwerke-norderstedt.de <mailto:DKlimek@stadtwerke-norderstedt.de>>: Hi,____ __ __ check your polling times. Most polling issues are sourced there…____ __ __ *Von:*observium [mailto:observium-bounces@observium.org <mailto:observium-bounces@observium.org>] *Im Auftrag von *Adam Armstrong *Gesendet:* Donnerstag, 13. September 2018 23:46 *An:* Observium <observium@observium.org <mailto:observium@observium.org>> *Betreff:* Re: [Observium] problem in graphics of more than 1000 devices____ __ __ Yes, I/O can also be an issue.____ This can be partially mitigated by using rrdcached.____ You can also disable modules that you don't really need, like FDB collection.____ Thanks,____ Adam.____ Sent from BlueMail <http://www.bluemail.me/r?b=13569> ____ On 13 Sep 2018, at 21:52, "Gray, Thomas" <tgray@infonline.de <mailto:tgray@infonline.de>> wrote:____ Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs. Thomas On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:____ hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics graph a month from a port graph to 6 hours a port the same situation occurs with the graphics of memory, processor, memory or any ports of all devices Wilbert. ____ ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>____ ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>____ _______________________________________________ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium> _______________________________________________ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
I think this written here more than 5 times.. why you still not done this?
Use RRDCACHED!
http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html
https://docs.observium.org/tuning/
Matthias Cramer wrote on 14/09/2018 18:34:
Hi Wilbert
I have a 4 Core with HT Xeon E3 1270 CPU @ 3.5Ghz 32GB Ram
Then 3 Raid 1 Arrays on Fast SSD Drives 1 for System, 1 for MySQL and 1 for RRD
The RRD one has to be on a Enterprise SSD which can handle big amounts of writes (several 10GB/5min) else it will get broken very fast. To reduce writes on the SSD I also recommend using rrdcached.
Regards
Matthias
On 14/09/2018 17:26, Wilbert Martínez López wrote:
What would be your recommendation the cpu, memory, hard disk since it is estimated to have up to 1500 devices?
Wilbert.
2018-09-14 10:19 GMT-05:00 Adam Armstrong <adama@memetic.org mailto:adama@memetic.org>:
This is probably I/O related then. Your system isn't fast enough, you need either more CPU, more storage throughout or both. Adam. Sent from BlueMail <http://www.bluemail.me/r?b=13569> On 14 Sep 2018, at 15:56, "Wilbert Martínez López" <wilbert@inetmexico.mx <mailto:wilbert@inetmexico.mx>> wrote: Hi, Derived from the first recommendation, update to ubuntu 18.04.01, also update to the latest version of Observium 18.9.9420, also update my version of php to 7.2 and after all that the behavior is the same. 2018-09-14 4:20 GMT-05:00 Klimek, Denis <DKlimek@stadtwerke-norderstedt.de <mailto:DKlimek@stadtwerke-norderstedt.de>>: Hi,____ __ __ check your polling times. Most polling issues are sourced there…____ __ __ *Von:*observium [mailto:observium-bounces@observium.org <mailto:observium-bounces@observium.org>] *Im Auftrag von *Adam Armstrong *Gesendet:* Donnerstag, 13. September 2018 23:46 *An:* Observium <observium@observium.org <mailto:observium@observium.org>> *Betreff:* Re: [Observium] problem in graphics of more than 1000 devices____ __ __ Yes, I/O can also be an issue.____ This can be partially mitigated by using rrdcached.____ You can also disable modules that you don't really need, like FDB collection.____ Thanks,____ Adam.____ Sent from BlueMail <http://www.bluemail.me/r?b=13569> ____ On 13 Sep 2018, at 21:52, "Gray, Thomas" <tgray@infonline.de <mailto:tgray@infonline.de>> wrote:____ Hi Wilbert, have a look at your I/O usage. Maybe your storage is too slow. I had similar problems and migrate from a VM to a physical server with SSDs. Thomas On Thu, 2018-09-13 at 14:52 -0500, Wilbert Martínez López wrote:____ hello can you please support me with the following situation I have, I installed Obervium in a virtual machine with 8 Gb of RAM and 8 cores, which have an Intel Xeon processor E5-1620 v3 at 3.50 Ghz, hard drive of 200 GB, has installed a version of Ubuntu 16.04.5 with Observium version CE17.9.0 when it had 30 devices added works well in terms of graphics, but when we added more antennas in total 1085 at the time it has become slow and also the graphics shows incomplete of any element, cpu, memory, ports, etc, the devices aggregates are radios of ubiquiti and mikrotik, I attach the images where it shows the commented of the graphics graph a month from a port graph to 6 hours a port the same situation occurs with the graphics of memory, processor, memory or any ports of all devices Wilbert. ____ ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>____ ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>____ _______________________________________________ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium> ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium> _______________________________________________ observium mailing list observium@observium.org <mailto:observium@observium.org> http://postman.memetic.org/cgi-bin/mailman/listinfo/observium <http://postman.memetic.org/cgi-bin/mailman/listinfo/observium>
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
This is probably because you're polling through a firewall which has session or other rate limits.
Adam.
Sent from BlueMail
On 16 Sep 2018, 01:14, at 01:14, Eduardo Silvestre eduardo@ptisp.pt wrote:
Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp
timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them
are with this behavior.
Any idea?
Best Regards,
--
*Melhores Cumprimentos | Best Regards,* *—* Eduardo Silvestre ptisp | wherever internet can take you…
eduardo@ptisp.pt mailto:eduardo@ptisp.pt helpdesk: +351 249 739 099 | mobile: +351 91 536 46 43 www.ptisp.pt http://www.ptisp.pt
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hello Adam,
this switch and server don't have firewall. This "issue" start happening after upgrade to last CE version.
Best Regards,
Adam Armstrong mailto:adama@memetic.org 16 September 2018 at 01:19 This is probably because you're polling through a firewall which has session or other rate limits.
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569 _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Eduardo Silvestre mailto:eduardo@ptisp.pt 16 September 2018 at 00:02 Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Mike Stupalov mailto:mike@observium.org 15 September 2018 at 11:20 I think this written here more than 5 times.. why you still not done this?
Use RRDCACHED!
http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html
https://docs.observium.org/tuning/
Matthias Cramer wrote on 14/09/2018 18:34:
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Matthias Cramer mailto:matthias.cramer@iway.ch 14 September 2018 at 16:34 Hi Wilbert
I have a 4 Core with HT Xeon E3 1270 CPU @ 3.5Ghz 32GB Ram
Then 3 Raid 1 Arrays on Fast SSD Drives 1 for System, 1 for MySQL and 1 for RRD
The RRD one has to be on a Enterprise SSD which can handle big amounts of writes (several 10GB/5min) else it will get broken very fast. To reduce writes on the SSD I also recommend using rrdcached.
Regards
Matthias
Wilbert Martínez López mailto:wilbert@inetmexico.mx 14 September 2018 at 16:26 What would be your recommendation the cpu, memory, hard disk since it is estimated to have up to 1500 devices?
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
You get what you pay for :)
Adam.
Sent from BlueMail
On 16 Sep 2018, 01:45, at 01:45, Eduardo Silvestre eduardo@ptisp.pt wrote:
Hello Adam,
this switch and server don't have firewall. This "issue" start happening after upgrade to last CE version.
Best Regards,
Adam Armstrong mailto:adama@memetic.org 16 September 2018 at 01:19 This is probably because you're polling through a firewall which has session or other rate limits.
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569 _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Eduardo Silvestre mailto:eduardo@ptisp.pt 16 September 2018 at 00:02 Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be
fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works
fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M
/opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Mike Stupalov mailto:mike@observium.org 15 September 2018 at 11:20 I think this written here more than 5 times.. why you still not done
this?
Use RRDCACHED!
http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html
https://docs.observium.org/tuning/
Matthias Cramer wrote on 14/09/2018 18:34:
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Matthias Cramer mailto:matthias.cramer@iway.ch 14 September 2018 at 16:34 Hi Wilbert
I have a 4 Core with HT Xeon E3 1270 CPU @ 3.5Ghz 32GB Ram
Then 3 Raid 1 Arrays on Fast SSD Drives 1 for System, 1 for MySQL and 1 for RRD
The RRD one has to be on a Enterprise SSD which can handle big
amounts
of writes (several 10GB/5min) else it will get broken very fast. To reduce writes on the SSD I also
recommend using rrdcached.
Regards
Matthias
Wilbert Martínez López mailto:wilbert@inetmexico.mx 14 September 2018 at 16:26 What would be your recommendation the cpu, memory, hard disk since it
is estimated to have up to 1500 devices?
Wilbert.
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
--
*Melhores Cumprimentos | Best Regards,* *—* Eduardo Silvestre ptisp | wherever internet can take you…
eduardo@ptisp.pt mailto:eduardo@ptisp.pt helpdesk: +351 249 739 099 | mobile: +351 91 536 46 43 www.ptisp.pt http://www.ptisp.pt
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hello Adam,
fair enough. in the paid version this issue is fixed?
Best Regards,
Adam Armstrong mailto:adama@memetic.org 16 September 2018 at 01:47 You get what you pay for :)
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569 _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Eduardo Silvestre mailto:eduardo@ptisp.pt 16 September 2018 at 01:46 Hello Adam,
this switch and server don't have firewall. This "issue" start happening after upgrade to last CE version.
Best Regards,
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Adam Armstrong mailto:adama@memetic.org 16 September 2018 at 01:19 This is probably because you're polling through a firewall which has session or other rate limits.
Adam.
Sent from BlueMail http://www.bluemail.me/r?b=13569 _______________________________________________ observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Eduardo Silvestre mailto:eduardo@ptisp.pt 16 September 2018 at 00:02 Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium Mike Stupalov mailto:mike@observium.org 15 September 2018 at 11:20 I think this written here more than 5 times.. why you still not done this?
Use RRDCACHED!
http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html
https://docs.observium.org/tuning/
Matthias Cramer wrote on 14/09/2018 18:34:
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
I don't think any version will fix that your devices sometimes fail to respond.
The particular query you posted is for ifXEntry. We've been collecting this table by BULKWALK the same way for about 9 years, with the addition of max-rep by default a few years ago.
This is most likely an issue with your devices.
adam. On 2018-09-16 01:53:27, Eduardo Silvestre eduardo@ptisp.pt wrote: Hello Adam,
fair enough. in the paid version this issue is fixed?
Best Regards,
Adam Armstrong [mailto:adama@memetic.org] 16 September 2018 at 01:47 You get what you pay for :)
Adam.
Sent from BlueMail [http://www.bluemail.me/r?b=13569] _______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]
Eduardo Silvestre [mailto:eduardo@ptisp.pt] 16 September 2018 at 01:46 Hello Adam,
this switch and server don't have firewall. This "issue" start happening after upgrade to last CE version.
Best Regards,
_______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]
Adam Armstrong [mailto:adama@memetic.org] 16 September 2018 at 01:19 This is probably because you're polling through a firewall which has session or other rate limits.
Adam.
Sent from BlueMail [http://www.bluemail.me/r?b=13569] _______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]
Eduardo Silvestre [mailto:eduardo@ptisp.pt] 16 September 2018 at 00:02 Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 [udp:********:161] ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
_______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]
Mike Stupalov [mailto:mike@observium.org] 15 September 2018 at 11:20 I think this written here more than 5 times.. why you still not done this?
Use RRDCACHED!
http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html [http://blog.best-practice.se/2014/10/using-rrdcached-with-observium.html]
https://docs.observium.org/tuning/ [https://docs.observium.org/tuning/]
Matthias Cramer wrote on 14/09/2018 18:34:
_______________________________________________ observium mailing list observium@observium.org [mailto:observium@observium.org] http://postman.memetic.org/cgi-bin/mailman/listinfo/observium [http://postman.memetic.org/cgi-bin/mailman/listinfo/observium]
--
Melhores Cumprimentos | Best Regards, — Eduardo Silvestre
ptisp | wherever internet can take you…
eduardo@ptisp.pt [mailto:eduardo@ptisp.pt] helpdesk: +351 249 739 099 | mobile: +351 91 536 46 43
www.ptisp.pt [http://www.ptisp.pt]
This is only device/network trouble. (Not configurations/observium/etc)
Try run manually same command and small modifications:
(same) snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry (with timeout 3 sec) snmpbulkwalk -t 3 -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry (set minimal repetition) snmpbulkwalk -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry (no bulk) snmpwalk -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry (no bulk with timeout 3) snmpwalk -t 3 -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry
you can stop on first responded.
Eduardo Silvestre wrote on 16/09/2018 02:02:
Hello Mike,
I've tried to setup observium with rrdcache but the polling stop working after apply configuration. However, permissions seems to be fine.
Nevertheless, when i've running poller.php -d -h myswitch" and got snmp timeouts but when run same command through command line works fine.
CMD[/usr/bin/snmpbulkwalk -Cr'100' -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry]
CMD EXITCODE[1] CMD RUNTIME[2.0631s] STDOUT[
] STDERR[ Timeout: No Response from udp:********:161 ] SNMP STATUS[FALSE] SNMP ERROR[#1002 - Request timeout]
I've a lot of switch with same IOS and same module but just two of them are with this behavior.
Any idea?
Best Regards,
--
*Melhores Cumprimentos | Best Regards,* *—* Eduardo Silvestre ptisp | wherever internet can take you…
eduardo@ptisp.pt mailto:eduardo@ptisp.pt helpdesk: +351 249 739 099 | mobile: +351 91 536 46 43 www.ptisp.pt http://www.ptisp.pt
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hello Mike,
All comands that you send me works fine without any timeout or error.
Best Regards,
Mike Stupalov wrote:
snmpwalk -t 3 -v2c -c *** -Pu -OQUs -m IF-MIB -M /opt/observium/mibs/rfc:/opt/observium/mibs/net-snmp 'udp':'********':'161' ifXEntry
Hello Mike,
All comands that you send me works fine without any timeout or error.
However, this seems to be related with size of ARP table in this switch.
Can I do anything to get around this?
Best Regards,
Mike Stupalov wrote:
Hello Mike,
All comands that you send me works fine without any timeout or error.
Best Regards,
Sure, go to the device properties->modules> find arp-table and click disable. This should disable ARP-tabel discovery for this device
/Markus
Den sön 16 sep. 2018 kl 17:43 skrev Eduardo Silvestre eduardo@ptisp.pt:
Hello Mike,
All comands that you send me works fine without any timeout or error.
However, this seems to be related with size of ARP table in this switch.
Can I do anything to get around this?
Best Regards,
Mike Stupalov wrote:
Hello Mike,
All comands that you send me works fine without any timeout or error.
Best Regards,
--
*Melhores Cumprimentos | Best Regards,* *—* Eduardo Silvestre ptisp | wherever internet can take you…
eduardo@ptisp.pt helpdesk: +351 249 739 099 | mobile: +351 91 536 46 43 www.ptisp.pt
observium mailing list observium@observium.org http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
Hello
We have in our company IceWarp as Mailserver.
This solution has a reduced OID Infomration, take a look here. https://esupport.icewarp.com/index.php?/Knowledgebase/Article/View/180/16/sn...
My problem is that I'm not able to add this host as a device, I think the add process need some other OID to recognize IceWarp as SNMP Device. If I could add them I would us the custom id to make something visible.
Am I right assuming, that in the current Observium Version this is not possible?
Regards Walter
participants (9)
-
Adam Armstrong -
Eduardo Silvestre -
Gray, Thomas -
Klimek, Denis -
Markus Klock -
Matthias Cramer -
Mike Stupalov -
walter＠onda.ch -
Wilbert Martínez López