Hi Adam,

                    I used 12 no. of threads with poller-wrapper.py script.

 

My VM is monitoring to approx. 180 device and having following resources:-

RAM = 8192 MB

CPU = 6 vCPU

 

But still it is giving gap or dropout in graphs.

Please help us.

 

 

 

………

With Best Regards,

Chaman Rathee

Mob. No:- +91-9560055816

 

 

 

 

 

 

From: observium [mailto:observium-bounces@observium.org] On Behalf Of Adam Armstrong
Sent: 14 November 2015 15:23
To: observium@observium.org
Subject: Re: [Observium] Hangs up on polling of device for long time..........!

 

The number of threads should be set to whatever your server can comfortable accommodate without slowing down the web ui.

 

It depends upon processors, ram and i/o. You can probably run 10-20 threads with SSD storage, and some number fewer with spinning disk i/o.

 

adam.

On 14/11/2015 09:50:55, Chaman Rathee <chaman.rathee@progression.com> wrote:

As I have deleted or disabled poller.php script from cron file.

But please describe ................. poller-wrapper.py script. Also explain threads that is by default 2.

I am monitoring 400 devices so please suggest no. of threads for me.


………
With Best Regards,
Chaman Rathee
Mob. No:- +91-9560055816



-----Original Message-----
From: Michael [mailto:observium@smarsz.com]
Sent: 14 November 2015 14:47
To: Observium Network Observation System; Chaman Rathee; 'Observium Network Observation System'
Cc: chmnrathee@gmail.com
Subject: Re: [Observium] Hangs up on polling of device for long time..........!

You have the standard poller AND the wrapper running every 5min.

Disable the standard poller.php.




On 14 November 2015 8:06:07 PM AEDT, Chaman Rathee wrote:
>I am also using poller-wrapper.py in crontab.
>Please check below snapshot for your reference.
>
>
>[cid:image001.png@01D11EE9.CC6C5500]
>
>
>………
>With Best Regards,
>Chaman Rathee
>Mob. No:- +91-9560055816
>
>
>
>From: observium [mailto:observium-bounces@observium.org] On Behalf Of
>Tom Laermans
>Sent: 14 November 2015 14:20
>To: Observium Network Observation System
>Subject: Re: [Observium] Hangs up on polling of device for long
>time..........!
>
>
>Or just use poller-wrapper obviously. Then one device acting up can't
>impact others.
>
>(As documented)
>
>Tom
>On Nov 14, 2015 9:32 AM, Adam Armstrong
><>> wrote:
>
>As I said last time, I think your server is simply too slow for the
>number of devices you're polling.
>
>Adam.
>
>Sent with AquaMail for Android
>http://www.aqua-mail.com
>
>On 14 November 2015 08:20:51 Chaman Rathee
><>>
>wrote:
>Hi Adam,
> Whenever any device is polled, it stores logs in observium.log file.
>And I checked in observium.log file and there is no logs of that
>timestamp when graph was missing.So, I observed that at that time my
>device was not polled that’s why no logs in observium.log file.
>
>
>………
>With Best Regards,
>Chaman Rathee
>Mob. No:- +91-9560055816
>
>
>From: observium [mailto:observium-bounces@observium.org] On Behalf Of
>Adam Armstrong
>Sent: 14 November 2015 13:44
>To: Observium Network Observation System
>Subject: Re: [Observium] Hangs up on polling of device for long
>time..........!
>
>
>What makes you think that not all of your servers are polled?
>
>Adam.
>
>Sent with AquaMail for Android
>http://www.aqua-mail.com
>
>On 14 November 2015 07:53:50 Chaman Rathee
><>>
>wrote:
>Hi Team,
>When I checked in observium.log file, I found that many time my all
>observium devices are not polled. ( Few were polled.)
> Due to which graph of unpolled device gives gap or dropouts.
>
>Please help us.
>
>Whereas my cron process runs always after 5 minutes interval.
>
>
>………
>With Best Regards,
>Chaman Rathee
>Mob. No:- +91-9560055816
>
>
>
>
>From: observium [mailto:observium-bounces@observium.org] On Behalf Of
>Adam Armstrong
>Sent: 10 November 2015 12:52
>To: Glyn Parker;
>observium@observium.org
>Subject: Re: [Observium] Hangs up on polling of device for long
>time..........!
>
>The poller wrapper doesn't have this issue. It'll run x number of
>pollers in parallel, if one hangs, it won't hold the rest up.
>
>You could see which query is hanging by running the poller for that
>device in debug mode: ./poller.php -h -d
>
>adam.
>
>On 10/11/2015 07:14:20, Glyn Parker
><>> wrote:
>
>
>
>
>I had a device that seemed to "hang", but actually it was the snmp
>timeout that was too short. If you manually run smnpwalk on the unit
>until it "hangs", then use 5s timeout on that specific OID to see if
>its this that is your problem.
>
>Worth a try before giving up.
>
>On 10-11-2015 08:10, Chaman Rathee wrote:
>Ok. But it should not be stuck here. He should move to next device for
>polling.
>
>
>.........
>With Best Regards,
>Chaman Rathee
>Mob. No:- +91-9560055816
>
>
>
>From: observium [mailto:observium-bounces@observium.org] On Behalf Of
>Adam Armstrong
>Sent: 10 November 2015 11:47
>To: observium@observium.org
>Subject: Re: [Observium] Hangs up on polling of device for long
>time..........!
>
>Since it's an unsupported UPS, it probably has a horribly broken SNMP
>stack.
>
>The solution is probably to not poll it :)
>
>adam.
>
>On 10/11/2015 06:00:23, Chaman Rathee
><>>
>wrote:
>Hi Team,
>
>During polling of devices, it hangs up on one device and not exceed for
>next device.
>
>Please suggest.
>
>
>
>
>.........
>With Best Regards,
>Chaman Rathee
>Mob. No:- +91-9560055816
>
>
>
>
>
>
>Privileged or confidential information may be contained in this
>message. If you are not the addressee indicated in this message (or
>responsible for delivery of the message to such person), please delete
>this message and kindly notify the sender by an emailed reply.
>Opinions, conclusions and other information in this message that do not
>relate to the official business of Progression and its associate
>entities shall be understood as neither given nor endorsed by them.
>
>
>
>Privileged or confidential information may be contained in this
>message. If you are not the addressee indicated in this message (or
>responsible for delivery of the message to such person), please delete
>this message and kindly notify the sender by an emailed reply.
>Opinions, conclusions and other information in this message that do not
>relate to the official business of Progression and its associate
>entities shall be understood as neither given nor endorsed by them.
>
>_______________________________________________
>observium mailing list
>observium@observium.org
>http://postman.memetic.org/cgi-bin/mailman/listinfo/observium
>
>
>
>[Fabrikam]
>
>Privileged or confidential information may be contained in this
>message. If you are not the addressee indicated in this message (or
>responsible for delivery of the message to such person), please delete
>this message and kindly notify the sender by an emailed reply.
>Opinions, conclusions and other information in this message that do not
>relate to the official business of Progression and its associate
>entities shall be understood as neither given nor endorsed by them.
>_______________________________________________
>observium mailing list
>observium@observium.org
>http:/​/​postman​.​memetic​.​org/​cgi-bin/​mailman/​listinfo/​observium
>
>
>
>
>
>[Fabrikam]
>
>Privileged or confidential information may be contained in this
>message. If you are not the addressee indicated in this message (or
>responsible for delivery of the message to such person), please delete
>this message and kindly notify the sender by an emailed reply.
>Opinions, conclusions and other information in this message that do not
>relate to the official business of Progression and its associate
>entities shall be understood as neither given nor endorsed by them.
>_______________________________________________
>observium mailing list
>observium@observium.org
>http:/​/​postman​.​memetic​.​org/​cgi-bin/​mailman/​listinfo/​observium
>
>
>
>
>
>
>[Fabrikam]
>
>Privileged or confidential information may be contained in this
>message. If you are not the addressee indicated in this message (or
>responsible for delivery of the message to such person), please delete
>this message and kindly notify the sender by an emailed reply.
>Opinions, conclusions and other information in this message that do not
>relate to the official business of Progression and its associate
>entities shall be understood as neither given nor endorsed by them.
>
>
>-----------------------------------------------------------------------
>-
>
>_______________________________________________
>observium mailing list
>observium@observium.org
>http://postman.memetic.org/cgi-bin/mailman/listinfo/observium






[Fabrikam]

Privileged or confidential information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), please delete this message and kindly notify the sender by an emailed reply. Opinions, conclusions and other information in this message that do not relate to the official business of Progression and its associate entities shall be understood as neither given nor endorsed by them.
_______________________________________________
observium mailing list
observium@observium.org
http://postman.memetic.org/cgi-bin/mailman/listinfo/observium





 
Fabrikam

Privileged or confidential information may be contained in this message. If you are not the addressee indicated in this message (or responsible for delivery of the message to such person), please delete this message and kindly notify the sender by an emailed reply. Opinions, conclusions and other information in this message that do not relate to the official business of Progression and its associate entities shall be understood as neither given nor endorsed by them.