Hi Lars,
try to increase the number of pollers if possible. You said that CPU is 70% idle. Another way would be checking which modules are activated. Sometimes you do not need all information i.e. FDB/ARP. On poller information site you can see which device and module takes x seconds to complete.
Mit freundlichem Gruß wilhelm.tel GmbH
Denis Klimek Carrier Manager & Professional Network Engineer IP-Systemtechnik Tel: +49 (0) 40 / 521 04 - 1049 -> Work From Home Mobil: +49 (0) 151 / 652 219 06
dklimek@stadtwerke-norderstedt.de www.wilhelm-tel.de
______________________________________________
[cid:image004.png@01D74C91.8B026D80]
Postanschrift: wilhelm.tel GmbH Heidbergstraße 101-111 22846 Norderstedt
Geschäftsführer: Jens Seedorff, Theo Weirich Vorsitzender des Aufsichtsrats: Christoph Mendel Handelsregister: HRB 4216 NO, Amtsgericht Kiel Umsatzsteuer ID: DE 81 299 7663
[ig]https://www.instagram.com/azubiteam/[fb]https://www.facebook.com/wilhelmtel.norderstedt/
Von: observium observium-bounces@observium.org Im Auftrag von Adam Armstrong via observium Gesendet: Dienstag, 18. Mai 2021 21:51 An: 'Observium' observium@observium.org Cc: Adam Armstrong adama@observium.org Betreff: Re: [Observium] Performance tuning
Just find some more/faster cores.
The per-core speed has an impact too because of MySQL.
Don't worry too much if a couple of very slow devices are sluggish and hang around over the 300 seconds. You're wanting to spread the majority of devices over the 300 seconds.
Adam.
From: observium <observium-bounces@observium.orgmailto:observium-bounces@observium.org> On Behalf Of Lars Joergensen via observium Sent: 18 May 2021 20:32 To: Observium <observium@observium.orgmailto:observium@observium.org> Cc: Lars Joergensen <DKLARJ@chr-hansen.commailto:DKLARJ@chr-hansen.com> Subject: [Observium] Performance tuning
Hi
We have one Observium server looking at
* 476 devices (mainly switches, some firewalls, some WAN routers) * 3666 ports * 8692 sensors
The server has 6 cores (~70% idle, ~3% iowait) and 8 GB memory. We have configured 15 poller threads with max 6 allowed wrapper processes.
The server is struggling to keep the polling cycle below 300 seconds (averaging 376 secs).
Is this workload realistic for one server, and if so where do you suggest we tune? Or do we need to use a distributed setup?
Lars ________________________________ Disclaimer: This e-mail, including any attachments, is for the intended recipient only. If you have received this e-mail by mistake please notify the sender immediately by return e-mail and delete this e-mail and any attachments, without opening the attachments, from your system. Access, disclosure, copying, distribution or reliance on any part of this e-mail by anyone else is prohibited. This e-mail is confidential and may be legally privileged. Chr. Hansen does not represent and/or warrant that the information sent and/or received by or with this e-mail is correct and does not accept any liability for damages related thereto. https://www.chr-hansen.com/en/legal-notice
