Ok, Tom, you asked.... ;-)
This is during a regular poll cycle. I'm using poller-wrapper to fire up 32 polling instances as it takes slightly under three minutes to complete the polling process. I'm polling 643 devices, 7375 ports, and 1320 sensors.
Tasks: 330 total, 13 running, 317 sleeping, 0 stopped, 0 zombie %Cpu(s): 39.3 us, 58.6 sy, 0.0 ni, 1.5 id, 0.1 wa, 0.0 hi, 0.6 si, 0.0 st KiB Mem: 12304220 total, 9841008 used, 2463212 free, 234880 buffers KiB Swap: 12580860 total, 0 used, 12580860 free. 8285488 cached Mem
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 1033 mysql 20 0 4510540 255200 11380 S 15.9 2.1 529:50.63 mysqld 4252 www-data 20 0 413696 55472 20652 S 0.0 0.5 0:02.53 apache2 14393 www-data 20 0 412028 53100 19948 S 0.0 0.4 0:01.94 apache2 27803 www-data 20 0 402056 41772 18600 S 0.0 0.3 0:01.25 apache2 15549 www-data 20 0 402056 41768 18600 S 0.0 0.3 0:01.71 apache2 27800 www-data 20 0 402056 41768 18600 S 0.0 0.3 0:01.57 apache2 13965 root 20 0 213288 27464 13576 R 9.0 0.2 0:00.72 php 3536 root 20 0 212760 26904 13500 S 3.3 0.2 0:01.11 php 13623 root 20 0 210944 26816 13428 S 11.3 0.2 0:00.81 php 13961 root 20 0 212440 26476 13620 S 11.9 0.2 0:00.81 php 14253 root 20 0 212184 26468 13692 S 10.6 0.2 0:00.72 php 3508 root 20 0 2480888 26372 6144 S 8.0 0.2 0:02.42 python 13166 root 20 0 211412 25716 13636 S 2.0 0.2 0:00.43 php 13884 root 20 0 211412 25672 13628 S 3.3 0.2 0:00.40 php 12710 root 20 0 211668 25656 13408 S 2.7 0.2 0:00.45 php 13402 root 20 0 211412 25644 13472 S 2.7 0.2 0:00.47 php 12369 root 20 0 211412 25564 13488 S 2.0 0.2 0:00.42 php 13650 root 20 0 211156 25564 13620 S 2.3 0.2 0:00.57 php 14925 root 20 0 211156 25508 13596 S 13.3 0.2 0:00.40 php 15018 root 20 0 209340 25500 13408 S 10.9 0.2 0:00.33 php 15220 root 20 0 209340 25492 13408 S 10.9 0.2 0:00.33 php 14003 root 20 0 211156 25480 13632 S 2.3 0.2 0:00.32 php 14006 root 20 0 211412 25480 13400 S 7.6 0.2 0:00.56 php 14958 root 20 0 211156 25472 13636 S 9.0 0.2 0:00.27 php 15250 root 20 0 211156 25468 13644 S 8.0 0.2 0:00.24 php 13867 root 20 0 211412 25448 13412 S 2.7 0.2 0:00.36 php 13886 root 20 0 211412 25428 13388 S 4.0 0.2 0:00.38 php 14493 root 20 0 211156 25428 13580 S 5.3 0.2 0:00.33 php 14818 root 20 0 211412 25428 13380 S 13.3 0.2 0:00.40 php 14519 root 20 0 211156 25412 13556 S 3.0 0.2 0:00.32 php 13397 root 20 0 211156 25372 13528 S 2.0 0.2 0:00.30 php 15426 root 20 0 209084 25172 13344 S 7.6 0.2 0:00.23 php 15360 root 20 0 211156 25152 13348 S 6.3 0.2 0:00.19 php 15482 root 20 0 211156 25092 13284 S 7.3 0.2 0:00.22 php 1795 syslog 20 0 206180 24652 12960 S 0.3 0.2 12:02.16 php 1420 root 20 0 385240 24084 18164 S 0.0 0.2 0:34.51 apache2 12774 root 20 0 162852 22284 11896 R 5.3 0.2 0:00.87 php 13589 root 20 0 140736 21780 10968 R 12.9 0.2 0:00.89 php 15670 root 20 0 195568 15476 12004 R 1.3 0.1 0:00.04 php 509 syslog 20 0 337948 14192 2800 S 0.3 0.1 4:35.01 rsyslogd 3164 www-data 20 0 385516 9604 3380 S 0.0 0.1 0:00.05 apache2 27012 www-data 20 0 385516 9600 3380 S 0.0 0.1 0:00.01 apache2 3497 www-data 20 0 385492 9540 3380 S 0.0 0.1 0:00.00 apache2 6992 www-data 20 0 385484 9500 3380 S 0.0 0.1 0:00.01 apache2 11916 www-data 20 0 385272 8008 2076 S 0.0 0.1 0:00.00 apache2
-----Original Message----- From: observium [mailto:observium-bounces@observium.org] On Behalf Of Tom Laermans Sent: Friday, December 04, 2015 1:37 PM To: Observium Network Observation System Subject: Re: [Observium] monitor tmpfs storage
My Observium host currently has 1.2GB in use - what are you doing to that poor thing that it has 10 gigs in use during normal operation?! :-)
Simon,
The RAMdisk only takes up its full size when you fill it up - but when it's full, it's just full like a full disk. No more new RRDs created. If it's not being shown in Observium, it's likely to be matching something in an ignored array I'd think? Observium can definitely monitor tmpfs, as long as your snmp reports it through the correct tables.
Tom
On 04/12/2015 17:44, Ron Marosko wrote:
Much more RAM, in fact you may already be knocking your system performance if you only have 12GB in the machine. I've got 12GB allocated to my observium VM, and the system is using 9-10GB of RAM just for regular operation. And then when I do this: root@observium:/opt/observium# du -hs rrd 27G rrd
Yeah, I'd have to allocate at least 40GB of DRAM to the VM in order to do the ramdisk thing, and that's kinda difficult to do when the VMhost only has 16GB of physical memory installed. Oops.
-----Original Message----- From: observium [mailto:observium-bounces@observium.org] On Behalf Of Simon Smith Sent: Friday, December 04, 2015 10:02 AM To: Observium Network Observation System Subject: [Observium] monitor tmpfs storage
Hi All,
just moved our RRDs to tmpfs storage, and WOW I noticed a speed difference, not only in the graphs, but also in the IO load
should of looked at http://observium.org/docs/persistent_ramdisk/ in the past DOH!
however i’ve just noticed the observium isn’t showing it as a storage mount?
can observium monitor actually monitor it at all ?
also I’ve set my limit at 10GB as the server has 12GB of RAM, and the RRD folder is currently at 6.8GB
bit of a stupid question but what happens if I’m close to the limit? as i knackered? more RAM server?
regards
Simon - Hestor Ltd _______________________________________________ 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