How to change interval of collecting date ?

[n0405]

first to say SexyGraf is very nice monitorg tool for esxi .

I am looking , but with no sucess , to change interval of collecting data from 5 minutes to 1 minut .

[rschitz]

Hi,
Thanks for your support.
To do that you'd need to change the carbon schema, the perfmanager queries and the cron tab.
I can tell you what to do but can i ask you what you need first and your env size pls?

[n0405]

Hi ,
my env size is 35 servers
I need to get data more frequent , for instance if is selected period for last 5 minutes on graph it shows only one values , and it will be nice that show a line of values for every minutes in that period .

[rschitz]

There is 2 aspects about this request.
First, for how long you need this 1min granularity because the default 5min granularity is kept for 48h, after that it decrease for space an efficiency.
Second, for which metrics you need that because some metrics come from properties and may not be refreshed every minute.

[n0405]

I suppose that one month will be enough, but only metrics regard to VMs and Host in cluster usage,

[rschitz]

1 month @1min is huge, you sure you need that much?
here is the default for reference : retentions = 5m:24h,10m:48h,80m:7d,4h:30d,12h:90d,48h:1y,96h:2y,288h:5y

[n0405]

yes , it is . but I need more data about cpu utilizacion , cpu ready

[maomaomaopan]

Hi, SexyGraf is very nice monitorg tool.
I have a same question, I want to see past 30 days data at 10minute intervals.
so I changed config file (/opt/graphite/conf/storage-schemas.conf), but no success

[anything_else]
pattern = .*
retentions = 5m:24h,10m:48h,80m:7d,240m:30d,720m:90d,2880m:1y,5760m:2y,17280m:5y
↓ 
retentions = 5m:24h,10m:48h,10m:7d,10m:30d,80m:90d,2880m:1y,5760m:2y,17280m:5y

Do you have any ideas?

[rschitz]

Hi @maomaomaopan and thanks for your support.
This will only be applied to the NEW whisper files so you have to delete your files so they will be recreated with the correct schema or you change the existing one with a command like this #217 (comment)

[rschitz]

@Nzaca i'm currently rewriting the vSphere collection part so let me try to figure something out

[maomaomaopan]

Hi @rschitz !
It worked as I was looking for
Thanks for your support!

[rschitz]

@Nzaca check the sexigraf pull exec time dashboard and how much time it takes to exec your actual pull?

[n0405]

Hi @rschitz
I checked and it's take 10s and happens every 5m.
last week I try with nohup /usr/bin/find -L /mnt/wfs/whisper/vmw/ -type f ( -name '*.wsp' ) -exec /usr/local/bin/whisper-resize.py {} 1m:24h,5m:48h,60m:7d,120m:30d,720m:90d,2880m:1y,5760m:2y,17280m:5y --nobackup ; &
and also change retentions in /opt/graphite/conf/storage-schemas.conf
retentions = 1m:24h,5m:48h,60m:7d,120m:30d,720m:90d,2880m:1y,5760m:2y,17280m:5y
but I didn't notice any change

[n0405]

Hi @rschitz ,
after a while I deal with these intervals again.
I stoped the service carbon-cache and execute nohup /usr/bin/find -L /mnt/wfs/whisper/vmw/ -type f ( -name '*.wsp' ) -exec /usr/local/bin/whisper-resize.py {} 1m:24h,1m:48h,5m:7d,60m:30d,60m:90d,60m:1y --nobackup ; &
and notice in nohup.out that indicate that my retention isn't good . So maybe that is problem from beginning

Traceback (most recent call last):
File "/usr/local/bin/whisper-resize.py", line 4, in
import('pkg_resources').run_script('whisper==1.1.7', 'whisper-resize.py')
File "/usr/local/lib/python3.5/dist-packages/pkg_resources/init.py", line 650, in run_script
self.require(requires)[0].run_script(script_name, ns)
File "/usr/local/lib/python3.5/dist-packages/pkg_resources/init.py", line 1446, in run_script
exec(code, namespace, namespace)
File "/usr/local/lib/python3.5/dist-packages/whisper-1.1.7-py3.5.egg/EGG-INFO/scripts/whisper-resize.py", line 69, in
for retentionDef in args[1:]]
File "/usr/local/lib/python3.5/dist-packages/whisper-1.1.7-py3.5.egg/EGG-INFO/scripts/whisper-resize.py", line 69, in
for retentionDef in args[1:]]
File "/usr/local/lib/python3.5/dist-packages/whisper-1.1.7-py3.5.egg/whisper.py", line 176, in parseRetentionDef
raise ValueError("Invalid retention specification '%s'" % points)
ValueError: Invalid retention specification '24h,1m:48h,5m:7d,60m:30d,60m:90d,60m:1y'

[rschitz]

Hi @Nzaca check this : https://graphite.readthedocs.io/en/latest/config-carbon.html#storage-schemas-conf

[n0405]

pull exec time

How can I change this interval ?

[rschitz]

In the /etc/cron.d/vi_* but it won't be accurate since the script is configured to catch the 15 latest 20sec metrics (5min total) and compute the median. In order to execute every 1min it would need lot of changes.