Prometheus metrics period #339
Comments
|
Hi, try using `--history-duration`. You can see all available cli options
by running `krr simple --help`
…On Mon, Sep 23, 2024 at 8:04 AM Najafov007 ***@***.***> wrote:
Hi again! Wanted to ask a ques, as I've understood robusta krr shows only
the last 14days of a pod resource requests and limits and gives a
suggestion. Is there any way to change that number to let's say fewer days?
Or it's as it is - 14 days? Thank you!)
—
Reply to this email directly, view it on GitHub
<#339>, or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AADYUB4U4QQOUV6ZD5FI3WDZX6OPJAVCNFSM6AAAAABOVIRWQ2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGU2DCNJYGI2TQMY>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Hello again! Thank you for your response, wanted to ask one more thing. Got it about history-duration, we need 7 days. Problem now is time of scratching the metrics... For example we give 7 days, is it scraping the data from prometehus for last ENTIRE 7 days? What I exactly mean, lets say our pods are actively working in between 8AM - 6PM, time before or after that interval pods are cooled down. We dont want robusta krr to take into consideration. We need it to watch only those hours and then give us a suggestion. You know why, it will be wrong from math side if we are considering not only actively working hours, and passive time. Thank you so much!! |
|
If the pods aren't running at all, there is no data and it wont be taken
into account. If the pods are running but have low cpu/memory usage, that
will be taken into account. You can tweak the P95 etc thresholds (see `krr
simple --help` for options) to adjust for this.
…On Wed, Oct 2, 2024 at 3:48 PM Najafov007 ***@***.***> wrote:
Hi, try using --history-duration. You can see all available cli options
by running krr simple --help
… <#m_-3088870937315420851_>
On Mon, Sep 23, 2024 at 8:04 AM Najafov007 *@*.*> wrote: Hi again! Wanted
to ask a ques, as I've understood robusta krr shows only the last 14days of
a pod resource requests and limits and gives a suggestion. Is there any way
to change that number to let's say fewer days? Or it's as it is - 14 days?
Thank you!) — Reply to this email directly, view it on GitHub <#339
<#339>>, or unsubscribe
https://github.com/notifications/unsubscribe-auth/AADYUB4U4QQOUV6ZD5FI3WDZX6OPJAVCNFSM6AAAAABOVIRWQ2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGU2DCNJYGI2TQMY
<https://github.com/notifications/unsubscribe-auth/AADYUB4U4QQOUV6ZD5FI3WDZX6OPJAVCNFSM6AAAAABOVIRWQ2VHI2DSMVQWIX3LMV43ASLTON2WKOZSGU2DCNJYGI2TQMY>
. You are receiving this because you are subscribed to this thread.Message
ID: @.*>
Hello again! Thank you for your response, wanted to ask one more thing.
Got it about history-duration, we need 7 days. Problem now is time of
scratching the metrics... For example we give 7 days, is it scraping the
data from prometehus for last ENTIRE 7 days? What I exactly mean, lets say
our pods are actively working in between 8AM - 6PM, time before or after
that interval pods are cooled down. We dont want robusta krr to take into
consideration. We need it to watch only those hours and then give us a
suggestion. You know why, it will be wrong from math side if we are
considering not only actively working hours, and passive time. Thank you so
much!!
—
Reply to this email directly, view it on GitHub
<#339 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AADYUB2QQ2K2YLJFGIEH63DZZPTQTAVCNFSM6AAAAABOVIRWQ2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDGOBYGU3DGMBSGQ>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Hi again! Wanted to ask a ques, as I've understood robusta krr shows only the last 14days of a pod resource requests and limits and gives a suggestion. Is there any way to change that number to let's say fewer days? Or it's as it is - 14 days? Thank you!)
The text was updated successfully, but these errors were encountered: