Re: [prometheus-users]

[Adso Castro]

Hey,

Well, you can search some guides on youtube, there's plenty of it, just 
search "Kubernetes + Prometheus" so you can learn the basics from 
AlertManager and Prometheus Rules as well. Also, don't forget the official 
docs:

AlertManager: https://prometheus.io/docs/alerting/latest/alertmanager/
Prometheus: https://prometheus.io/docs/introduction/overview/

Em segunda-feira, 20 de julho de 2020 às 13:44:06 UTC-3, 
melissa...@gmail.com escreveu:

> Hi,
>
> I use Prometheus to monitor Kubernetes clusters. I would like to 
> receive an alert when Kubernetes API server is not running. How to do it?
>
> Thank you,
>
> Melissa
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/94914a2b-08e1-4533-a791-242eb0a77b28n%40googlegroups.com.

[prometheus-users] Heavy Queries and Grafana Dashboards

[Adso Castro]

Hey all,

Does anyone know how can I identity heavy queries and heavy (Grafana) 
dashboards so I can create records to relieve it from my Prometheus stack? 
Maybe a query or something else? 

Couldn't find anything close to it.

Thank you.

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/56ab5526-2228-46c6-aa1d-436de147586an%40googlegroups.com.

[prometheus-users] Spring MVC client example

[Coder Noob]

Hello,

I am trying to integrate my Spring MVC 5 application to Prometheus. But as 
far as I see the data at /prometheus endpoint is not pushed to prometheus. 
All the tutorials are for Spring Boot and I cannot migrate my app to 
SpringBoot right now.
Can you explain me how can I integrate the Spring MVC application to 
Prometheus?

Best regards,

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/6197ca55-7819-4760-9c04-cedb3246e8d7o%40googlegroups.com.

Re: [prometheus-users] What causes "counter jumped backwards" in node-exporter

[Bruce Merry]

On Monday, July 20, 2020 at 8:24:59 PM UTC+2 sup...@gmail.com wrote:

> Thanks for posting this. I've been looking for more cases where people see 
> these issues.
>
> Linux does not do any mutex locking of the CPU metric counters. So we 
> added some to the node_exporter in order to detect and mitigate spurious 
> counter resets.
>
> In all of my testing and evidence, I've only see this happen on iowait 
> data. But it's interesting that you see it on other events is useful. I 
> don't think your docker use has any impact. I'd suspect it has more to do 
> with the underlying server environment.
>

Looking the last week of logs, I see 41% for user, 31% for idle, 26% for 
system, 1.4% for iowait, 0.3% for softirq.
 

> Is this bare metal? VMs? What hypervisor?
>

It's bare metal. What's also odd is that it's restricted to one batch of 
machines that have the same hardware and run the same workloads. Other 
machines (with different hardware and workloads) don't exhibit these 
warnings despite having been deployed at the same time with the same kernel 
and OS. The affected machines have 12 cores and hyperthreading enabled (so 
24 virtual cores) while the other machines generally have up to 8 cores and 
no HT, so possibly the affected machines have a higher chance of running 
into race conditions for the kernel data structures.

Let me know if there are other details you'd like to investigate.

Cheers
Bruce

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/f75a9bcf-0b59-4184-8191-c0da05886b38n%40googlegroups.com.

Re: [prometheus-users] What causes "counter jumped backwards" in node-exporter

[Ben Kochie]

Thanks for posting this. I've been looking for more cases where people see
these issues.

Linux does not do any mutex locking of the CPU metric counters. So we added
some to the node_exporter in order to detect and mitigate spurious counter
resets.

In all of my testing and evidence, I've only see this happen on iowait
data. But it's interesting that you see it on other events is useful. I
don't think your docker use has any impact. I'd suspect it has more to do
with the underlying server environment. Is this bare metal? VMs? What
hypervisor?

See the discussion and code in this issue:
https://github.com/prometheus/node_exporter/issues/1686



On Mon, Jul 20, 2020 at 8:11 PM Bruce Merry  wrote:

> We've recently upgraded a fleet of machines to both Ubuntu 18.04 (Linux
> 5.3) and node-exporter to 1.0.1 (run under Docker). We're now seeing a lot
> of messages like these:
>
> CPU Idle counter jumped backwards, possible hotplug event, resetting CPU
> stats
> CPU User counter jumped backwards
> CPU System counter jumped backwards
> CPU Iowait counter jumped backwards
> CPU SoftIRQ counter jumped backwards
>
> I examined a handful of messages and the counter seems to go backwards by
> a very small amount (0.01). Over the last 3 days, 74% of the messages are
> from one machine and 24% from another, with multiple machines making up the
> rest.
>
> I assume that this something odd in the system itself and that
> node-exporter is just reporting what it sees. But any tips on what might
> cause it / how to fix it, and whether it is worth worrying about? Is it
> likely to be caused by running under Docker?
>
> Here's the uname and docker-compose.yml for completeness:
>
> Linux imgr1 5.3.0-62-generic #56~18.04.1-Ubuntu SMP Wed Jun 24 16:17:03
> UTC 2020 x86_64 x86_64 x86_64 GNU/Linux
>
> version: '2'
> services:
>   node-exporter:
> image: prom/node-exporter:v1.0.1
> command:
>   - '--collector.textfile.directory=/textfile'
>   - '--no-collector.zfs'
>   - '--no-collector.wifi'
>   - '--no-collector.bcache'
>   - '--path.rootfs=/host'
>   - '--collector.vmstat.fields=^(oom_kill|pgpg|pswp|nr|pg.*fault).*'
>   -
> '--collector.filesystem.ignored-fs-types=^(tmpfs|autofs|binfmt_misc|cgroup|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|mqueue|fuse.lxcfs|nfs|nsfs|ceph|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|sysfs|tracefs)$$'
> network_mode: host
> pid: host
> volumes:
>   - /var/spool/node-exporter:/textfile:ro
>   - /:/host:ro,rslave
>
> Thanks
> Bruce
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prometheus Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to prometheus-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/prometheus-users/2aac7b8b-8780-4a2f-973d-b0344720b876n%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CABbyFmoBsgQB%2Bcxb0FPvGfB5t1vTS%2BJ-ajxSM-Jbs1uQVR1G0w%40mail.gmail.com.

[prometheus-users] What causes "counter jumped backwards" in node-exporter

[Bruce Merry]

We've recently upgraded a fleet of machines to both Ubuntu 18.04 (Linux 
5.3) and node-exporter to 1.0.1 (run under Docker). We're now seeing a lot 
of messages like these:

CPU Idle counter jumped backwards, possible hotplug event, resetting CPU 
stats
CPU User counter jumped backwards
CPU System counter jumped backwards
CPU Iowait counter jumped backwards
CPU SoftIRQ counter jumped backwards

I examined a handful of messages and the counter seems to go backwards by a 
very small amount (0.01). Over the last 3 days, 74% of the messages are 
from one machine and 24% from another, with multiple machines making up the 
rest.

I assume that this something odd in the system itself and that 
node-exporter is just reporting what it sees. But any tips on what might 
cause it / how to fix it, and whether it is worth worrying about? Is it 
likely to be caused by running under Docker?

Here's the uname and docker-compose.yml for completeness:

Linux imgr1 5.3.0-62-generic #56~18.04.1-Ubuntu SMP Wed Jun 24 16:17:03 UTC 
2020 x86_64 x86_64 x86_64 GNU/Linux

version: '2'
services:
  node-exporter:
image: prom/node-exporter:v1.0.1
command:
  - '--collector.textfile.directory=/textfile'
  - '--no-collector.zfs'
  - '--no-collector.wifi'
  - '--no-collector.bcache'
  - '--path.rootfs=/host'
  - '--collector.vmstat.fields=^(oom_kill|pgpg|pswp|nr|pg.*fault).*'
  - 
'--collector.filesystem.ignored-fs-types=^(tmpfs|autofs|binfmt_misc|cgroup|configfs|debugfs|devpts|devtmpfs|fusectl|hugetlbfs|mqueue|fuse.lxcfs|nfs|nsfs|ceph|overlay|proc|procfs|pstore|rpc_pipefs|securityfs|sysfs|tracefs)$$'
network_mode: host
pid: host
volumes:
  - /var/spool/node-exporter:/textfile:ro
  - /:/host:ro,rslave

Thanks
Bruce

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/2aac7b8b-8780-4a2f-973d-b0344720b876n%40googlegroups.com.

[prometheus-users]

[dealvips.com]

Hi,

I use Prometheus to monitor Kubernetes clusters. I would like to receive an
alert when Kubernetes API server is not running. How to do it?

Thank you,

Melissa

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CANLicqMn-HB9BdAEiWCddzwBk3vH4b9r0y9LHbE%3DtJX0pHo5rQ%40mail.gmail.com.

[prometheus-users]

[dealvips.com]

Hi,

I use Prometheus to monitor Kubernetes Cluster. I would like to receive an
email alert if Kubernetes etcd is not running. How to do it?


Thank you,


Melissa

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CANLicqNBfHYCe2rZx-RKQAUfBO5kUkByX1yaChayh93tuQ%2B_RQ%40mail.gmail.com.

[prometheus-users]

[dealvips.com]

Hi,

I use Prometheus to monitor Kubernetes Cluster. I would like to receive an
alert if a node is down. How to do it?

Thanks,

Melissa

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CANLicqMwMcRggSTjdM-pgvXu3TGQ5xHR_YaAnL8uSdP9YFxO5Q%40mail.gmail.com.

Re: [prometheus-users] SNMP exporter using supermicroserver

[Mohamed Es-saoudi]

Hey Ben,

Vielen Dank.

do you have any tutorial video how to this, the gihub link seems 
complicated :/

Am Montag, 20. Juli 2020 17:03:19 UTC+2 schrieb Ben Kochie:
>
> You're probalby looking for the ipmi_exporter.
>
> https://github.com/soundcloud/ipmi_exporter#power-consumption
>
> On Mon, Jul 20, 2020 at 5:00 PM Mohamed Es-saoudi  > wrote:
>
>> Hello,
>>
>> did someone now what is the OID for the power usage of the supermicro 
>> server?
>>
>> thanks
>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "Prometheus Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to promethe...@googlegroups.com .
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/prometheus-users/0d2dd202-f5c1-4939-8c2b-a06c0792a653o%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/b3582e46-7473-4c92-a162-e215dcd5dde7o%40googlegroups.com.

Re: [prometheus-users] Multiple exporters per machine

[Ben Kochie]

In addition to what Stuart Clar said, I want to talk about one specific
question:

> It seems non-optimal to have prometheus scrape  4 different ports on
thousands of machines

The Prometheus design is highly parallel. It uses lightweight goroutines to
manage the scrapes of targets. So it's actually more optimal to have many
small targets, vs few large ones. Having many exporters allows Prometheus
to scale up the scrape functions and ingest data smoothly. While this seems
like it would be a down side, that's mostly a problem for other monitoring
systems. For Prometheus, this is working as designed.

Of course, there are limits. Thousands of targets per job is fine, but
millions would be too much for a single Prometheus server.

On Mon, Jul 20, 2020 at 3:17 PM Isaac Graf  wrote:

> I am new to prometheus and we are looking to to install it on an existing
> complex production system.
> It looks like we will need multiple exporters per machine- e.g.
> node_exporter, statsd exporter, custom metric exporter, and some machine
> types will need other exporters (such as redis exporter).
> It seems non-optimal to have prometheus scrape  4 different ports on
> thousands of machines.
> Is there a better way to approach this?
>
> Thanks very much.
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prometheus Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to prometheus-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/prometheus-users/fd3691bc-6015-4e56-b894-d13cccf35ad4o%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CABbyFmovRWZ0MtX_UbgWjFCSWgPja%3DVvM1%3DkYgu_%2B%2BzYJZmh5w%40mail.gmail.com.

Re: [prometheus-users] SNMP exporter using supermicroserver

[Ben Kochie]

You're probalby looking for the ipmi_exporter.

https://github.com/soundcloud/ipmi_exporter#power-consumption

On Mon, Jul 20, 2020 at 5:00 PM Mohamed Es-saoudi 
wrote:

> Hello,
>
> did someone now what is the OID for the power usage of the supermicro
> server?
>
> thanks
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prometheus Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to prometheus-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/prometheus-users/0d2dd202-f5c1-4939-8c2b-a06c0792a653o%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CABbyFmoZgUG2mK4iGLokFz5shYc9P6P0R4thz9_giWJuTE6gpw%40mail.gmail.com.

[prometheus-users] SNMP exporter using supermicroserver

[Mohamed Es-saoudi]

Hello,

did someone now what is the OID for the power usage of the supermicro 
server?

thanks

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/0d2dd202-f5c1-4939-8c2b-a06c0792a653o%40googlegroups.com.

[prometheus-users] Monitor NodeJs

[Yasmine Mbarek]

Hello everyone , 
Is there any other way to monitor NodeJS application without touching 
anything in the app.js file ?

Thank you.

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/86021092-bc33-4b5f-be8b-ecde332e53fco%40googlegroups.com.

[prometheus-users] Problem with multiple receivers in alertmanager

[PASS IT Consulting GmbH Research]

Hello

I have the following situation: I am running several shops  on a server.
I have three groups of persons (receivers)

   -  The tech admins, which should receive all alarms
   - The shop team  which should see all shop related alarm
   - The shop owner who should only see the results for his shop


I tried to achieve this using the following route definition

# The root route on which each incoming alert enters.
route:

  # A default receiver
  receiver: tech-admins

  # All the above attributes are inherited by all child routes and can 
  # overwritten on each.


  

  # The child route trees.
  routes:

  # All  alerts
  # are dispatched to the technical admins
  - receiver: 'tech-admins'
continue: true
group_wait: 10s
match_re:
  job: '.*'

routes: 
 
# All shop alerts
# are dispatched to the shop team.
- receiver: 'shop-admins'
  group_by: [product]
  continue: true
  group_wait: 10s
  match_re:
product: 'Shop.*'
  routes:


  # Alerts for individual shops are send to th the shop admins 
  - receiver: 'ownerTest' 
match:
  product: 'ShopTest'
  
When I now trigger an alarm for ShopTest te shop team and the shop owner 
receive email but not the tech-admins

I do not understand why this happens?

Is there a way to debug or test the alarms

P.S.  I did not made the rule for the techadmins the default route, since 
there a contionue true is forbidden.

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/644aba78-c95e-4fd3-be10-ee3dcdd37de3n%40googlegroups.com.

Re: [prometheus-users] Multiple exporters per machine

[Stuart Clark]

On 2020-07-20 14:17, Isaac Graf wrote:

I am new to prometheus and we are looking to to install it on an
existing complex production system.
It looks like we will need multiple exporters per machine- e.g.
node_exporter, statsd exporter, custom metric exporter, and some
machine types will need other exporters (such as redis exporter).
It seems non-optimal to have prometheus scrape  4 different ports on
thousands of machines.
Is there a better way to approach this?


The design of Prometheus is for applications to ideally expose metrics 
about their operation. Where applications don't have native Prometheus 
or OpenMetrics support exporters are used. These are designed to be 
specific to the application being monitored (e.g. MySQL or the Linux 
kernel).


As a result you may end up with several exporters running on a single 
machine, depending on what applications you have installed that don't 
have native metrics support.


In addition the separation over multiple exporters ensures you don't 
have a single point of failure causing total metrics loss (at most you'd 
just lose the metrics that a specific exporter was handling).


--
Stuart Clark

--
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/8bd21057e363a2027050aee672819396%40Jahingo.com.

[prometheus-users] Multiple exporters per machine

[Isaac Graf]

I am new to prometheus and we are looking to to install it on an existing 
complex production system.
It looks like we will need multiple exporters per machine- e.g. 
node_exporter, statsd exporter, custom metric exporter, and some machine 
types will need other exporters (such as redis exporter).
It seems non-optimal to have prometheus scrape  4 different ports on 
thousands of machines.
Is there a better way to approach this? 

Thanks very much.

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/fd3691bc-6015-4e56-b894-d13cccf35ad4o%40googlegroups.com.

Re: [prometheus-users] Re: absent(nonexistent{job="myjob",instance=~".*"}) why miss label instance

[576501...@qq.com]

What  mean?

在 2020年7月20日星期一 UTC+8下午2:59:06，Julien Pivotto写道：
>
> up offet 1h unless up
>
>
> That would fire for an hour with instance label.
>
> Le lun. 20 juil. 2020 à 08:47, 5765...@qq.com  <
> 5765...@qq.com > a écrit :
>
>> Our requirement is the alarm for up metric, using 
>> absent(up{job="myjob",instance=~ ".* "}). but you won't get the instance 
>> information. Do you have any other suggestions?
>>
>> 在 2020年7月20日星期一 UTC+8下午2:29:22，Brian Candler写道：
>>>
>>> 1. There are an infinite number of possibilities which could match 
>>> instance=~".*"
>>>
>>> 2. The result of that query is vector of timeseries.  Each timeseries 
>>> must have concrete labels like foo="bar", whereas foo=~"bar|baz" is a 
>>> PromQL querying construct only.  That is, it's not a label, it's an 
>>> expression that selects labels.
>>>
>> -- 
>> You received this message because you are subscribed to the Google Groups 
>> "Prometheus Users" group.
>> To unsubscribe from this group and stop receiving emails from it, send an 
>> email to promethe...@googlegroups.com .
>> To view this discussion on the web visit 
>> https://groups.google.com/d/msgid/prometheus-users/1c3e069d-79d5-442b-9671-b92d810bda0eo%40googlegroups.com
>>  
>> 
>> .
>>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/d5d68c74-9396-4643-bdf4-1f6582912a35o%40googlegroups.com.

Re: [prometheus-users] snmp_exporter: lookups doesn't work as expected with PowerNet MIB

[Egor Vasilyev]

Thank you very much, it works!

пятница, 17 июля 2020 г., 20:15:54 UTC+3 пользователь Ben Kochie написал:
>
> The index for these metrics has two index components, so when doing 
> lookups you have to include both.
>
> modules:
>   apc_cooling:
> walk:
> - coolingUnitStatusAnalogTable
> lookups:
> - source_indexes:
>   - coolingUnitTableIndex
>   - coolingUnitStatusAnalogTableIndex
>   lookup: coolingUnitStatusAnalogDescription
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/4c93f569-159e-4bea-945e-61d375a2d88do%40googlegroups.com.

[prometheus-users] Re: snmp_exporter: lookups doesn't work as expected with PowerNet MIB

[Egor Vasilyev]

Thank you very much, it works!

среда, 15 июля 2020 г., 12:33:56 UTC+3 пользователь Egor Vasilyev написал:
>
> I use APC PowerNet MIB 4.2.6. My generator.yaml:
>
> modules:
>   template_acrd_602:
> version: 3
> max_repetitions: 25
> retries: 3
> timeout: 30s
> walk:
>   - coolingUnitStatusAnalogTableIndex
>   - coolingUnitStatusAnalogDescription
>   - coolingUnitStatusAnalogValue
>
>
> I try:
> # curl -sw '' '
> http://127.0.0.1:9116/snmp?module=template_acrd_602=hostname'
>
> And receive (the original output contains dozens of metrics, I shrank it):
> coolingUnitStatusAnalogDescription{coolingUnitStatusAnalogDescription="Air 
> Filter Run Hrs",coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 1
> coolingUnitStatusAnalogTableIndex{coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 44
> coolingUnitStatusAnalogValue{coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 4472
>
>
> And it's ok. But if I add a lookup section into generator.yaml like this:
>
> modules:
>   template_acrd_602:
> version: 3
> max_repetitions: 25
> retries: 3
> timeout: 30s
> walk:
>   - coolingUnitStatusAnalogTableIndex
>   - coolingUnitStatusAnalogDescription
>   - coolingUnitStatusAnalogValue
> lookups:
>   - source_indexes: [coolingUnitStatusAnalogTableIndex]
> lookup: coolingUnitStatusAnalogDescription
>
>
> I expect to receive an output something like this:
>
> coolingUnitStatusAnalogDescription{coolingUnitStatusAnalogDescription="Air 
> Filter Run Hrs",coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 1
> coolingUnitStatusAnalogTableIndex{coolingUnitStatusAnalogDescription="Air 
> Filter Run Hrs"coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 44
> coolingUnitStatusAnalogValue{coolingUnitStatusAnalogDescription="Air 
> Filter Run Hrs"coolingUnitStatusAnalogTableIndex="44",
> coolingUnitTableIndex="1"} 4472
>
>
> But actually I receive:
> coolingUnitStatusAnalogDescription{coolingUnitStatusAnalogDescription="",
> coolingUnitStatusAnalogTableIndex="44",coolingUnitTableIndex="1"} 1
> coolingUnitStatusAnalogTableIndex{coolingUnitStatusAnalogDescription="",
> coolingUnitStatusAnalogTableIndex="44",coolingUnitTableIndex="1"} 44
> coolingUnitStatusAnalogValue{coolingUnitStatusAnalogDescription="",
> coolingUnitStatusAnalogTableIndex="44",coolingUnitTableIndex="1"} 4472
>
>
> What am I doing wrong? Help please. Thank you!
>
> OID description you could get on web site 
> 
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/d9fe594a-6f15-44d8-a716-bd7656693caeo%40googlegroups.com.

Re: [prometheus-users] How to (correctly) upsize a HA prometheus setup?

[sayf eddine Hammemi]

Hello,
Doesn't it mean that your alert expression is not reliable enough? maybe
you are trying to solve an XY problem here. did you try to make an
expression that does not trigger without enough history?

On Mon, Jul 20, 2020 at 10:26 AM 'Holger Freyther' via Prometheus Users <
prometheus-users@googlegroups.com> wrote:

> tl;dr: How to turn-up new Prometheus instances and not send (or silence)
> false alerts?
>
> My HA setup consists out of running several independent instances that
> scrape, evaluate the recording rules and alert. That should be pretty
> standard and works great until I need to increase the retention size (or
> turn-up in a new cluster, maintenance, ..). To upsize I get compute and new
> storage and spawn a new instance of Prometheus.
>
> My troubles start with alerts using larger windows on increase/rate (e.g.
> check progress of a job over a longer period of time). The new instance has
> an empty TSDB and just started to scrape it is likely to trigger this alert.
>
> I wondered if I can silence but AFAIK we make the label sets match across
> all instances of the HA set-up to allow Alertmanager to dedup.
>
> Is there a best practice for this?
>
> holger
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "Prometheus Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to prometheus-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/prometheus-users/c9d8c05d-ec07-4adc-b4bf-10cd88b0b5cen%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CALnV8WhBPbmX2TzzDy6i_dX%3DbS0Z0Vmz%2BkqzNWojgh%3DZ28rUKA%40mail.gmail.com.

[prometheus-users] How to (correctly) upsize a HA prometheus setup?

['Holger Freyther' via Prometheus Users]

tl;dr: How to turn-up new Prometheus instances and not send (or silence) 
false alerts?

My HA setup consists out of running several independent instances that 
scrape, evaluate the recording rules and alert. That should be pretty 
standard and works great until I need to increase the retention size (or 
turn-up in a new cluster, maintenance, ..). To upsize I get compute and new 
storage and spawn a new instance of Prometheus. 

My troubles start with alerts using larger windows on increase/rate (e.g. 
check progress of a job over a longer period of time). The new instance has 
an empty TSDB and just started to scrape it is likely to trigger this alert.

I wondered if I can silence but AFAIK we make the label sets match across 
all instances of the HA set-up to allow Alertmanager to dedup.

Is there a best practice for this?

holger


-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/c9d8c05d-ec07-4adc-b4bf-10cd88b0b5cen%40googlegroups.com.

Re: [prometheus-users] Re: absent(nonexistent{job="myjob",instance=~".*"}) why miss label instance

[Julien Pivotto]

up offet 1h unless up


That would fire for an hour with instance label.

Le lun. 20 juil. 2020 à 08:47, 576501...@qq.com <576501...@qq.com> a écrit :

> Our requirement is the alarm for up metric, using
> absent(up{job="myjob",instance=~ ".* "}). but you won't get the instance
> information. Do you have any other suggestions?
>
> 在 2020年7月20日星期一 UTC+8下午2:29:22，Brian Candler写道：
>>
>> 1. There are an infinite number of possibilities which could match
>> instance=~".*"
>>
>> 2. The result of that query is vector of timeseries.  Each timeseries
>> must have concrete labels like foo="bar", whereas foo=~"bar|baz" is a
>> PromQL querying construct only.  That is, it's not a label, it's an
>> expression that selects labels.
>>
> --
> You received this message because you are subscribed to the Google Groups
> "Prometheus Users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to prometheus-users+unsubscr...@googlegroups.com.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/prometheus-users/1c3e069d-79d5-442b-9671-b92d810bda0eo%40googlegroups.com
> 
> .
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/CAFJ6V0piwsF5VGctLEHpG7GESOK2BaYEoyZZie17KW0SUyXUSw%40mail.gmail.com.

[prometheus-users] Re: absent(nonexistent{job="myjob",instance=~".*"}) why miss label instance

[576501...@qq.com]

Our requirement is the alarm for up metric, using 
absent(up{job="myjob",instance=~ ".* "}). but you won't get the instance 
information. Do you have any other suggestions?

在 2020年7月20日星期一 UTC+8下午2:29:22，Brian Candler写道：
>
> 1. There are an infinite number of possibilities which could match 
> instance=~".*"
>
> 2. The result of that query is vector of timeseries.  Each timeseries must 
> have concrete labels like foo="bar", whereas foo=~"bar|baz" is a PromQL 
> querying construct only.  That is, it's not a label, it's an expression 
> that selects labels.
>

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/1c3e069d-79d5-442b-9671-b92d810bda0eo%40googlegroups.com.

[prometheus-users] Re: absent(nonexistent{job="myjob",instance=~".*"}) why miss label instance

[Brian Candler]

1. There are an infinite number of possibilities which could match 
instance=~".*"

2. The result of that query is vector of timeseries.  Each timeseries must 
have concrete labels like foo="bar", whereas foo=~"bar|baz" is a PromQL 
querying construct only.  That is, it's not a label, it's an expression 
that selects labels.

-- 
You received this message because you are subscribed to the Google Groups 
"Prometheus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to prometheus-users+unsubscr...@googlegroups.com.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/prometheus-users/5bad054e-035f-4cac-8926-2ce0c5c61d2bo%40googlegroups.com.