by

Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    Cory Parent
    @goedelsoup
    that TF provider is gonna be a blessing
    Adelbert Chang
    @adelbertc
    @timperrett are you folks using Prometheus to scrape metrics for Nelson itself? you guys use Wavefront right?
    Cory Parent
    @goedelsoup
    i do myself
    i do prom server per DC + weave/grafana cloud
    Adelbert Chang
    @adelbertc
    i think i am going to have to add some InfluxDB stuff
    Timothy Perrett
    @timperrett
    @adelbertc we use prometheus importer for telegraph, then we send that out to wavefront. We use telegraf for everything
    Adelbert Chang
    @adelbertc
    yep thats exactly what i ended up doing
    i forgot to update here :sweat_smile:
    Cory Parent
    @goedelsoup
    do you guys notice any significant benefits from the push model of influx on what you can action upon, view, etc?
    Adelbert Chang
    @adelbertc
    i have not noticed anything yet but i am a novice user
    Timothy Perrett
    @timperrett
    after influx went private source i havent paid much attention
    i guess they are getting traction on private / hybrid cloud?
    Cory Parent
    @goedelsoup
    i'm kinda just wondering if influx's design really offers much more than prom w/ pushgateway in the right places. i doubt i'll reverse course, but now would be the time so i'm interrogating random parties :laughing:
    Timothy Perrett
    @timperrett
    prom has specific issues also - its not hugely scalable without significant engienering work (see uber m3)
    Cory Parent
    @goedelsoup
    not too big a deal for me -- interesting though, i wonder how this compares to weave/thanos ecosystem that most people are on now
    Timothy Perrett
    @timperrett
    Weave? wasnt taht the overlay network from ages ago
    “most people” is a trigger for me :P
    Cory Parent
    @goedelsoup
    yeah, weave is "multi-tenancy for prom". thanos is a bit more interesting and virtualizes the metrics store to provide transparent access to hot/cold metrics alike. https://github.com/thanos-io/thanos
    Adelbert Chang
    @adelbertc
    a marvel-themed project name? sold.
    Christopher Davenport
    @ChristopherDavenport
    You all using epimetheus to expose Prometheus metrics?
    Adelbert Chang
    @adelbertc
    heh nah, just the simple java client
    i dont think epimetheus existed when Nelson was created ;)
    Timothy Perrett
    @timperrett
    right we predate all that
    Matt Explosion
    @aldiyen
    hey sorry to trouble you guys, but is anyone ever going to merge and release getnelson/helm#7 I’m still running a home grown version in my prod environment
    Timothy Perrett
    @timperrett
    oh yes, we can absolutely do that
    it will be version 6.0.0 as its an incompatible dependency change
    @aldiyen its running on release CI now - will post here when its done
    Timothy Perrett
    @timperrett
    @aldiyen 6.0.13 is released!
    Matt Explosion
    @aldiyen
    awesome, thanks @timperrett !
    Adelbert Chang
    @adelbertc
    is there a way to change thhe logging level dynamically and/or by restarting the service?
    Cory Parent
    @goedelsoup
    you should be able to pass root.level via a properties file based on how the logback config is assembled
    Adelbert Chang
    @adelbertc
    java logging will be the end of me
    Cory Parent
    @goedelsoup
    Adelbert Chang
    @adelbertc
    thats what i ended up doing
    Timothy Perrett
    @timperrett
    lol
    Matt Explosion
    @aldiyen
    java logging is an exciting world of configuration files
    Adelbert Chang
    @adelbertc
    @timperrett @goedelsoup have either of you observed the Nelson Deployment Monitor thread periodically dying
    Cory Parent
    @goedelsoup
    i'm still not deployed :cry: still working on getting all of our artifact tags right
    Adelbert Chang
    @adelbertc
    D:
    Cory Parent
    @goedelsoup
    i literally wrote a baby nelson this week. it feels like the last stepping stone though.
    gets everything on nomad (templating via levant) and then we just need to start versioning like adults.
    Matt Explosion
    @aldiyen
    just noticed that someone opened a PR to the Verizon/helm repo back in May. would you folks be open to me reaching out to Verizon to try to get ownership of the repo transferred to the getnelson github organization reduce confusion? looks like the existing fork would need to be renamed in order for the transfer to be accepted, so probably some annoyance there…
    Christopher Davenport
    @ChristopherDavenport
    There are people still at Verizon?
    Matt Explosion
    @aldiyen
    oh god, are there not?
    that’s even more horrifying
    Cory Parent
    @goedelsoup
    not to my knowledge. i have been making PRs and having conversations to try to get references shifted over.
    i don't even know if we are able to get a reference link into the READMEs at this point though. tim might be the best pathway since i think he negotiated the open-sourcing.
    btw, thank you @aldiyen for pushing that. i was doing the same and was very lazy about it.
    Matt Explosion
    @aldiyen
    i sent an email