Andreas Gerstmayr

> @andreasgerstmayr it does sound a little risky, and maybe also something that will be difficult to make reliable/deterministic across many kernel versions. FWIW, there's also /proc/pid/oom_score that may help...

Hm that's strange indeed. For example *Context switches per second* (`kernel.all.pswitch`) and *Runnable* (`kernel.all.runnable`) are both metrics without indom, from the same PMDA (`pmdalinux`) but the context switches graph is...

Yep, your stream.expire and stream.maxlen settings look fine. A couple more commands to diagnose: ``` # Get metadata about the stored series # note the series id (the hash before...

No problem. I think this issue can be closed then?

fwiw, there's `jq` (https://github.com/stedolan/jq, available in Fedora, RHEL, Debian, etc.) which does JSON pretty printing (with color support) among other things (e.g. transformations)

> our json lib also has jsonpointer code, which would be very useful in QA once we write the front-end code ah, that's the part where I currently resort to...

> To confirm, the log you pasted is the entirety of the verifier output you're seeing? Yes. > Also, which are you building against? The latest commit of this repo...

@planetharsh: I tested it with the latest kernel available on CentOS Stream 9, which is: `5.14.0-145.el9.ppc64le` as of today, and it still fails. Do you have a RPM package of...

Hi @chenhengqi, this is the output with `-v`: ``` # ./runqlat -v libbpf: loading object 'runqlat_bpf' from buffer libbpf: elf: section(2) .text, size 2024, link 0, flags 6, type=1 libbpf:...

> | [...] 20 MB of JSON [...] Still very high fwiw, > > @andreasgerstmayr we haven't yet implemented http compression in pmproxy and this feels like it might provide...