SuperQue

11 ms·

61.
by SuperQue 7y ago
Yup, and that's kinda intentional. The design is to be a monitoring system, not a generic TSDB. What it does needs to be simple, fast, and reliable so that your alerts get sent out. The original design inspiration, borgmon, also was a
62.
by SuperQue 7y ago
Backfilling data is on the roadmap. It's already possible, the tools are being worked on. Also, accurate reporting and backfill are not related.
63.
by SuperQue 7y ago
The videos will be uploaded soon. Don't worry, none of the slides or other content was lost. We're just waiting for our video company to finish the final renders with all the correct color and lighting quality adjustments made. Ea
64.
by SuperQue 8y ago
We generate 1T of logs per day. Doing this with PostgreSQL would be hilariously bad.
65.
by SuperQue 8y ago
This is not metrics, this is logs. Logs contain a lot of detail that are not appropriate for metrics. Logs output tends to be expensive, it costs a lot of time and IO to emit, compress, send, store logs. Logs also tend to contain PII things
66.
by SuperQue 8y ago
I was thinking it would be funny to call it Logki.
67.
by SuperQue 8y ago
Are you talking about the Cortex Prometheus service? They're working on transferring that to Grafana branding.
68.
by SuperQue 8y ago
Yup, as a hiring manager I lost a couple of good hires due to the comp calculator. Especially in markets, like where I live. The market for engineer salaries is much higher, due to the demand. But the cost of living is very low, by comparis
69.
by SuperQue 8y ago
Probably cheaper, as you can start negotiating discounts when you get to this size.
70.
by SuperQue 8y ago
Yea, I haven't used it myself, but the reports are that it works better than the original PromCache proxy. It's been on my TODO list for a while, but way lower on the priority list. But you know, when the internet decides it'
71.
by SuperQue 8y ago
It is a separate instance from our internal one. We have a cron job that automatically copies the internal Grafana dashboards to the public one, so you still see exactly what we see. We used to use Federation, but now we just have the publ
72.
by SuperQue 8y ago
Yea, I'm working on that, will be deploying this nice caching proxy to speed up the dashboard. Thanks to Comcast for creating Trickster. https://github.com/Comcast/trickster
73.
by SuperQue 9y ago
Percona just released a migration tool. https://github.com/Percona-Lab/prom-migrate
74.
by SuperQue 9y ago
We just upgraded one of GitLab's Prometheus servers (we have HA redundant servers). We saw a better than 10x reduction in CPU use. https://imgur.com/3aJtYu9 This server doesn't take any query load, but it does tak
75.
by SuperQue 9y ago
It's not about push or pull or exporters, it's about where you aggregate your events. As soon as you are working with a non-trivial number of events, you need to think more carefully about how deal with them. With standard statsd
76.
by SuperQue 9y ago
Yup, I've had more than one current Google SRE state "You can have my Borgmon when you pry it from my cold dead hands.". Borgmon may be dead in the eyes of some people, but I know for a fact that it's still the only thin