4 ms·
Nothing against choosing this set of apps really, but I'm curious why collectd and not telegraf which does the same kind of metrics probes and is a part of the
by viraptor 6y ago
Nothing against choosing this set of apps really, but I'm curious why collectd and not telegraf which does the same kind of metrics probes and is a part of the TICK stack.
- bbrks 6y agoAs somebody new to both, are you able to elaborate on what telegraf/TICK provides that the collection of programs in the post does not? I see lots of posts complaining about the stack in the post, but all of the alternatives posed don't really explain why.
- viraptor 6y agoTelegraf and collectd do roughly the same thing. They run some plugins to get data and push the results to a given metrics sink. I asked because TICK (Telegraf, InfluxDB, Chronograf and Kapacitor) is a known solution and a fairly standard way to add elements of monitoring to your system.
- jimktrains2 6y ago> TICK (Telegraf, InfluxDB, Chronograf and Kapacitor) is a known solution and a fairly standard way to add elements of monitoring to your system. Amusing that I've never heard of any of this but have heard and used collectd. It's obviously no where near as common as a lamp stack or anywhere near common at all, so asking why it over something else is answered by "someone made it up so it's better".
- viraptor 6y agoCollectd is only one part here. Did you need a solution for inline processing/aggregation and alerting? If not, you wouldn't run into TICK. It's not common overall, because few environments need to go that far. I don't get the comparison to LAMP popularity. Insects are more common than cars too. They're different things ¯\_(ツ)_/¯
- jimktrains2 6y agoI mentioned lamp as an example of an acronym and stack that would be more common to ask why it wasn't chosen. In retrospect that probably wasn't the ideal way of getting to my point that tick isn't a common stack so asking why not it because it's common seems weird. > If not, you wouldn't run into TICK. It's not common overall, because few environments need to go that far. That's my point though. The post I replied to acted as though everyone knows of and uses it, but provided no information on what makes it a better choice for such usecases.
- cbzbc 6y agocollectd comes packaged by default in a bunch of distributions and telegraf does not -- for basic functionality telegraf doesn't really buy anything and takes more configuration.
- mhall119 6y ago`snap install telegraf` is pretty easy though
- cbzbc 6y agoNot everything uses or even can run ubuntu. I have a similar setup and have a number of openwrt devices that also feed things into grafana via collectd - so I already have that up and running.
- mhall119 6y agoWhat kind of packaging does openwrt use? We publish tarballs of Telegraf that should work, but if there's a different packaging format that would make it easier to install we can look into adding that.
- sciurus 6y agoAlthough I've successfully used collectd in the past, I would choose telegraf over it now. This is especially true if I'm sending data to InfluxDB,s ince it and telegraf are primarily developed by InfluxData. Both telegraf and collectd are mature programs with a large base of well-documented plugins. However, telegraf development is more active, and I find its golang codebase much more approachable than the C codebase of collectd. Obviously, YMMV. In general though, I wouldn't advise adopting the entire TICK stack. Chronograf is inferior to Grafana. Kapacitor has a super-steep learning curve and is being deprecated by InfluxData. TIG (telegraf, influxdb, grafana) can be a good choice. Prometheus plus Grafana is worth evaluating as well; there's arguably more momentum in that ecosystem than in InflxuDB's.