Content
@
1 reply
0 recast
2 reactions
✳️ dcposch on daimo
@dcposch.eth
is there anything better than honeycomb for service monitoring & tracing? i see datadog but have never used it
9 replies
0 recast
25 reactions
Lemma
@lemma
Like others have said, Datadog is good but expensive. I personally like Honeycomb a lot too. Self-hosted Elastic stack is also good and is not a ton of work to run yourself IMO. I built and ran Elastic previously as a single person doing infra. Some small outages over the years but overall very positive experience.
1 reply
0 recast
0 reaction
Lemma
@lemma
E.g. I ran a small 3 node cluster that worked well for services with large number of users (think dozens of requests per second which isn’t a ton but fits the scale for most startups). Upgraded the node size once, and beyond that scaled horizontally (just add more nodes). Kibana is a pretty nice interface too.
1 reply
0 recast
0 reaction
Lemma
@lemma
Overall cost went up to a maximum of $300-$500/mo I think on AWS, but we were retaining data for months and had internal services built on top of that data so there was read load beyond just developers using it to debug issues or view logs.
0 reply
0 recast
0 reaction