Conversation
d8e2122 to
0130f2b
Compare
Doesn't it make more sense to count this per collection? |
|
I made this metric not per-collection to keep consistent with the other metrics. Additionally, we can compare this parameter against |
e1f6f5b to
424a232
Compare
a2e1318 to
479a1e8
Compare
f3b7c36 to
c099206
Compare
d8fcfa8 to
0ef6150
Compare
timvisee
left a comment
There was a problem hiding this comment.
I don't think we correctly handle resharding. See my review remark below.
Happy to assist you with this one to form the logic here.
| .filter(|i| i.status == TrackerStatus::Optimizing) | ||
| .count(); | ||
|
|
||
| let min_max_active_replicas = collection |
There was a problem hiding this comment.
let's move replica counting into a separate function/file. This one becomes bulky
There was a problem hiding this comment.
I'd suggest to do this in a separate PR once the other ones are merged too. This way rebasing and merging with the other currently open PRs is significantly easier.
timvisee
left a comment
There was a problem hiding this comment.
Happy to approve once the above review remarks have been resolved ✔️
ee21c86 to
33545d9
Compare
7044a84 to
df0c11e
Compare
* Add replica metrics * Clippy * Properly ignore shards with resharding status * Include 0 as value in metrics * Remove total_active_replicas and total_expected_replicas * Improve minimum active replica count in case of resharding * Reorder metrics calls min < max (to trigger CI) * Fix metrics help * Fix typo --------- Co-authored-by: timvisee <tim@visee.me>
Adds the following new metrics to the /metrics api