* Remove NewPossibleNonCounterInfo until it can be made more efficient, and avoid creating empty annotations as much as possible
Signed-off-by: Jeanette Tan <jeanette.tan@grafana.com>
* Bump prometheus common to v0.44.0
Signed-off-by: Yannick te Kulve <738464+YannickTeKulve@users.noreply.github.com>
* Fix golang_protobuf_extensions sum
Signed-off-by: Yannick te Kulve <738464+YannickTeKulve@users.noreply.github.com>
* Remove unused deps
Signed-off-by: Yannick te Kulve <738464+YannickTeKulve@users.noreply.github.com>
---------
Signed-off-by: Yannick te Kulve <738464+YannickTeKulve@users.noreply.github.com>
When Prometheus restarts it creates every series read in from the WAL,
but many of those series will be finished, and never receive any more
samples. By defering allocation of the txRing slice to when it is first
needed, we save 32 bytes per stale series.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
Broken by https://github.com/prometheus/prometheus/pull/12738. We have to update both global variables (as GlobalConfig is not a pointer here).
DefaultConfig is used when no global: section is provided, whereas DefaultGlobalConfig is used when it's provided and for individual scrape configs.
Reported on #prometheus-dev (thanks to @beorn7): https://cloud-native.slack.com/archives/C01AUBA4PFE/p1697733267205649
Tested manually, it would be nice to add test at some point (quick fix for now).
Signed-off-by: bwplotka <bwplotka@gmail.com>
In proto3, this doesn't change anything. However, since the
`CreatedTimestamp` field is generated as a pointer
(`*types.Timestamp`), we are still able to detect the unset state.
(This is in contrast to the `timestamp_ms` field, which is a plain
int64, for which we cannot enforce generation as a pointer, see
comment updated in the previous commit for future actions.)
Signed-off-by: beorn7 <beorn@grafana.com>
When reading the WAL this method is called with buffers from a pool, on
multiple goroutines. Pre-allocating sufficient size avoids slow growth
and many reallocations in `append`.
Signed-off-by: Bryan Boreham <bjboreham@gmail.com>
On a 32 bit architecture the size of int is 32 bits. Thus converting from
int64, uint64 can overflow it and flip the sign.
Try for yourself in playground:
package main
import "fmt"
func main() {
x := int64(0x1F0000001)
y := int64(1)
z := int32(x - y) // numerically this is 0x1F0000000
fmt.Printf("%v\n", z)
}
Prints -268435456 as if x was smaller.
Followup to #12650
Signed-off-by: György Krajcsovits <gyorgy.krajcsovits@grafana.com>
build(deps): bump github.com/prometheus/prometheus from 0.45.0 to 0.47.2 in /documentation/examples/remote_storage
Signed-off-by: Levi Harrison <git@leviharrison.dev>
Improve promtool tsdb analyze
- Make it more suitable for variable size float chunks.
- Add support for histogram chunks.
---------
Signed-off-by: Ziqi Zhao <zhaoziqi9146@gmail.com>
Signed-off-by: Levi Harrison <git@leviharrison.dev>
* Additionally wrap WBL replay error
Although WBL replay is already wrapped with errLoadWbl,
there are other errors that can happen during a WBL replay.
We should not try to repair WAL in those cases.
This commit additionally wraps the final error in Head.Init again
with errLoadWbl so that WBL replay errors can be identified properly.
Signed-off-by: Ganesh Vernekar <ganeshvern@gmail.com>
Signed-off-by: Jesus Vazquez <jesusvzpg@gmail.com>
Co-authored-by: Jesus Vazquez <jesusvzpg@gmail.com>
Signed-off-by: Levi Harrison <git@leviharrison.dev>
zenador