feat(replay): Bump limit for minReplayDuration

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run

[chargome]

With this PR users can set their min replay duration to max 50s, previously this was capped at 15s.

We cannot bump this value further as this would lead to dropping buffered replays (we keep max. 60s in-memory at this point)

closes#18109

[chargome]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

[billyvg]

@billyvg just wondering, would we not also drop buffered replays here since they can be anywhere from 0-60s?

Yeah i think we just have to strongly warn users about this if we are going to add it.

Though looking at [the code]... it seems like we only consider the session start to current flush time as the "duration" - I can't recall if, while in buffer mode, we mutate the session.start time at allwe do

[billyvg]

We should add the comment in the replay options type so that it's visible to users. We should also be clear that this can drop onError-sampled replays

[github-actions]

node-overhead report 🧳

Note: This is a synthetic benchmark with a minimal express app and does not necessarily reflect the real-world performance impact in an application.

Scenario	Requests/s	% of Baseline	Prev. Requests/s	Change %
GET Baseline	10,765	-	8,786	+23%
GET With Sentry	1,586	15%	1,357	+17%
GET With Sentry (error only)	7,312	68%	6,118	+20%
POST Baseline	1,137	-	1,190	-4%
POST With Sentry	521	46%	524	-1%
POST With Sentry (error only)	997	88%	1,049	-5%
MYSQL Baseline	3,838	-	3,304	+16%
MYSQL With Sentry	464	12%	450	+3%
MYSQL With Sentry (error only)	3,168	83%	2,716	+17%

View base workflow run