Cloudflare Developers•3y ago

but what the user felt

daveOP•2/28/23, 4:51 PM

cause health analytics is showing my Worker takes a whooping ~150ms on average

Ddave but what the user felt

Hello, I’m Allie!•2/28/23, 4:54 PM

Maybe with RUM?

Hello, I’m Allie!•2/28/23, 4:54 PM

But that necessitates client-side JS to measure.

daveOP•2/28/23, 4:56 PM

I want to see like the RTT if I used curl too

Ddave I want to see like the RTT if I used curl too

Hello, I’m Allie!•2/28/23, 4:57 PM

That would require network speed metrics, which you basically can’t get without a client-side script(or running a ping toward the users IP address)

Hello, I’m Allie!•2/28/23, 4:57 PM

At most you can know when a Request enters the CF edge, and when the Response leaves

HHello, I’m Allie!At most you can know when a Request enters the CF edge, and when the Response le...

daveOP•2/28/23, 5:00 PM

that's what I want to know

Ddave that's what I want to know

Hello, I’m Allie!•2/28/23, 5:01 PM

Might be able to do that with LogPush. Do you have Enterprise?

HHello, I’m Allie!Might be able to do that with LogPush. Do you have Enterprise?

daveOP•2/28/23, 5:03 PM

Pro

Ddave Pro

Hello, I’m Allie!•2/28/23, 5:07 PM

Don’t think you can do it then

daveOP•2/28/23, 5:29 PM

hmm

daveOP•2/28/23, 5:29 PM

I see EventTimestampMs in logpush for the start of the Worker I think

daveOP•2/28/23, 5:29 PM

but doesn't give me the end time sadly

Ddave I see EventTimestampMs in logpush for the start of the Worker I think

Hello, I’m Allie!•2/28/23, 5:30 PM

Yeah, I think you need full LogPush to measure that, which is Ent and above, afaik

Noel•2/28/23, 5:32 PM

is wrangler tail currently broken?

Noel•2/28/23, 5:33 PM

i was using 2.10.0 and got no logs so i updated to 2.12.0 (latest) and i still don't get logs

daveOP•2/28/23, 5:34 PM

enable worker_devworker_dev once, publish, and then you can disable worker_devworker_dev

daveOP•2/28/23, 5:34 PM

I found that to be the best fix

kian•2/28/23, 5:35 PM

the current worker incident would impact tail

daveOP•2/28/23, 5:35 PM

better answer lol;

Noel•2/28/23, 5:37 PM

oh, there is an ongoing incident?

daveOP•2/28/23, 5:37 PM

yep

daveOP•2/28/23, 5:38 PM

Can I enable Logpush for Page Functions?

Noel•2/28/23, 5:39 PM

so it's also possible that workers doesn't update and i don't get errors during deploy?

Ddave Can I enable Logpush for Page Functions?

Hello, I’m Allie!•2/28/23, 5:39 PM

I believe if you have LogPush enabled, it should cover functions

NNoel so it's also possible that workers doesn't update and i don't get errors during ...

daveOP•2/28/23, 5:39 PM

yes, it's annoying

Noel•2/28/23, 5:41 PM

that's a bit relieving to know that this time it isn't on me

HHello, I’m Allie!*I believe* if you have LogPush enabled, it *should* cover functions

daveOP•2/28/23, 5:42 PM

https://developers.cloudflare.com/workers/platform/logpush/#enable-logging-on-your-worker

Logpush · Cloudflare Workers docs

Cloudflare Logpush supports the ability to send Workers Trace Event Logs to a supported destination. Worker’s Trace Events Logpush includes metadata …

daveOP•2/28/23, 5:42 PM

it has to be enabled manually, no?

daveOP•2/28/23, 5:42 PM

for all my Worker scripts anyway

Walshy•2/28/23, 5:44 PM

Not possible, I need to see how we go about making that available

daveOP•2/28/23, 5:44 PM

ty, I will stop looking in the dashboard then

daveOP•2/28/23, 6:02 PM

Does just having a D1 binding (even if not using it) cause any slowdown?

charl.dev•2/28/23, 6:04 PM

with the reported workers issue, has it been resolved - we seeing a worker looks like its cached / not updating

charl.dev•2/28/23, 6:04 PM

its our stg env and after 8pm here so not hectic but need to sort it out

Original message was deleted

charl.dev•2/28/23, 6:08 PM

sigh if this was our prod instance, PSD2s all round

Unsmart•2/28/23, 6:09 PM

Issues happen

UUnsmart Issues happen 🤷

charl.dev•2/28/23, 6:09 PM

so do fines

Unsmart•2/28/23, 6:10 PM

If you get fined for down time maybe consider never deploying code so ur never down

charl.dev•2/28/23, 6:11 PM

guess thats why we have staging lol

UUnsmart If you get fined for down time maybe consider never deploying code so ur never d...

daveOP•2/28/23, 6:11 PM

except for last night, where V8 decided to change how dates are printed at runtime lol

Ddave except for last night, where V8 decided to change how dates are printed at runti...

Unsmart•2/28/23, 6:12 PM

well I mean locale data changes so you cant really always expect a locale to print a certain way

charl.dev•2/28/23, 6:12 PM

so we should stop deploying if we want it to come back, catch up?

charl.dev•2/28/23, 6:16 PM

any rough eta, so we can let other teams know

charl.dev•2/28/23, 6:28 PM

what was odd for us and is still happening, after a deploy at 6pm, the worker has something to check the req pathname and then basically set that pathname in a cookie.

For some reason, the request for our main.js was being picked up in the worker as the req pathname and being set client side.

The actual worker code wasnt changed, but only the KV assets. Im guessing it couldnt find the main.js or something because the manifest in the worker didnt tie up to the hashed main.js

Something like that.

Is there no way to tell when a specific worker has caught up with deploys?

Unsmart•2/28/23, 6:30 PM

you can add headers with a version you know and you will know when it updated

Unsmart•2/28/23, 6:33 PM

but what the user felt

Similar Threads

Similar Threads

Similar Threads