Very good point. Skye is correct that performance is not the #1 priority for BetterKV, pricing is, b

Very good point. Skye is correct that performance is not the #1 priority for BetterKV, pricing is, but I do also definitely see your point. I’ll probably add your suggestions as a performance mode that people can enable

HHello, I’m Allie!Very good point. Skye is correct that performance is not the #1 priority for Bet...

beelzabub•7/14/23, 11:13 PM

It's also net a cost savings. Would you be open to making it the default & the user needing to turn it off? Even if you don't want to do the probabilistic piece, can you shorten the cacheTTL so that it's under what you give KV? Otherwise, BetterKV is not really different than cache+R2.

Bbeelzabub My goal is to win the bet & get Sid (TL for pages) to shave his head bald if we ...

Walshy•7/14/23, 11:13 PM

I can't wait for this

Bbeelzabub It's also net a cost savings. Would you be open to making it the default & the u...

Hello, I’m Allie!OP•7/14/23, 11:16 PM

Qq, what’s your username on Discord, so I can credit you for the changes?

HHello, I’m Allie!Qq, what’s your username on Discord, so I can credit you for the changes?

beelzabub•7/14/23, 11:16 PM

vlovich

beelzabub•7/14/23, 11:17 PM

Oh that's my github username

beelzabub•7/14/23, 11:17 PM

beelzabub#0369 on Discord

Hello, I’m Allie!OP•7/14/23, 11:21 PM

Wait

Hello, I’m Allie!OP•7/14/23, 11:21 PM

No, I meant GH

Hello, I’m Allie!OP•7/14/23, 11:21 PM

I guess I’m just brain-tired from the heat…

Hello, I’m Allie!OP•7/14/23, 11:21 PM

beelzabub•7/14/23, 11:23 PM

But yeah, being more likely to refreshing the value as you approach the expriy end date is a significant cost saving measure. Counter-intuitive but true. It's how KV has to operate under the hood for similar reasons. We technically use a bounded exponentially increasing algorithm where after 5s there's an infinitesimaly small probability of a refresh that doubles with every passing second up to 100% at something like 45s or 50s at which point every read has a 100% probability of causing a background refresh. That way only a small subset of requests end up triggering a refresh while the rest just always serve from the cache without any stampeding herd. We have other tricks we're layering in as well, but they're not available externally yet.

Bbeelzabub Basically the cacheTTL you use for the fronting cache is the same as the cacheTT...

zehawk•7/15/23, 8:56 AM

How old is this writeup? I did not even know that putting cache in front of KV was a thing at some point, and from the time I started using (~2 years now), I always assumed KV is parallel to Cache. In other words, since KV is already at the edge, what's the point of cache in front of a cache?

Zzehawk How old is this writeup? I did not even know that putting cache in front of KV w...

Unsmart•7/15/23, 11:36 AM

To reduce billing since KV cache hits are still billed but the cache api is free

Bbeelzabub Basically the cacheTTL you use for the fronting cache is the same as the cacheTT...

DaniFoldi•7/15/23, 11:48 AM

Now that you have exposed the cache status of reads, this becomes a lot easier to implement - previously when I thought about a probabilistic update, using the time the read took as an indicator for hit/miss was rather messy, so thanks for implementing MeowHeartCloudflare

DaniFoldi•7/15/23, 11:49 AM

(one small issue I noticed with the docs PR, the ToC seems to cut off at some point)

SShak Do KV writes have to replicate from the local POP to the KV origin/storage serve...

beelzabub•7/15/23, 7:32 PM

The latter

エ

エキノコックスジャパン•7/16/23, 12:42 PM

Hi.
Is there any problem with publishing Cloudflare kv id, preview_id on github?

kv_namespaces = [
  { binding = "TEST", id = "xxx", preview_id = "yyy" },
]

kv_namespaces = [
  { binding = "TEST", id = "xxx", preview_id = "yyy" },
]

I have defined id, preview_id in wrangler.toml like this, is it ok to upload it as is where it can be seen by unspecified people? Or should I put them in .dev.vars?

zegevlier•7/16/23, 12:48 PM

Everything in

wrangler.toml

wrangler.toml

is safe to upload to version control

エ

エキノコックスジャパン•7/16/23, 12:55 PM

@zegevlier Thank you.
Is it OK to write directly in wrangler.toml and make it public for things like api key provided by cloudflare, not just kv?
I'm curious because GCP and AWS keep things like api_key secret.

エエキノコックスジャパン @zegevlier Thank you. Is it OK to write directly in wrangler.toml and make it pu...

zegevlier•7/16/23, 1:04 PM

No. API keys should never be made public. You should not be storing these in your

wrangler.toml

wrangler.toml

. KV ids are not API keys, so they can be made public.

Zzegevlier No. API keys should *never* be made public. You should not be storing these in y...

エ

エキノコックスジャパン•7/16/23, 1:14 PM

Thank you very much.
I wasn't sure how to tell which items should be secret, but I think your advice helped me understand a little better.

I checked the cloudflare dashboard and the api tokens clearly says "Protect this key like a password!" to keep it secret.

I'll try to keep this kind of warning secret, but if there is anything else I should be aware of, could you please let me know?

エエキノコックスジャパン Thank you very much. I wasn't sure how to tell which items should be secret, but...

zegevlier•7/16/23, 1:27 PM

In general (this might not always be true, but it's usually a good rule of thumb. As far as I know it always holds for Cloudflare stuff), IDs are safe to share, keys and secrets are not.

Zzegevlier In general (this might not always be true, but it's usually a good rule of thumb...

エ

エキノコックスジャパン•7/16/23, 1:30 PM

Thank you again and again for your kindness. I understand.

Jadugar_Jaggu•7/18/23, 6:40 AM

Guys there is issue in kVs

Jadugar_Jaggu•7/18/23, 6:40 AM

it crashed my whole server almost 5 times today

Jadugar_Jaggu•7/18/23, 7:19 AM

Fixed now

Erisa•7/18/23, 11:09 AM

Hi @bhadoo, can you share with me your affected account ID?

Erisa•7/18/23, 11:11 AM

Thank you, I am looking

BBhadoo Error: Error: KV GET failed: 500 Internal Server Error at async getAccessTok...

Erisa•7/18/23, 11:20 AM

i see an increase in errors around that time but it should be back to normal now. Please do let me know if you see this again

Erisa•7/18/23, 12:25 PM

Checking

BBhadoo Just checked, happening again. Last 3 minutes ago. Received 6 notifications of ...

Erisa•7/18/23, 12:46 PM

This has been escalated

Erisa•7/18/23, 12:47 PM

The right teams are looking at it now

zehawk•7/18/23, 12:59 PM

Is there a disconnect in the docs between:

https://developers.cloudflare.com/workers/runtime-apis/kv/#cache-ttl "It is not recommended if your data is updated often and you need to see updates shortly after they are written, because writes that happen from other global network locations will not be visible until the cached value expires."

vs

The new decoupled arch as described in https://blog.cloudflare.com/faster-workers-kv-architecture/

(I would like to be sure that increasing the cacheTTL value will not backfire on me.)

James•7/18/23, 1:01 PM

The new arch isn’t fully rolled out yet, but I imagine once it is, that will be updated. For now, I wouldn’t build anything with the new arch in mind and expect things to work as they did in the past.

JJames The new arch isn’t fully rolled out yet, but I imagine once it is, that will be ...

zehawk•7/18/23, 1:01 PM

Ah ok, good that I checked.

BBhadoo Alright

Erisa•7/18/23, 1:07 PM

It's now an incident on our side, you can track here: https://www.cloudflarestatus.com/incidents/k5nmwlwzntcj
Thanks for reporting

Increased Workers KV 500 HTTP Errors in multiple regions

Very good point. Skye is correct that performance is not the #1 priority for BetterKV, pricing is, b

Similar Threads