API Limit changed? #313
Comments
|
Facing the same issue from yesterday, requests are getting cancelled randomly. |
|
Facing the same issue, glad I saw this, I was getting confused with what is happening. |
|
Not related but whenever the dose 1 capacity is >0 I am getting a 403 at the same time. I know how 403 works, but unusual behavior. After the slots become 0 the error stops coming. |
|
Finally, I was also facing the same issue since yesterday, that it all of a sudden throws an error that the request is blocked by CORS. This is what I got just now: No 'Access-Control-Allow-Origin' header is present on the requested resource. |
|
403 is intermittent. It happens for one 5 min cycle and then stops.
… On 22-May-2021, at 6:19 PM, Divya Vishwakarma ***@***.***> wrote:
Finally, I was also facing the same issue from yesterday, that it all of a sudden throws an error that the request is blocked by CORS.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A>.
|
But it wasn't happening earlier, right? |
|
Right. It has increased today. I have to restart the run but it happens again. Surprisingly it happens sometime when I pull a particular district out of 6 districts.
It’s becoming pointless day by day to run a public service tracker.
I have been helping so many people in need of vaccination,
But after 18th, several front end and backend changes have happened, making me think twice whether to run this service or not.
… On 22-May-2021, at 6:21 PM, Divya Vishwakarma ***@***.***> wrote:
403 is intermittent. It happens for one 5 min cycle and then stops.
… <x-msg://12/#>
On 22-May-2021, at 6:19 PM, Divya Vishwakarma @.***> wrote: Finally, I was also facing the same issue from yesterday, that it all of a sudden throws an error that the request is blocked by CORS. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#313 (comment) <#313 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A>.
But it wasn't happening earlier, right?
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF5UCGPKSSP2KNHVQ6TTO6SGVANCNFSM45KQTE7A>.
|
Same dude. Well, I built this application for my city only but has helped a number of folks. But after these API's are getting failed at regular intervals irritates a lot. |
|
No one from cowin is interested in replying to freeloaders like us, they might sell their connections to marketing companies soon.
Their public data is so old now that it’s become impossible to book a slot.
AS UI also now stops multiple button clicks, it’s like they are happy that they have stopped a HACK, but it was saving lives of real people.
Multi bene bookings stopped.
Time out reduced.
A user gets logged out faster than before.
It’s becoming an over engineered product with lacklustre results.
… On 22-May-2021, at 6:26 PM, Divya Vishwakarma ***@***.***> wrote:
Right. It has increased today. I have to restart the run but it happens again. Surprisingly it happens sometime when I pull a particular district out of 6 districts. It’s becoming pointless day by day to run a public service tracker. I have been helping so many people in need of vaccination, But after 18th, several front end and backend changes have happened, making me think twice whether to run this service or not.
… <x-msg://13/#>
On 22-May-2021, at 6:21 PM, Divya Vishwakarma @.> wrote: 403 is intermittent. It happens for one 5 min cycle and then stops. … x-msg://12/# On 22-May-2021, at 6:19 PM, Divya Vishwakarma @.> wrote: Finally, I was also facing the same issue from yesterday, that it all of a sudden throws an error that the request is blocked by CORS. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#313 <#313> (comment) <#313 (comment) <#313 (comment)>>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A> https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A>. But it wasn't happening earlier, right? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#313 (comment) <#313 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUDIZF5UCGPKSSP2KNHVQ6TTO6SGVANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF5UCGPKSSP2KNHVQ6TTO6SGVANCNFSM45KQTE7A>.
Same dude. Well, I built this application for my city only but has helped a number of folks. But after these API's are getting failed at regular intervals irritates a lot.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF2RST5TFA4455DOFNDTO6SYPANCNFSM45KQTE7A>.
|
Yeah dude, like I came here because their official website has this Github Repo link to lodge issues. But it seems like nobody gives a heck. |
Same man, i am also running vaccine notification service for 20+ districts, had to put out an notice that can't provide notification all of a sudden due to cowin in every district. |
|
With the 100 calls /5min restriction how are you providing an instant slot count change? Wouldn’t repeating every district take like a lot of time?
I am unable to scale up beyond 6 districts, the slot disappears so fast.
This api is giving false information at time.
… On 22-May-2021, at 6:48 PM, Harsh Vijay ***@***.***> wrote:
Right. It has increased today. I have to restart the run but it happens again. Surprisingly it happens sometime when I pull a particular district out of 6 districts. It’s becoming pointless day by day to run a public service tracker. I have been helping so many people in need of vaccination, But after 18th, several front end and backend changes have happened, making me think twice whether to run this service or not.
… <x-msg://14/#>
On 22-May-2021, at 6:21 PM, Divya Vishwakarma @.> wrote: 403 is intermittent. It happens for one 5 min cycle and then stops. … x-msg://12/# On 22-May-2021, at 6:19 PM, Divya Vishwakarma @.> wrote: Finally, I was also facing the same issue from yesterday, that it all of a sudden throws an error that the request is blocked by CORS. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#313 <#313> (comment) <#313 (comment) <#313 (comment)>>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A> https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF7BHZXLPBGC5GU7U53TO6R4NANCNFSM45KQTE7A>. But it wasn't happening earlier, right? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#313 (comment) <#313 (comment)>>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AUDIZF5UCGPKSSP2KNHVQ6TTO6SGVANCNFSM45KQTE7A <https://github.com/notifications/unsubscribe-auth/AUDIZF5UCGPKSSP2KNHVQ6TTO6SGVANCNFSM45KQTE7A>.
Same man, i am also running vaccine notification service for 20+ districts, had to put out an notice that can't provide notification all of a sudden due to cowin in every district.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF2YW5IHXZ5SIXHSFTDTO6VI3ANCNFSM45KQTE7A>.
|
|
The issue seems to be because AWS cloudfront error? Has it got anything to do with invalidating tokens? Many Telegram bots are still working fine, wondering how bot scripts are handled |
|
Yes it’s cloudfront but the policy is managed by admins, that’s why we are reporting it. Bot’s are nothing but codes like ours, if you follow closely, they will miss out on updates too.
This affects all using public ip.
Companies with protected apis may have good access.
… On 22-May-2021, at 7:29 PM, klpradeepkl ***@***.***> wrote:
The issue seems to be because AWS cloudfront error? Has it got anything to do with invalidating tokens? Many Telegram bots are still working fine, wondering how bot scripts are handled
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZFYSAVRZQDEJ4HAL65TTO62EVANCNFSM45KQTE7A>.
|
|
On a different note, seems like a lot of us are running notification/alert systems via web/email/telegram etc. Can we have a list created of such systems that are already in place (if we don't have one already)? This will not only help the people out there who want to use the systems, but also reduce the redundant systems which in turn can reduce the number of hits API must be getting. Just a thought! |
|
Guys The api's are throttled now like 30 requests per 5 minutes or so. |
|
I think not, the errors have stopped since 7 pm, I was using 6 second delay, which gives me 60 requests per 5 min.
It’s intermittent.
… On 22-May-2021, at 7:56 PM, Vishal ***@***.***> wrote:
Guy's The api's are throttled now like 30 requests per 5 minutes or so.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF7NCTIFUDXB7K4VCVDTO65LDANCNFSM45KQTE7A>.
|
|
These errors have started again.
May be it’s time to shutdown such a helpful service that I created over weeks.
… On 22-May-2021, at 7:41 PM, Ricky Bhatia ***@***.***> wrote:
On a different note, seems like a lot of us are running notification/alert systems via web/email/telegram etc. Can we have a list created of such systems that are already in place (if we don't have one already)? This will not only help the people out there who want to use the systems, but also reduce the redundant systems which in turn can reduce the number of hits API must be getting. Just a thought!
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF4WGSBZX574VZTQJLDTO63RVANCNFSM45KQTE7A>.
|
|
Please try increasing the delay.. r u running it on cloud? |
|
Slots will vanish with so much delay added.. This is no way of running any tracker, defeats all purpose of tracking.
Waste of time for users.
To run this with max delay one can only run 1-2 districts.
Correct me if I am wrong.
However as I type this, the errors have disappeared. So it’s intermittent as I have understood.
… On 22-May-2021, at 8:18 PM, Vishal ***@***.***> wrote:
try increasing the delay.. r u running it on cloud?
|
|
As per my analysis the check implemented is like if requests_in_last_five_minutes > 25-30 then throw error. |
|
So you are saying that 30 requests are safe?
However 30 requests with a public api hold no meaning, as I said, slots vanish even, over and above this api is cached by 30 min.
… On 22-May-2021, at 8:26 PM, Vishal ***@***.***> wrote:
As per my analysis the check implemented is like if requests_in_last_five_minutes > 25-30 then throw error.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF72WAAIAINW2ENDRL3TO7AZJANCNFSM45KQTE7A>.
|
|
Yes 30 reqs per IP seems safe. If you can deploy multiple cloud instances and sync them then it is workable. |
|
You mean multi ip cloud instances?
Sync them how? Or why sync?
Please explain.. I have GCP.
… On 22-May-2021, at 8:41 PM, Vishal ***@***.***> wrote:
Yes 30 reqs per IP seems safe. If you can deploy multiple cloud instances and sync them then it is workable.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZFZ6APQHCVYDL5Q7HPDTO7CTZANCNFSM45KQTE7A>.
|
|
No, Multiple cloud working together in sync. like one cloud is sending request at 8:50:00 then other one will send at 8:50:05 and whoever gets the slot triggers the telegram msg. |
|
Sorry this is not clear to me.. is this a good practice?
Can I read about it anywhere, with regards to google cloud.
… On 22-May-2021, at 8:48 PM, Vishal ***@***.***> wrote:
o, Multiple cloud working together in sync. like one cloud is sending request at 8:50:00 then other one will send at 8:50:05 and whoever gets the slot triggers the telegram msg.
|
|
Yes, why not this is how scaling works. |
|
I have created a vaccine tracker using "calendarByPin" which would give people information of vaccine depending on their pin code I hosted it on the digital ocean Bangalore server and it worked only for a max of 30 min to 1 hour since then it's throwing 403 error. It's been 5-6 hrs. Do you guys have any solutions? You can check the error yourself at https://vaccine.codewithbishal.com/ Update: I have shifted to AWS now and it's working... |
|
Now it seems to be working fine guys. |
|
@vishwakarmad1999 not working for me..I think I have to shift to GCP or AWS |
|
I am on gcp… that will not help.. its cloud front policy. Errors have reduced as the day has passed.
Use bearer token method, its better.
… On 23-May-2021, at 12:44 AM, Bishal ***@***.***> wrote:
@vishwakarmad1999 <https://github.com/vishwakarmad1999> not working for me..I think I have to shift to GCP or AWS
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZFYX3VNHPIYOG3UEFJTTO77AZANCNFSM45KQTE7A>.
|
|
@CodeWithBishal , could you check it again? |
|
@nikhiltalreja can you please explain? |
|
@vishwakarmad1999 yes I have checked just now...still not working |
|
There is a thread on how to work with bearer token - #278
… On 23-May-2021, at 1:55 AM, Bishal ***@***.***> wrote:
@nikhiltalreja <https://github.com/nikhiltalreja> can you please explain?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#313 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AUDIZF5LQC2PT3I5S75ZQHLTPAHJ7ANCNFSM45KQTE7A>.
|
|
@nikhiltalreja yes, but how do we even register for prod server API keys |
|
Check the thread.
…On Sun, 23 May 2021 at 2:06 AM, Bishal ***@***.***> wrote:
@nikhiltalreja <https://github.com/nikhiltalreja> yes, but how do we even
register for prod server API keys
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#313 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUDIZF46V435KUAPT5KVG5LTPAIVDANCNFSM45KQTE7A>
.
|
|
Hello folks |
|
It it just me or everyone is facing this issue. And this has become very frequent now. |
|
Looks like no one is interested in fixing it. As an api provider it’s their
job and duty to provide support.
So many are dependent on our tracker for obvious reasons that Cowin and
arogya don’t provide a notification module.
Many have received timely vaccine due to my tracker, so api coders are
adding value.
…On Sun, 23 May 2021 at 12:44 PM, Divya Vishwakarma ***@***.***> wrote:
It it just me or everyone is facing this issue. And this has become very
frequent now.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#313 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUDIZF7NFXY5AJURDRVIRD3TPCTLZANCNFSM45KQTE7A>
.
|
|
I am also facing the same issue, @cowinapi please help. |
|
I've been following this thread ever since its inception and share the same plight. The @cowinapi team is just being tight-lipped, for reasons they'd know best, but this isn't done. I still remember how active their replies on TTL and rate limits were, back in the day. As regards the portal, I get a mouthful of a message - "Your transaction didn't go through, please try again later." while attempting to manually secure a slot. The behavior is peculiar with their APIs as well as their web portal. Developers and users suffer alike. If they continue to be as blase, I'd propose a bunch of us developers write up on social media about the limitations we're working within. At least the end users of our services would know why our platforms are failing. |
|
If there is no reply or fix then it’s best we shut our service. Maybe they
are new to issuing apis to public and now can’t handle the load.
Maybe they used us to do all the debugging and improvisation they wanted
for themselves.
This has gone From open source to closed source.
…On Mon, 24 May 2021 at 6:40 PM, Rahul Pillai ***@***.***> wrote:
I've been following this thread ever since its inception and share the
same plight. The @cowinapi <https://github.com/cowinapi> team is just
being tight-lipped, for reasons they'd know best, but this isn't done. I
still remember how active their replies on TTL and rate limits were, back
in the day.
As regards the portal, I get a mouthful of a message - *"Your transaction
didn't go through, please try again later."* while attempting to manually
secure a slot. The behavior is peculiar with their APIs as well as their
web portal. Developers and users suffer alike.
If they continue to be as blase, I'd propose a bunch of us developers
write up on social media about the limitations we're working within, also
citing the team's nonchalance. At least the end users of our services would
know *why* our platforms are failing.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#313 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUDIZFYMYDUKUZH5PDEBA53TPJF4PANCNFSM45KQTE7A>
.
|
|
Agree. Nonetheless, If not for their APIs, I'd at least request the @cowinapi team to fix the foibles of their web portal. Errors messages like "Your transaction didn't go through" or "Something is not right" doesn't suffice. |
|
Well guys, looks like the @cowinapi team doesn't give a heck about our application and now that the government has also enabled on-spot registration for adults then we can shut our application off because running it in the midst of 403 just irritates a lot. |
|
On Spot? Walkins?
I am getting disappointing reports from my friends in Bangalore on km long large queues for centers with just 40-50 slots.
…-Nikhil
On 24-May-2021, at 7:04 PM, Divya Vishwakarma ***@***.***> wrote:
Well guys, looks like the @cowinapi team doesn't give a heck about our application and now that the government has also enabled on-spot registration for adults then we can shut our application off because running it in the midst of 403 just irritates a lot.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or unsubscribe.
|
|
Centers seem to be kicking off their own rudimentary token-based registration system. I can say so for Pune, at least. Defeats the whole purpose of the web portal but it is what it is. |
|
So it’s a luck by chance thing back again.
I support the online booking system as it keeps away disappointment of
waiting, and corrupt practices of the VIP culture.
Yes, sadly it is what it is.
…On Mon, 24 May 2021 at 7:38 PM, Rahul Pillai ***@***.***> wrote:
Centers seem to be kicking off their own rudimentary token-based
registration system. I can say so for Pune, at least. Defeats the whole
purpose of the web portal but it is what it is.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#313 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AUDIZF5OFL6WAY27QKSVINLTPJMWTANCNFSM45KQTE7A>
.
|
|
Clearly the need this #174 They are force to put such stringent limits because people are creating redundant trackers, they can't keep up with the demand so they have now started to tighten up the limits. This is not a sustainable model for them, they need to publish data when slots are available not vice-versa where plethora of clients are basically spamming their public endpoint. IMHO this was never a saleable solution from the get go. |
|
Then why haven’t we heard of any official notification service? Making trackers doesn’t take much.
Govt of India has a policy for api in many areas, you can check apisetu.
Yes this is unmanageable solution only if they don’t fix the 403 errors.
…-Nikhil
On 26-May-2021, at 3:34 PM, Abhijeet Singh Rathore ***@***.***> wrote:
Clearly the need this #174
They are force to put such stringent limits because people are creating redundant trackers, they can't keep up with the demand so they have now started to tighten up the limits.
This is not a sustainable model for them, they need to publish data when slots are available not vice-versa where plethora of clients are basically spamming their public endpoint.
IMHO this was never a saleable solution from the get go.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or unsubscribe.
|
I am trying to access the API but getting 403 error every 5 mins.
The original API limit was 100/5 mins, and I am sending 75 requests per 5 mins. Getting error for 60 requests per 5 mins too.
Can anyone confirm if there is any change in limit?
The text was updated successfully, but these errors were encountered: