Ok so I needed to add around 22 trillion cycles but then was able to run:
dfx canister update-settings --compute-allocation 1 data_canister --ic
And currently I have no ingress expiry error:
Will see how it goes.
after this command were you issues fixed or not?
@jamesbeadle
I dunno, this works on one of my canisters through the candid UI but I need to redeploy all and see if it was just computationally expensive canister or every canister I need has to be set with this, including the frontend canisters…
So far if you set this on canister you don’t get the ingress expiry error through the candid UI as i did before.
once you check it on all canister then please share your feedback also, thanks
Sorry you are experiencing issues! Could we please in all reports include
- canister id
- subnet id
- exact error message
- which agent are you using and which version
so we can appropriately address the issues?
@jamesbeadle which canister id did you see problems with? Was it addressed with the compute allocation?
2 Likes
I came across this solution as well, but it typically implies we’re bidding for compute resources, which will significantly increase the cycles burn rate? As far as I know, setting “compute-allocation 1” still consumes cycles, when idle? Clarity on compute-allocation
I am very concerned about how we manage our projects if errors arise on ICP after updates. My project has been stuck for more than a month due to an update, and I am continuously paying development costs.
@Manu @jennifertran, please look into this project. Our platform is based on content publishing and community contribution, and if our community is unable to contribute, it defeats the purpose of our platform.
Regards
Hello @HinzaAsif,
-
Is there a specific error that you are running into?
-
When did this start occurring?
-
What is the ID of the canister that is error is appearing in?
@Alishan is the person working on the project, and he created this post 9 days ago.
@jennifertran, I would really appreciate it if you could resolve this issue. Our production work is on hold, and the whole team is stuck.
Regards
1 Like
Hello @Alishan,
I know that you sent me the subnet ID previously. However, can you please also send me the additional information? Please feel free to DM.
- canister id
- Which agent are you using and which version?
- Are you setting any custom expirations on messages?
@jennifertran
canister= ogidd-6aaaa-aaaal-qcezq-cai
agent version is= “@dfinity/agent”: “^0.18.1”,
and i am not setting any custom expirations
1 Like
@jennifertran I’d like to mention that our canisters: rxqpk-hyaaa-aaaam-actvq-cai and pvvqb-hiaaa-aaaam-ab6xq-cai have the same code as the canister in question: 7uioq-vyaaa-aaaal-ac6ea-cai. but the later is on a different subnet. This is the subnet in which our working canisters are : https://dashboard.internetcomputer.org/subnet/4ecnw-byqwz-dtgss-ua2mh-pfvs7-c3lct-gtf4e-hnu75-j7eek-iifqm-sqe
and this is the subnet in which our canister with issue is : https://dashboard.internetcomputer.org/subnet/6pbhf-qzpdk-kuqbr-pklfa-5ehhf-jfjps-zsj6q-57nrl-kzhpd-mu7hc-vae. there is a very visible difference between the number of canister in the two subnets: 23’915 v 6’281.
1 Like
Thank you to @Alishan, @Mysticx, and any others who sent me the info via DM. It is known that the canisters that you are having issues with are on subnets with heavy compute load.
Here is an update on our end:
2 Likes
Also which one are you using? The Javascript one?
Yes, i am using JavaScript .
1 Like
Hi @Alishan
If you are really using the JS agent it means you are using 1 year old agent.
Is there chance that you upgrade to the most recent one ?
1 Like
I have two projects and both are using same same agent one is working and other is not, and there canisters are on different subnet , canister on which i am facing this error is on such subnet which has 24k+ canisters and the other canister which which is working fine is on such subnet that have only 6k+ canisters. I can think it might be due to load of canisters,
And the solution provided by @jennifertran i think temporary because if i migrate my canister to another subnet after one or two years later load increased also on that subnet again i will face this issue, and if increase computational rate/ cycles burn rate , it may also cause of issue when others people also increase it,
I am looking a professional way to sort it out.
Thank @rumenov
Hello @Alishan
Have you reviewed this post in detail? Subnets with heavy compute load: what can you do now & next steps
If you are not looking to migrate the canisters to a different subnet (which might be overloaded in the future), you can try to use compute allocation.
You can also wait for the long term solutions that we are looking to implement.
Thank @jennifertran.
I am using next js and my website is article base and their seo really needed. For that purpose i have to generate static pages to add metadata because crawler didn’t pick metadata of clients page mean if i add metadata to the page after loading JavaScript, for that reason i have to generate static pages each time and deploy them again on IC . I have to repeat this process once in a day so that articles which were created on website to be generated as static.
What nextjs do
it get all ids of my articles and generate static page of each article or id, during this process it send query to canister to get data of article by id and generate its page. As this process really fast and send to many requests to canister for example if there are 500 articles, it will call a canister 200 time which i think not a good approach. Do you suggest me alternative solution for seo. What i do so that my problem of again and again next js build generation and deployment to be sorted.
Reason to do this only for SEO please suggest me the best solution which were already implemented and working. @jennifertran @rumenov @peterparker and other all people,
Thanks in advance.