Deploy service with existing disk time's out

1

I have existing disk, and when i clear deploy cache and deploy, I get this log and then nothing until timeout.

how do I get my service back up

Jul 13 05:42:53 PM ==> Cloning from GitHub - render-examples/mongodb: Run MongoDB on Render…
Jul 13 05:42:54 PM ==> Checking out commit c3d8d8cdc7246e3afbe644eb3d003588ac0d0b11 in branch master
Jul 13 05:42:55 PM #1 [internal] load build definition from Dockerfile
Jul 13 05:42:55 PM #1 sha256:0d7ba67677a8977911e5f842f649c8d312fea27bed165188ab823f0958951da8
Jul 13 05:42:55 PM #1 transferring dockerfile: 52B done
Jul 13 05:42:55 PM #1 DONE 0.1s
Jul 13 05:42:55 PM
Jul 13 05:42:55 PM #2 [internal] load .dockerignore
Jul 13 05:42:55 PM #2 sha256:1343dbbd171786bef94de3dd3f22cdb610039297d16f246fe326ce947bc5f9bb
Jul 13 05:42:55 PM #2 transferring context: 2B done
Jul 13 05:42:55 PM #2 DONE 0.0s
Jul 13 05:42:55 PM
Jul 13 05:42:55 PM #3 [internal] load metadata for docker.io/library/mongo:4.2
Jul 13 05:42:55 PM #3 sha256:29d5e019e50dcfcb4379342a6141c082bbcb37e8281775de215983cf3bc30a04
Jul 13 05:42:56 PM #3 …
Jul 13 05:42:56 PM
Jul 13 05:42:56 PM #4 [auth] library/mongo:pull token for registry-1.docker.io
Jul 13 05:42:56 PM #4 DONE 0.0s
Jul 13 05:42:56 PM
Jul 13 05:42:56 PM #3 [internal] load metadata for docker.io/library/mongo:4.2
Jul 13 05:42:56 PM #3 sha256:29d5e019e50dcfcb4379342a6141c082bbcb37e8281775de215983cf3bc30a04
Jul 13 05:42:57 PM #3 DONE 1.5s
Jul 13 05:42:57 PM
Jul 13 05:42:57 PM #5 [1/1] FROM docker.io/library/mongo:4.2@sha256:e2edead3bd49a2cb1b47e487378912f8bbe7ce2d708a5d2aec83d0ded526ae5d
Jul 13 05:42:57 PM #5 resolve docker.io/library/mongo:4.2@sha256:e2edead3bd49a2cb1b47e487378912f8bbe7ce2d708a5d2aec83d0ded526ae5d 0.0s done
Jul 13 05:42:59 PM #5 DONE 1.9s
Jul 13 05:42:59 PM
Jul 13 05:42:59 PM #6 exporting to oci image format
Jul 13 05:42:59 PM #6 sha256:69a2560eef4d3ece902c3b5149d142e9bd132f25db0bc9e35b94201534c415d2
Jul 13 05:42:59 PM #6 exporting layers done
Jul 13 05:42:59 PM #6 exporting manifest sha256:016d1d85178246cc9bc47b8790bc1900090f3eb7f0b8516f01bbaac51d0905a2 0.0s done
Jul 13 05:42:59 PM #6 exporting config sha256:ef810eb8a0721dd563fac706a308b61c339e499f78f158d51ebb71f9986ac9e9 0.0s done
Jul 13 05:43:02 PM #6 DONE 4.7s
Jul 13 05:43:02 PM
Jul 13 05:43:02 PM #7 exporting cache
Jul 13 05:43:02 PM #7 sha256:2700d4ef94dee473593c5c614b55b2dedcca7893909811a8f2b48291a1f581e4
Jul 13 05:43:02 PM #7 DONE 0.0s
Jul 13 05:43:03 PM ______________________________
Jul 13 05:43:03 PM < Pushing image to registry… >
Jul 13 05:43:03 PM ------------------------------
Jul 13 05:43:03 PM \ ^^
Jul 13 05:43:03 PM \ (oo)_____
Jul 13 05:43:03 PM ()\ )/
Jul 13 05:43:03 PM ||----w |
Jul 13 05:43:03 PM || ||
Jul 13 05:43:06 PM ______
Jul 13 05:43:06 PM < Done >
Jul 13 05:43:06 PM ------
Jul 13 05:43:06 PM \ ^^
Jul 13 05:43:06 PM \ (oo)_______
Jul 13 05:43:06 PM (__)\ )/
Jul 13 05:43:06 PM ||----w |
Jul 13 05:43:06 PM || ||

2

Hi @happytim,

I took a look at your service and it looks like the next deploy didn’t timeout and the service is live. Let me know if you have any other questions!

3

tyler,

thanks. on my 3rd attempt, it deployed. phew

i guess my question is - if a service fails to deployed with a timeout, why is the existing service not left in place. ie, why did i get downtime because of this?

also, I am nervous now, how can i test to see if the service with re-deploy again (without the risk of downtime on live)

thanks loads for your help

tim

4

The reason you had downtime was due to the persistent disk that is attached to this service.

Adding a disk to a service prevents zero downtime deploys . A new deploy first stops your app to prevent it from writing to disk, and only then brings up the new version so the app can start writing to the disk again.

Unfortunately, there is no way to prevent that downtime. We are working on a fix to make sure that deploys do not timeout, my apologies for the inconvenience.

5

yes, that makes sense, so i see the reason for the downtime. but, do you know what is timing out?

6

Hi @happytim,

We are still investigating the reasons for the timeout but we have seen some instances where the disk has issues mounting to the new service. We are working on a fix for this.

7

Tyler,

I am now in a situation where i have over-sized my service plan - is it currently Standard Plus ($50/m) whereas it should be Starter Plus ($15/m). I did this when i was desperately trying to get my service up following last time I deployed. I am too scared to down-size in case I cannot deploy due to timeout issue.

Last time I deployed a service with a disk, I had over hour of downtime due to timeout issue, and I can’t go through this again!

Please can you advise if the timeout issue will be resolved soon? If it is going to be a long time, will you credit me for the oversizing ($35/m)?

thanks

tim

8

Tyler,

just to let you know. I downgraded from Standard Plus to Starter Plus today without any issues - very fast and without any timeout.

Thanks for your help on this