-
Notifications
You must be signed in to change notification settings - Fork 39
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Seeking configuration for Read timeout to XXX after 120000 ms #880
Comments
you just have to go in your service settings page and change the value for the timeouts. Tell me if everything works fine ! |
I have modified that configuration, with different value for each fields to identifiy them if they were trigger. But I got the the same timeout in my log :
That why I was trying the solution with config file of Otoroshi. Note : I have try to curl the URL that otoroshi call from the commande line on Otoroshi pod (Docker) and it's respond a code HTTP 200 (but with more than 120 sec). I have also try to desactive the section circuit-breaker to let it pass as it. But, I got a Bad Gateway after 60 sec. |
That's weird ! Is it possible for you to try the same thing with the latest 1.5.0 version (https://github.com/MAIF/otoroshi/releases/tag/v1.5.0-alpha.18) and see if it behaves the same ? |
hey @pkolodziejczyk i fixed a bug (#883) that could cause the issue (not 100% sure as I can't reproduce it). some things you can try to fix your bug
|
hey @pkolodziejczyk you can try https://github.com/MAIF/otoroshi/releases/tag/v1.5.0-alpha.19 and tell me how it behaves |
I will try your solution. Question : why the restart of the otoroshi instance ? Because, I didn't restart otoroshi after saving the new configuration. for my others modifications. For the test of the new version, if my schedule allow it, I will test the image 1.5.0-alpha.19-jdk8 |
When launching the image 1.5.0-alpha.19-jdk8 I will try with the 1.5.0-alpha.19-jdk11 |
container_1.5.0-alpha.19-jdk11.relyens.1.1.log Look like, the new "checking otoroshi updates" is making otoroshi crash in my environnement. My pods don't have acces to the net. |
@pkolodziejczyk the restart is because of a caching bug that should be fixed by now. You can try the fixed version by trying the following docker image: If you still have issues, can you try to add the following env. variable:
and show me the full log ? About the JDK8 version, I wasn't aware of that one. I will drop JDK8 support a little bit sooner that I expected (september 2021) I guess. About update check crashing the image, I shouldn't be the case. The log can be a little bit raw but everything should continue to work. |
Will test it on Monday and give you the results. Thanks for the time. |
As there is no more comments, I'm closing this one, but it can be reopened |
Sorry for the late reply. I have tested with the docker image 1.5.0-beta.1 And still have :
Still don't understand where come from that 120000 ms I haven't check if there is new value to fill the interface in 1.5.0 Edit : Made a new test with the custom timeout settings part filled
Still that 120000 ms value in place. |
Ok, so I guess the issue is fixed, you now have an issue with otoroshi protocol @pkolodziejczyk right (#898) ? |
No. The timeout at 120 000ms is here. The other issue concern an other call on an other service. (#898) I just wait to have the next testable version with (#897) to check if the configuration is right in 1.5. All value are at 300 00xms in 1.4.22 and the log show those values. before the time out :
|
Can you remind me your client configuration ? |
Ok, the log about 120000ms is still there, but otoroshi send a response back before 120000ms right ? it should return after 30008 ms |
@pkolodziejczyk i found the issue, totally forgot about this one. just set env var. |
Thanks, will try it. |
I have try with the 1.5.0-beta.1
The otoroshi response after 1min 0.42sec :
Downservice log :
|
More log with 1.5.0-beta.2 I got a timeout at 120 secondes.
But, I am not sur if it's otoroshi or our proxy on front of it. (I asked my OPS Teams to check it.) Note : the new interface with the direct access to APIKEY is just nice. |
@pkolodziejczyk everything should be fixed with beta.3 that is out now |
Same behaviour from Otoroshi version 1.5.0-beta.3 A second call after 60 secondes and timeout at 120 secondes
My downstream service didn't responded in time. So, the warn is logical. But, should be alone. Since "Client attempts" is set to 1and "C.breaker retry delay" is set to 300005 delay ms Second Test : But, I am not sur if the first warn is not from the previous test.
I am doing more testing. But there is something with the double call, that i don't explain with the configuration. |
@pkolodziejczyk I have trouble to reproduce the issue. Here is a service descriptor that call a service that generates long responses. In the video I call the service on a beta.3 version timeout.movservice-descriptor-my-service-1628265802741.json.zip |
Here is my test : service-descriptor-TaskServiceSynapse-1628507477222.zip The timeout is probably on my OPS Teams side. But, I am less sure about the double trigger. |
OTOROSHI_880_TEST_02.mp4Better format to look here. And with useAkkaHttpClient true. |
I got a response for my Operation Team. Thanks a lot for your support and your product. |
Hi Mathieu, I'm one of the architects working with Patrick Did you have an idea when you will release a new version with this correction ? Depending on the date we will decide if we install the beta or wait for the release. |
Hi @pkolodziejczyk and @KpTn6974 sorry for the delay, I was in vacations. I'm glad the issue is fixed on your side. I understand that you don't want to deploy a beta version in production. I have no specific date in mind about the release as I am quite busy right now, but I guess it will be before the end of 2021. However, I can assure you that the beta version are quite stable and should not introduce breaking stuff until the release, only bug fixes and a new documentation. |
Hi @mathieuancelin, Thanks for your reply. And no worries for the delay, vacations are a good thing also ;-) We will discuss it internally and I think deploy the beta on testing and preproduction environment to validate it. Before going to production. |
Hi,
Version of otoroshi : 1.4.22 as docker image (1.4.22-jdk8)
I have a timeout that i don't know how to change the default configuration :
As I understand it. It's configuration of playframework that need to be change. Since otoroshi use it.
I have try the solution proposed here :
playframework/play-ws#202
and put the configuration in otoroshi configuration but it failed.
The text was updated successfully, but these errors were encountered: