You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@openwhisk.apache.org by OpenWhisk Team Slack <ra...@apache.org> on 2019/11/02 09:15:26 UTC
[slack-digest] [2019-11-01] #general
2019-11-01 09:48:18 UTC - Bill Zong: Don’t get any experiences with it.
You might be better to [post an issue](<https://github.com/apache/openwhisk-runtime-python/issues>) for formal discussions.
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572601698175900?thread_ts=1572561821.175400&cid=C3TPCAQG1
----
2019-11-01 09:58:30 UTC - Adrian Schuepbach: Hi Klaus, in my case it seems that somehow something got updated partially (for whatever reason) and this casued the problem.
After manually installing Kubernetes 1.16 and manualyl deploying the newest OpenWhisk, it worked again.
One problem, which I still have, is that actions sometimes take too long to download the docker image and launch the container (>1min default timeout). In this case, I still see the error `The action did not initialize or run as expected. Log data might be missing.`. I think the behaviour was different in older versions.
It seems that in older OpenWhisk versions, downloading the image and launching the container was not part of the max allowed time to run the action, but now it seems that it is.
I extended the timeout to 5 minutes for all of my actions and then it works fine.
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572602310176100?thread_ts=1571750437.034300&cid=C3TPCAQG1
----
2019-11-01 13:45:47 UTC - Pepi Paraskevoulakou: i invoked again the action and i got a result like this
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572615947176700
----
2019-11-01 13:46:39 UTC - Pepi Paraskevoulakou: afterwards i checked the activation logs and i saw that. Do i need to add time in timeout?
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572615999177200
----
2019-11-01 13:56:21 UTC - Rodric Rabbah: when you invoke a sequence, the internal scheduler waits for a predetermined time for the response to be ready
what the error message states is that the internal response took too long, this isn’t a value you can change at user level you have to modify the control plane and rebuild/deploy. *it might help if you set the individual action timeout to be the max.*
suggest running again, it might work, if for example this is a docker action and there was a lot of time spent fetching the docker image.
you’d have to inspect the individual activations to see which timed out and why
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572616581179400
----
2019-11-01 14:07:24 UTC - Pepi Paraskevoulakou: and what about that? i tried to invoke again the action
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572617244179700
----
2019-11-01 14:18:56 UTC - Rodric Rabbah: i have not seen that before
https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1572617936180200
----