2019-07-18 02:07:54 UTC - chris: Guys, i have a question.... the script `wsk -i package list /whisk.system` why the package list is empty ? the openwhisk is installed by docker-compose `make quick-start` https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563415674026500 ---- 2019-07-18 02:12:50 UTC - Rodric Rabbah: hi Chris https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563415970026700 ---- 2019-07-18 02:13:08 UTC - Rodric Rabbah: dont think the packages are installed by default - try `make add-catalog` first https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563415988027200?thread_ts=1563415988.027200&cid=C3TPCAQG1 ---- 2019-07-18 02:13:47 UTC - Rodric Rabbah: by web api do you mean a web action https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563416027027400?thread_ts=1563381349.022500&cid=C3TPCAQG1 ---- 2019-07-18 02:16:15 UTC - chris: @Rodric Rabbah oh my god!!!! Love u so much :heart: i just found it in the gihub ....
These are not included by default with the devtools make quick-start command. If you want to install these packages, run the following make command. `make add-catalog` https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563416175028400?thread_ts=1563415988.027200&cid=C3TPCAQG1 ---- 2019-07-18 02:20:27 UTC - Rodric Rabbah: :+1: https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563416427028600?thread_ts=1563415988.027200&cid=C3TPCAQG1 ---- 2019-07-18 10:21:40 UTC - Satwik Kolhe: Please correct me if I am wrong - I continuously had issues with apigateway -- ``` ==> /var/log/api-gateway/management.log <== 2019/07/18 09:36:30 [error] 38#0: *40 lua entry thread aborted: runtime error: /etc/api-gateway/scripts/lua/management/lib/tenants.lua:130: attempt to compare nil with number stack traceback: coroutine 0: /etc/api-gateway/scripts/lua/management/lib/tenants.lua: in function 'getTenantAPIs' /etc/api-gateway/scripts/lua/management/routes/apis.lua:62: in function 'getAPIs' /etc/api-gateway/scripts/lua/management/routes/apis.lua:41: in function 'requestHandler' access_by_lua(management_apis.conf:36):5: in function <access_by_lua(management_apis.conf:36):1>, client: 10.44.0.0, server: management_gw, request: "GET /v2/23bc46b1-71f6-4ed5-8c54-816aa4f8c502/apis?limit=undefined&skip=undefined&basePath=%2F HTTP/1.1", host: "10.110.90.93:9000" ``` TO which I ended up checking the openwhisk/apigateway image tag - which is.. ``` root@ow-master-1:~# kl get pods/slsow-apigateway-7b7879b7bb-vr64v -o yaml -n slsow | grep image image: openwhisk/apigateway:0.10.0-incubating imagePullPolicy: IfNotPresent image: openwhisk/apigateway:0.10.0-incubating imageID: <docker-pullable://openwhisk/apigateway@sha256:a399e8dc0efcf430178cf3b8096a0ddd1b8ba10d6bc4a7ea742eb03958ad745b> ``` And possible fixes for the issue on <http://github.com/apache/incubator-openwhisk-apigateway|github.com/apache/incubator-openwhisk-apigateway> - Where I found that this issue (<https://github.com/apache/incubator-openwhisk-apigateway/pull/343/files>) is already reported and fixed and merged into master. The build process is failing for master (which is also seen in TravisCI) - and these changes are not in openshisk/apigateway:0.10.0-incubating. I have made necessary changes in my running image of apigateway! How can this be solved permanently? apache/incubator-openwhisk-apigateway --branch 0.10.0-incubating >> is not available on github! https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563445300042300?thread_ts=1563445300.042300&cid=C3TPCAQG1 ---- 2019-07-18 10:27:54 UTC - Jiang PengCheng: sorry, but I didn't get what you want, it seems that you need to get the code of `0.10.0-incubating` and make some changes on it? then there is a tag `0.10.0-incubating` on github https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563445674042700?thread_ts=1563445300.042300&cid=C3TPCAQG1 ---- 2019-07-18 10:36:52 UTC - Satwik Kolhe: Hey @Jiang PengCheng, The image that is used (openwhisk/apigateway:0.10.0-incubating) does not have the necessary changes that are mentioned here (<https://github.com/apache/incubator-openwhisk-apigateway/pull/343/files>) - which are merged in master branch. I tried to build the apigateway image from this master branch, but it failed - same is signified from TravisCI logs. Because of all this I had to make temporary changes in my deployment of openwhisk, where I updated the changes mentioned here (<https://github.com/apache/incubator-openwhisk-apigateway/pull/343/files>) How can we make this permanent? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563446212042900?thread_ts=1563445300.042300&cid=C3TPCAQG1 ---- 2019-07-18 10:56:54 UTC - Satwik Kolhe: @Jiang PengCheng Got it - Thanks, I mixed branch and tags, will pull 0.10.0-incubating and make necessary changes to this stable version of apigateway https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563447414043100?thread_ts=1563445300.042300&cid=C3TPCAQG1 ---- 2019-07-18 11:53:17 UTC - Michael Schmidt: I actually found this was just a mistake by my k8s auto service discovery https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563450797043300?thread_ts=1563381349.022500&cid=C3TPCAQG1 ---- 2019-07-18 13:56:10 UTC - Roberto Santiago: I am still trying to debug an issue I am having with my invokers failing (kubernetes deploy). What I am seeing is that that the invokers are losing connection with the Docker Daemon. Specifically, if I ask for the logs from the invoker I get errors that look like the following: https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563458170044900 ---- 2019-07-18 13:56:13 UTC - Roberto Santiago: ```[2019-07-17T22:45:58.164Z] [ERROR] [#tid_sid_invokerHealth] [DockerClientWithFileAccess] info: command was unsuccessful, code: 125 (unsuccessful), stdout: , stderr: /usr/bin/docker: An error occurred trying to connect: Post <http://%2Fvar%2Frun%2Fdocker.sock/v1.24/containers/create?name=wskgke-playground-invoker-node-pool-0-d9ad70ba-qvrd16_68_whisksystem_invokerHealthTestAction0>: read unix @->/var/run/docker.sock: read: connection reset by peer. [2019-07-17T22:45:58.166Z] [ERROR] [#tid_sid_invokerWarmup] [DockerClientWithFileAccess] info: command was unsuccessful, code: 125 (unsuccessful), stdout: , stderr: /usr/bin/docker: An error occurred trying to connect: Post <http://%2Fvar%2Frun%2Fdocker.sock/v1.24/containers/create?name=wskgke-playground-invoker-node-pool-0-d9ad70ba-qvrd16_69_prewarm_nodejs10>: read unix @->/var/run/docker.sock: read: connection reset by peer.``` https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563458173045100 ---- 2019-07-18 13:56:58 UTC - Roberto Santiago: I've seen a number of reasons why this type of error can happen including things like being out of memory or disk space. But I don't have a smoking gun yet. https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563458218045900 ---- 2019-07-18 13:59:26 UTC - Roberto Santiago: Any ideas where to look next? Also, if anyone is running OW through kubernetes deploy could you shell into one of your invoker pods and run `docker ps`? Right now I see quite a large number of containers (like 70-100 containers). Is it the same for you? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563458366047800?thread_ts=1563458366.047800&cid=C3TPCAQG1 ---- 2019-07-18 16:19:59 UTC - Michele Sciabarra: just doing top what says https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563466799048000 ---- 2019-07-18 16:20:11 UTC - Michele Sciabarra: looks like simply docker crashed https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563466811048400 ---- 2019-07-18 16:21:35 UTC - Michele Sciabarra: are you able to ssh to the docker node and do a docker ps? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563466895048900 ---- 2019-07-18 16:21:50 UTC - Michele Sciabarra: and a top to check memory usage? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563466910049200 ---- 2019-07-18 23:43:00 UTC - yihuaf: Is it normal for `ow-utils` image to take more than 20-30 minutes to build? `./gradlew distDocker` stuck on this step and took more than 840 seconds with out response. https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563493380050400 ---- 2019-07-18 23:51:22 UTC - Rodric Rabbah: no that seems very wrong https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563493882050700 ---- 2019-07-18 23:52:04 UTC - Rodric Rabbah: note that the gradlew build retries on failure, to compensate for network hiccups that might happen did you inspect the log output for errors? https://openwhisk-team.slack.com/archives/C3TPCAQG1/p1563493924051400 ----
