We need both options IMO. There are already snapshot logs available. (which is the full log option) We need tailing (streaming) option alongside. Lets get help from UX team on the UX aspects.
Were you able to try out that both tailing option (that means streaming) with giving last number of lines or last 5 (or some number) mins options are possible ? These options are possible with snapshot logs. I have tested and back implementation is done. But I'm not sure whether this options are possible with streaming. If possible I'm +1 to go with last number of lines. Because there are always logs for any last number of lines, but not for all last number of minutes. On Mon, May 23, 2016 at 11:24 PM, Danushka Fernando <danush...@wso2.com> wrote: > Currently we have a page to load all logs. With Streaming I guess we need > to go for tailing certain amount of time could be preconfigured and later > we can let the user configure the time he wants to tail. > There are two options > Keep current full log option and stream last no of lines of the log in a > separate page (That means have two log watching places one for full log and > other for tailing last 5 mins) > or > Keep only tailing logs option. > IMO we need both options. WDYT? > > Thanks & Regards > Danushka Fernando > Senior Software Engineer > WSO2 inc. http://wso2.com/ > Mobile : +94716332729 > > On Fri, May 20, 2016 at 9:53 PM, Kasun De Silva <kas...@wso2.com> wrote: > >> Hi Danushka, >> >> Streaming log option is similar to tailf in unix systems. I guess you >> application is not actively producing any logs. That might be the reason >> you don't get any output or experiencing this. >> >> First I would try following kube command in k8s console against my >> respective pod. >> >> # Begin streaming the logs of a pod >> ./kubectl logs -f <pod_name> >> >> Go to 249 or 251 and try that. You can try more options as mentioned in >> [3]. but make sure you have the "*-f "* option. >> >> Also you can try [1] from you browser against 249 or 251, that is the >> direct api call to the k8s to get streaming of the logs. This is the same >> call that is generated by the fabric8 when we are invoking >> streamRuntimeLogs() >> method. >> >> Also I would have a web app that actively generate logs to test this >> option. Something like [2]. You can just download the war file from [2] and >> deploy it in AppCloud setup you are testing. Then just hit " >> *<launch_url>/logme*" from the browser and the application will start >> generate logs continuously. This kind of application will help you test >> this feature. >> >> >> Finally I would try to see whether if we can pass tailing lines or >> .sinceSeconds() >> or .tailingLines() options to streamRuntimeLogs() builder pattern, >> better if something like .tailingLines(10) then streaming will always >> starts with last 10 lines. In that case we need to modify the back end code >> similar to "getRuntimeLogs(LogQuery query)" method. >> >> This is something I would try at the very end, if everything goes >> smoothly above. >> >> Hope this information is helpful to you regarding this JIRA. Let me know >> if you have any concerns. >> >> >> [1] http://*<k8s_ip>:<port>*/api/v1/namespaces/*<namespace>*/pods/ >> *<pod_name>*/log?container=tomcat-server&*follow=true* >> &previous=false×tamps=false >> >> [2] >> https://github.com/kasundsilva/jsp-loggin/tree/master/sample.log.app.one/target >> >> [3] http://kubernetes.io/docs/user-guide/kubectl/kubectl_logs/ >> >> >> Thanks >> Kasun >> >> *Kasun de Silva* >> Software Engineer >> >> WSO2 Inc <http://wso2.com>*. *|* E*mail : kas...@wso2.com | Mobile: +94 >> 77 794 4260 >> >> > -- *Kasun de Silva* Software Engineer WSO2 Inc <http://wso2.com>*. *|* E*mail : kas...@wso2.com | Mobile: +94 77 794 4260
_______________________________________________ Dev mailing list Dev@wso2.org http://wso2.org/cgi-bin/mailman/listinfo/dev
