Twinkle, I think while clearing a task u can press upstream and downstream buttons, this will clear all tasks of that dag run and will eventually re-run whole dag.
On Fri, Oct 21, 2016 at 4:29 PM, twinkle sachdeva < twinkle.sachd...@gmail.com> wrote: > @Shin: Nice find. > > @Alex : Thanks for sharing the thread. I think it will be a nice > enhancement for the teams, specially switching from the technical stack > where it has been a feature. > > On Fri, Oct 21, 2016 at 4:05 PM, <wget.n...@gmail.com> wrote: > > > Nope, that’s all Airflow can do right now per UI. We had a discussion > > about in July: > > > > https://mail-archives.apache.org/mod_mbox/incubator- > > airflow-dev/201607.mbox/browser > > > > > > > > cheers, > > > > --alex > > > > > > > > -- > > B: mapredit.blogspot.com > > > > > > > > *From: *twinkle sachdeva <twinkle.sachd...@gmail.com> > > *Sent: *Friday, October 21, 2016 12:23 PM > > > > *To: *wget.n...@gmail.com > > *Cc: *dev@airflow.incubator.apache.org > > *Subject: *Re: Regarding force running the whole graph from UI > > > > > > > > > > > > It is not triggering the rest of the graph. > > > > It leads to execution of first task only. > > > > > > > > Are there some other flags / Layout change which needs to be done? > > > > > > > > > > > > On Fri, Oct 21, 2016 at 3:45 PM, <wget.n...@gmail.com> wrote: > > > > In Airflows main window, click at the DAG and you see the graph view. > > Clicking on one of the tasks should open a window with the possibility to > > “Run” that task. When you use the first task, then the whole DAG will be > > executed (given no errors occur). > > > > > > > > I hope that’s what you’re looking for 😊 > > > > > > > > Cheers, > > > > --alex > > > > > > > > -- > > B: mapredit.blogspot.com > > > > > > > > *From: *twinkle sachdeva <twinkle.sachd...@gmail.com> > > *Sent: *Friday, October 21, 2016 12:09 PM > > > > > > *To: *wget.n...@gmail.com > > *Cc: *dev@airflow.incubator.apache.org > > *Subject: *Re: Regarding force running the whole graph from UI > > > > > > > > Hi Alex, > > > > > > > > We are using Celery Executor. > > > > I am not able to identify how can i do it from UI. Is it possible? > > > > > > > > Command line is awesome. > > > > > > > > Regards, > > > > Twinkle > > > > > > > > > > > > On Fri, Oct 21, 2016 at 3:36 PM, <wget.n...@gmail.com> wrote: > > > > Hey Twinkle, > > > > > > > > Triggering DAG’s per UI works only by using the CeleryExecutor in > > airflow.cfg, probably with the mesos one, too. Both are execute tasks > > remotely. > > > > > > > > P.S: Personally I favorite LocalExecutor and trigger DAG’s per CLI. > > > > > > > > Cheers, > > > > --alex > > > > > > > > -- > > B: mapredit.blogspot.com > > > > > > > > *From: *twinkle sachdeva <twinkle.sachd...@gmail.com> > > *Sent: *Friday, October 21, 2016 11:58 AM > > *To: *wget.n...@gmail.com > > *Cc: *dev@airflow.incubator.apache.org > > *Subject: *Re: Regarding force running the whole graph from UI > > > > > > > > Hi Alex, > > > > > > > > I mean the complete DAG. > > > > In technical terms, being able to run 'trigger_dag' command from the UI. > > > > > > > > Regards, > > > > Twinkle > > > > > > > > > > > > On Fri, Oct 21, 2016 at 2:47 PM, <wget.n...@gmail.com> wrote: > > > > Hey, > > > > > > > > Something like this: > > > > https://pythonhosted.org/airflow/cli.html ? > > > > > > > > What do you mean with the whole graph? The complete DAG, or a task from a > > specific DAG? > > > > > > > > --alex > > > > > > > > -- > > > > B: https://mapredit.blogspot.com > > > > > > > > *From: *twinkle sachdeva <twinkle.sachd...@gmail.com> > > *Sent: *Friday, October 21, 2016 8:30 AM > > *To: *dev@airflow.incubator.apache.org > > *Subject: *Regarding force running the whole graph from UI > > > > > > > > Hi, > > > > > > > > Is there a way by which we can force run the whole graph from Airflow UI? > > > > > > > > Also, is there any documentation available regarding all the options > which > > > > are there in the pop-up dialog for running the graph? > > > > > > > > Thanks & Regards, > > > > Twinkle > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >
