You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spark.apache.org by Marcelo Vanzin <va...@cloudera.com.INVALID> on 2018/08/13 18:49:17 UTC

Re: Cleaning Spark releases from mirrors, and the flakiness of HiveExternalCatalogVersionsSuite

On this topic... when I worked on 2.3.1 and caused this breakage by
deleting and old release, I tried to write some code to make this more
automatic:

https://github.com/vanzin/spark/tree/SPARK-24532

I just found that the code was a little too large and hacky for what
it does (find out the latest releases on each branch). But maybe it
would be worth to do that?

In any case, agree with Mark that checking signatures would be good, eventually.


On Sun, Jul 15, 2018 at 1:51 PM, Sean Owen <sr...@gmail.com> wrote:
> Yesterday I cleaned out old Spark releases from the mirror system -- we're
> supposed to only keep the latest release from active branches out on
> mirrors. (All releases are available from the Apache archive site.)
>
> Having done so I realized quickly that the HiveExternalCatalogVersionsSuite
> relies on the versions it downloads being available from mirrors. It has
> been flaky, as sometimes mirrors are unreliable. I think now it will not
> work for any versions except 2.3.1, 2.2.2, 2.1.3.
>
> Because we do need to clean those releases out of the mirrors soon anyway,
> and because they're flaky sometimes, I propose adding logic to the test to
> fall back on downloading from the Apache archive site.
>
> ... and I'll do that right away to unblock HiveExternalCatalogVersionsSuite
> runs. I think it needs to be backported to other branches as they will still
> be testing against potentially non-current Spark releases.
>
> Sean



-- 
Marcelo

---------------------------------------------------------------------
To unsubscribe e-mail: dev-unsubscribe@spark.apache.org