We originally added the Facebook mirror support back when we were still supporting 0.17, which was only available in archive.apache.org. We were getting download failures from archive.apache.org all the time, since ivy is kind of sensitive and archive.apache.org isn't heavily provisioned.
If www.apache.org/dist always had all of the releases needed by Hive, its availability would probably be fine, but I'm not sure that will always be the case since they're fairly strict about purging all but the most recent releases. JVS On Sep 30, 2011, at 12:22 PM, Ashutosh Chauhan wrote: > I am wondering shall we also add 'official' apache site and make that as > default, since then if there is a problem some one from community can go > take a look. > http://www.apache.org/dist//hadoop/common/hadoop-0.20.204.0/ I believe it > was not done this way earlier because there there was no apache release > which had security in it, now with 204 that problem is solved, shall we move > there? > > Ashutosh > > On Mon, Sep 26, 2011 at 15:08, John Sichi <jsi...@fb.com> wrote: > >> Oops, down again, and the "big downtime" is now, possibly taking days >> before recovery. >> >> JVS >> >> On Sep 26, 2011, at 2:42 PM, wrote: >> >>> It's back up now, but apparently there is some underlying flakiness in >> the array which will need a big downtime to address some time soon. It >> seems like http://archive.cloudera.com/hive-deps is still there...I wonder >> why the automatic fallback didn't work? >>> >>> JVS >>> >>> On Sep 26, 2011, at 1:52 PM, wrote: >>> >>>> Looking into it...last time this happened, we added a Cloudera site as a >> backup...is it not kicking in? >>>> >>>> JVS >>>> >>>> On Sep 26, 2011, at 1:36 PM, Ashutosh Chauhan wrote: >>>> >>>>> $ ant package >>>>> gets stuck and prints following after a long time: >>>>> [ivy:retrieve] >>>>> [ivy:retrieve] :: problems summary :: >>>>> [ivy:retrieve] :::: ERRORS >>>>> [ivy:retrieve] Server access Error: Connection timed out url= >>>>> >> http://mirror.facebook.net/facebook/hive-deps/hadoop/core/hadoop-0.20.1/hadoop-0.20.1.tar.gz >>>>> >>>>> I also tried visiting the url directly from browser and expectedly that >> was >>>>> unreachable too. Can anyone take a look at this. Because of this its >>>>> impossible to compile Hive. >>>>> -Doffline=true didn't help either. >>>>> >>>>> Thanks, >>>>> Ashutosh >>>> >>> >> >>
