[01.org Android-IA] repo sync fails

Betina Berg betzy1985 at gmx.net
Mon Jan 28 07:19:39 PST 2013


Hello Matt,

thanks for your response. Since two weeks I'm trying to download, again and again and again, without any success. Looping doesn't work for me.
I also have tried with git clone to download platform/frameworks/base. It fails as expected. The interesting thing is that git clone fails always on 93% after the download reachs 1GiB. Has the server any download size limit?
It seems that your server resets the connection after 1GiB is downloaded.
Maybe this limitation is only for people from outside, this could be the reason why it works for some and not for others.

Regards,
Betzy

> betzy1985 at gmx.net wrote:
> > Hello,
> >
> > I'm trying to download the Android on Intel source code, but everytime
> > repo sync fails at platform/frameworks/base. These are the output
> > messages:
> > Fetching projects:  98% (316/322)  Receiving objects:  43%
> > (298509/680356), 43.2Fetching projects:  99% (319/322)  Receiving
> objects:
> >  45% (306161/680356), 43.9fatal: The remote end hung up unexpectedly.00
> > GiB | 975 KiB/s
> > fatal: early EOF
> > fatal: index-pack failed
> > remote: Counting objects: 680356, done
> > remote: Finding sources: 100% (680356/680356)
> > error: RPC failed; result=56, HTTP code = 20015 MiB | 25 KiB/s
> > fatal: The remote end hung up unexpectedly
> > fatal: early EOF
> > fatal: index-pack failed
> > error: Cannot fetch platform/frameworks/base
> >
> > error: Exited sync due to fetch errors
> >
> > It's always on project 319/322, the receiving objects never gets to
> 100%,
> > sometimes it aborts near 20%, sometimes near 80%. It differs in each
> run.
> >
> > What can I do to download the full source code?
> 
> Hello Betzy,
> 
> This issue has been a persistent thorn in our sides and we don't know why
> it happens for some people and not for others. At this point, all I know
> to recommend is to re-try the download in a continuous loop until it
> completes successfully.
> 
> If anybody else has had success getting around this issue, please let us
> know.
> 
> thanks,
> Matt




More information about the Celadon mailing list