Merge lp://qastaging/~pfalcon/linaro-android-build-tools/ec2-s3-mirror into lp://qastaging/linaro-android-build-tools

Proposed by Paul Sokolovsky
Status: Merged
Merged at revision: 415
Proposed branch: lp://qastaging/~pfalcon/linaro-android-build-tools/ec2-s3-mirror
Merge into: lp://qastaging/linaro-android-build-tools
Diff against target: 12 lines (+3/-0)
1 file modified
node/setup-build-android (+3/-0)
To merge this branch: bzr merge lp://qastaging/~pfalcon/linaro-android-build-tools/ec2-s3-mirror
Reviewer Review Type Date Requested Status
Paul Sokolovsky Approve
Loïc Minier (community) Approve
Данило Шеган Pending
Georgy Redkozubov Pending
Review via email: mp+94055@code.qastaging.launchpad.net

Description of the change

Switch to beta S3 Ubunti EC2 mirror, per suggestion in lp:932088.

To post a comment you must log in.
Revision history for this message
Loïc Minier (lool) wrote :

This would work, but why care about encoding the first part ".*://[^.]*"? Isn't it simpler to just sed all .ec2.archive.ubuntu.com occurrences? Is this to avoid matching comments?
sed -r 's,(\.ec2\.archive\.ubuntu\.com)/,\1.s3.amazonaws.com/,'

(BTW you might want to escape the dots)

Looks like you could drop the two lines below too since an apt-get update is definitely run.

review: Approve
Revision history for this message
Paul Sokolovsky (pfalcon) wrote :

On Wed, 22 Feb 2012 08:45:28 -0000
Loïc Minier <email address hidden> wrote:

> Review: Approve
>
> This would work, but why care about encoding the first part
> ".*://[^.]*"? Isn't it simpler to just sed
> all .ec2.archive.ubuntu.com occurrences? Is this to avoid matching
> comments? sed -r
> 's,(\.ec2\.archive\.ubuntu\.com)/,\1.s3.amazonaws.com/,'

Well, I essentially took the command from Scott Moser's mail, checked
that it indeed works, and used as is. So, let's call it
upstream-recommended solution ;-).

>
> (BTW you might want to escape the dots)
>
> Looks like you could drop the two lines below too since an apt-get
> update is definitely run.

Learned it the hard way (real-world test) that it's actually needed.
Don't know why, as package indexes should be the same, but well, turns
out some other magic is involved.

--
Best Regards,
Paul

Linaro.org | Open source software for ARM SoCs
Follow Linaro: http://www.facebook.com/pages/Linaro
http://twitter.com/#!/linaroorg - http://www.linaro.org/linaro-blog

Revision history for this message
Loïc Minier (lool) wrote :

On Wed, Feb 22, 2012, Paul Sokolovsky wrote:
> Learned it the hard way (real-world test) that it's actually needed.
> Don't know why, as package indexes should be the same, but well, turns
> out some other magic is involved.

 But they show up as commented out in the mp?

 (I've seen it sometimes that two apt-get updates are needed to pick up
 the GPG signatures the first time and consider the packages trusted the
 second time, but this might actually be a bug)

--
Loïc Minier

Revision history for this message
Paul Sokolovsky (pfalcon) wrote :

Well, the commented apt-get update is another matter. The situation, we do initial update yet on jenkins side of thing, to make sure that we use the latest bzr to fetch the script as subject of this review. I commented it but left in the file to have a reminder that if anything changes on Jenkins side, it should be revived here to make sure that we use latest packages.

Now newly added apt-get update is needed to make sure that manually rewritten /etc/apt/sources.list actually works (maybe it's keys matter as you suggest). So, there 3 updates, needed in different combination on different circumstance, and I'd like to keep them that way, even if some commented. Because circumstances may and do change, and then it takes extra time to re-tweak setup scripts, whereas having general solution before the eyes (even if commented) is great helper in such cases.

Ok, I hope that answers the concerns, going to merge.

review: Approve

Preview Diff

[H/L] Next/Prev Comment, [J/K] Next/Prev File, [N/P] Next/Prev Hunk
The diff is not available at this time. You can reload the page or download it.

Subscribers

People subscribed via source and target branches