Segue Startup Failure

51 views
Skip to first unread message

Hugh MacMullan

unread,
Jun 4, 2012, 3:08:41 PM6/4/12
to Segue for R
Hi Team!

Let me start of by saying "huzzah for Segue!" What a cool project.
Simple is the way for my users for sure.

I'm trying to start a new cluster (following the details on Jeffrey
Breen's awesome "Abusing" article), and it seems to be failing at the
R installation portion of the process, a dependency check failure (see
OUTPUT below).

Any workaround? Any other details I can provide?

AMI was: ami-a9b362c0 (if that's not set in stone?)

Thought I would report!

-Hugh

P.S. -- what I did:
> library(segue)
library(segue)
Loading required package: rJava
Loading required package: caTools
Loading required package: bitops
Segue did not find your AWS credentials. Please run the
setCredentials() function.
> setCredentials('MYCRED','MYCRED2')
setCredentials('MYCRED', 'MYCRED2')
> myCluster <- createCluster(numInstances=2)
myCluster <- createCluster(numInstances=2)

then STARTING ... BOOTSTRAPPING ... SHUTTING_DOWN ... FAILED

=========== OUTPUT (stdout is the kicker) ===========

controller:
2012-06-04T17:13:43.823Z INFO Fetching file 's3://rtmpveqqfdfzmifbjs0w-
segue/bootstrapLatestR.sh'
2012-06-04T17:13:44.881Z INFO Working dir /mnt/var/lib/bootstrap-
actions/1
2012-06-04T17:13:44.881Z INFO Executing /mnt/var/lib/bootstrap-actions/
1/bootstrapLatestR.sh
2012-06-04T17:14:11.902Z INFO Execution ended with ret val 100
2012-06-04T17:14:11.903Z ERROR Execution failed with code '100'

stderr:
+ sudo tee /etc/apt/sources.list.d/R.list
+ gpg --keyserver pgp.mit.edu --recv-key 381BA480
gpg: directory `/home/hadoop/.gnupg' created
gpg: new configuration file `/home/hadoop/.gnupg/gpg.conf' created
gpg: WARNING: options in `/home/hadoop/.gnupg/gpg.conf' are not yet
active during this run
gpg: keyring `/home/hadoop/.gnupg/secring.gpg' created
gpg: keyring `/home/hadoop/.gnupg/pubring.gpg' created
gpg: requesting key 381BA480 from hkp server pgp.mit.edu
gpg: /home/hadoop/.gnupg/trustdb.gpg: trustdb created
gpg: key 381BA480: public key "Johannes Ranke (CRAN Debian archive)
<jra...@uni-bremen.de>" imported
gpg: no ultimately trusted keys found
gpg: Total number processed: 1
gpg: imported: 1
+ gpg -a --export 381BA480
+ sudo apt-key add jranke_cran.asc
+ sudo apt-get update
+ sudo DEBIAN_FRONTEND=noninteractive apt-get install -t testing --yes
--force-yes --no-install-recommends r-base
E: Broken packages

stdout:
# Change these lines if you don't want to use the main CRAN mirror.
# debian R upgrade
deb http://cran.r-project.org/bin/linux/debian squeeze-cran/
# deb-src http://cran.r-project.org/cran/bin/linux/debian squeeze-
cran/

OK
Get:1 http://security.debian.org squeeze/updates Release.gpg [836 B]
Ign http://security.debian.org/ squeeze/updates/contrib Translation-en
Ign http://security.debian.org/ squeeze/updates/main Translation-en
Ign http://security.debian.org/ squeeze/updates/non-free Translation-
en
Get:2 http://security.debian.org squeeze/updates Release [86.9 kB]
Get:3 http://http.us.debian.org squeeze Release.gpg [1672 B]
Ign http://http.us.debian.org/debian/ squeeze/contrib Translation-en
Ign http://http.us.debian.org/debian/ squeeze/main Translation-en
Ign http://http.us.debian.org/debian/ squeeze/non-free Translation-en
Get:4 http://http.us.debian.org testing Release.gpg [836 B]
Get:5 http://http.us.debian.org/debian/ testing/contrib Translation-en
[35.0 kB]
Get:6 http://http.us.debian.org/debian/ testing/main Translation-en
[3857 kB]
Get:7 http://security.debian.org squeeze/updates/main i386 Packages
[229 kB]
Get:8 http://http.us.debian.org/debian/ testing/non-free Translation-
en [62.7 kB]
Get:9 http://http.us.debian.org squeeze Release [111 kB]
Get:10 http://http.us.debian.org testing Release [190 kB]
Get:11 http://cran.r-project.org squeeze-cran/ Release.gpg [198 B]
Ign http://cran.r-project.org/bin/linux/debian/ squeeze-cran/
Translation-en
Get:12 http://security.debian.org squeeze/updates/contrib i386
Packages [674 B]
Get:13 http://security.debian.org squeeze/updates/non-free i386
Packages [14 B]
Get:14 http://cran.r-project.org squeeze-cran/ Release [3651 B]
Get:15 http://http.us.debian.org squeeze/main i386 Packages [6568 kB]
Get:16 http://cran.r-project.org squeeze-cran/ Packages [16.9 kB]
Get:17 http://http.us.debian.org squeeze/contrib i386 Packages [54.1
kB]
Hit http://http.us.debian.org squeeze/non-free i386 Packages
Get:18 http://http.us.debian.org testing/main i386 Packages/DiffIndex
[7876 B]
Get:19 http://http.us.debian.org testing/contrib i386 Packages/
DiffIndex [7819 B]
Get:20 http://http.us.debian.org testing/non-free i386 Packages/
DiffIndex [7819 B]
Get:21 http://http.us.debian.org testing/main i386 Packages [5837 kB]
Get:22 http://http.us.debian.org testing/contrib i386 Packages [43.9
kB]
Get:23 http://http.us.debian.org testing/non-free i386 Packages [71.4
kB]
Fetched 17.2 MB in 18s (952 kB/s)
Reading package lists...
Reading package lists...
Building dependency tree...
Reading state information...
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
libc6-dev : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be
installed

James Long

unread,
Jun 5, 2012, 7:57:18 AM6/5/12
to seg...@googlegroups.com
Thanks for the head's up! I'm testing currently. I'm suspicious of the
m1.small instance which is the default. It's 32 bit and all the other
instances are 64bit. It may be time to block m1.small and force using
64 bit instances.... but first I have to ensure that's the real issue.

-J

James Long

unread,
Jun 5, 2012, 8:04:16 AM6/5/12
to seg...@googlegroups.com
scratch that. I'm getting failures on m1.large.

-J

James Long

unread,
Jun 5, 2012, 12:33:34 PM6/5/12
to seg...@googlegroups.com
A few things needed some love. But I could not get m1.small working
right. I changed a few things that would cause failures, I set segue
to reject m1.small and default to m1.large. And I changed the version
to 0.04. Give it a whirl now and let me know if it explodes.

-J

Hugh MacMullan

unread,
Jun 5, 2012, 4:29:20 PM6/5/12
to Segue for R
It's alive! Thanks for the speedy response James, looking good.
-Hugh
> >>> Get:7http://security.debian.orgsqueeze/updates/main i386 Packages
> >>> [229 kB]
> >>> Get:8http://http.us.debian.org/debian/testing/non-free Translation-
> >>> en [62.7 kB]
> >>> Get:9http://http.us.debian.orgsqueeze Release [111 kB]
> >>> Get:10http://http.us.debian.orgtesting Release [190 kB]
> >>> Get:11http://cran.r-project.orgsqueeze-cran/ Release.gpg [198 B]
> >>> Ignhttp://cran.r-project.org/bin/linux/debian/squeeze-cran/
> >>> Translation-en
> >>> Get:12http://security.debian.orgsqueeze/updates/contrib i386
> >>> Packages [674 B]
> >>> Get:13http://security.debian.orgsqueeze/updates/non-free i386
> >>> Packages [14 B]
> >>> Get:14http://cran.r-project.orgsqueeze-cran/ Release [3651 B]
> >>> Get:15http://http.us.debian.orgsqueeze/main i386 Packages [6568 kB]
> >>> Get:16http://cran.r-project.orgsqueeze-cran/ Packages [16.9 kB]
> >>> Get:17http://http.us.debian.orgsqueeze/contrib i386 Packages [54.1
> >>> kB]
> >>> Hithttp://http.us.debian.orgsqueeze/non-free i386 Packages
> >>> Get:18http://http.us.debian.orgtesting/main i386 Packages/DiffIndex
> >>> [7876 B]
> >>> Get:19http://http.us.debian.orgtesting/contrib i386 Packages/
> >>> DiffIndex [7819 B]
> >>> Get:20http://http.us.debian.orgtesting/non-free i386 Packages/
> >>> DiffIndex [7819 B]
> >>> Get:21http://http.us.debian.orgtesting/main i386 Packages [5837 kB]
> >>> Get:22http://http.us.debian.orgtesting/contrib i386 Packages [43.9
> >>> kB]
> >>> Get:23http://http.us.debian.orgtesting/non-free i386 Packages [71.4

Aaron

unread,
Jul 7, 2012, 8:28:33 PM7/7/12
to seg...@googlegroups.com
Hello There,

Been playing around as well.  Made some steller progress as the simplicity of the design but I'm getting the same error tonight?  Using Segue files as posted on July 5th.  Tried it on my Mac and Ubuntu boxes with the same results.    



<snip from the logs>
Get:23 http://http.us.debian.org testing/non-free amd64 Packages [74.7 kB]
Fetched 17.3 MB in 13s (1242 kB/s)
Reading package lists...
Reading package lists...
Building dependency tree...
Reading state information...
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
 libc6-dev : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be installed
 libgcc1 : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be installed
 libgfortran3 : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be installed
 libgomp1 : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be installed
 libstdc++6 : Breaks: gcc-4.4 (< 4.4.6-4) but 4.4.5-8 is to be installed

James Long

unread,
Jul 9, 2012, 9:30:53 AM7/9/12
to seg...@googlegroups.com
Aaron,

It looks like there are some issues with the bootstrap script that
loads up the instances with needed packages. I'll start debugging.

Do you know which instance size you were using when you got the error?

-J

Aaron

unread,
Jul 9, 2012, 10:08:49 AM7/9/12
to Segue for R
I used the default, then tried m1.xlarge, i also tried the m1.small
but Segue overrode this with m1.large as per the 32 bit issue you
avoid.

On Jul 9, 9:30 am, James Long <jdl...@gmail.com> wrote:
> Aaron,
>
> It looks like there are some issues with the bootstrap script that
> loads up the instances with needed packages. I'll start debugging.
>
> Do you know which instance size you were using when you got the error?
>
> -J
>
>
>
>
>
>
>
> On Sat, Jul 7, 2012 at 9:28 PM, Aaron <aaron.walt...@gmail.com> wrote:
> > Hello There,
>
> > Been playing around as well.  Made some steller progress as the simplicity
> > of the design but I'm getting the same error tonight?  Using Segue files as
> > posted on July 5th.  Tried it on my Mac and Ubuntu boxes with the same
> > results.
>
> > <snip from the logs>
> > Get:23http://http.us.debian.orgtesting/non-free amd64 Packages [74.7 kB]
> >> > >>> Get:7http://security.debian.orgsqueeze/updates/maini386 Packages
> >> > >>> [229 kB]
> >> > >>> Get:8http://http.us.debian.org/debian/testing/non-freeTranslation-
> >> > >>> en [62.7 kB]
> >> > >>> Get:9http://http.us.debian.orgsqueezeRelease [111 kB]
> >> > >>> Get:10http://http.us.debian.orgtestingRelease [190 kB]
> >> > >>> Get:11http://cran.r-project.orgsqueeze-cran/Release.gpg [198 B]
> >> > >>> Ignhttp://cran.r-project.org/bin/linux/debian/squeeze-cran/
> >> > >>> Translation-en
> >> > >>> Get:12http://security.debian.orgsqueeze/updates/contribi386
> >> > >>> Packages [674 B]
> >> > >>> Get:13http://security.debian.orgsqueeze/updates/non-freei386
> >> > >>> Packages [14 B]
> >> > >>> Get:14http://cran.r-project.orgsqueeze-cran/Release [3651 B]
> >> > >>> Get:15http://http.us.debian.orgsqueeze/maini386 Packages [6568 kB]
> >> > >>> Get:16http://cran.r-project.orgsqueeze-cran/Packages [16.9 kB]
> >> > >>> Get:17http://http.us.debian.orgsqueeze/contribi386 Packages [54.1
> >> > >>> kB]
> >> > >>> Hithttp://http.us.debian.orgsqueeze/non-freei386 Packages
> >> > >>> Get:18http://http.us.debian.orgtesting/maini386 Packages/DiffIndex
> >> > >>> [7876 B]
> >> > >>> Get:19http://http.us.debian.orgtesting/contribi386 Packages/
> >> > >>> DiffIndex [7819 B]
> >> > >>> Get:20http://http.us.debian.orgtesting/non-freei386 Packages/
> >> > >>> DiffIndex [7819 B]
> >> > >>> Get:21http://http.us.debian.orgtesting/maini386 Packages [5837 kB]
> >> > >>> Get:22http://http.us.debian.orgtesting/contribi386 Packages [43.9
> >> > >>> kB]
> >> > >>> Get:23http://http.us.debian.orgtesting/non-freei386 Packages [71.4

James Long

unread,
Jul 9, 2012, 10:21:41 AM7/9/12
to seg...@googlegroups.com
Cool. Thanks.

I'm able to reproduce the error and am currently trying to cook up a
solution. Thanks for giving me such good error info. It makes ramping
up the debugging much easier.

-J
Reply all
Reply to author
Forward
0 new messages