Unable to launch AWS EC2 instances (version 2.5)

[Dale-Kurt Murray]

I'm unable to launch AWS EC2 instances, the scalr instance has been configured with the 'Event handler URL' and 'Server IP address'.

The environment has also been configured for AWS with the account number, Access Key, Secret Key, X.509 Certificate file and X.509 Private Key file.

A farm and a app-apache-ubuntu-ebs role has been create, however when the farm is launched no instances are launched.

The cronjob are also setup, in fact I manually executed the following command /usr/bin/php -q /var/scalr/app/cron-ng/cron.php --DeployManager

If there is anything that I have missed please let me know.

[Enrico Detoma]

I have the same problem with Scalr 3.0.
Everything seems to be configured, but no instances are launched when I launch the farm.
Running /usr/bin/php -q /var/www/app/cron-ng/cron.php --DeployManager manually, in the log I found:
Found 0 deployment tasks
What should I look for, to solve the problem?
Thanks

[Chris Mills]

I have the exact same problem, but with Scalr 2.2.1. I noticed in a
previous post you referenced a guide to self-hosting - I realized I
also used the same guide, and I am willing to bet the OP did as well.

It seems to me that we all may have the same issue, as a result of a
single oversight in this guide. Perhaps if someone more experienced
can look at it, we can all solve this issue.

On May 5, 7:13 am, Enrico Detoma <enrico.det...@gmail.com> wrote:
> I have the same problem with Scalr 3.0.
> Everything seems to be configured, but no instances are launched when I
> launch the farm.
> Running /usr/bin/php -q /var/www/app/cron-ng/cron.php --DeployManager
> manually, in the log I found:

> *Found 0 deployment tasks
> *What should I look for, to solve the problem?

[Srini]

1.  You can start by looking at the logs.  Login to scalr with the admin user and you can view the system logs.  Hopefully you will find something there.

2.  Double check and make sure all cron jobs (cron and cron-ng) are configured and running.  Either verify cron log or you can simply run "top" and check if php processes are triggered regularly.

3.  Just to make sure its not an issue that occurs after sometime, restart mysqld and apache and try again.  If it works then it could be database connection issues.  (The fact that it says "0 deployment tasks" is disturbing.)

Srini

[Chris Mills]

Actually, I rescind my comment.

After fixing one php error I encountered before I finally got that
output I last replied with, my Scalr installation appears to be
functioning perfectly.

I am not sure what exactly fixing that error (in RRD, I believe) did,
but now my installation of Scalr performs all functions I have tried
so far exactly as I expect it to.

Sorry to bother you, and thank you for your assistance.

[Chris Mills]

Well, this is where my problem diverges from the two above - running
the command "/usr/bin/php -q /var/scalr/app/cron-ng/cron.php --
DeployManager" resolves into the following error:

Option "DeployManager" is not recognized.

Usage: /var/www/app/cron-ng/cron.php [options]

--MessagingQueue Manage messages queues
--Poller Manage poller
--SNMPStatsPoller SNMP stats Poller
--ScalarizrMessaging Process ingoing Scalarizr messages
--farm-id [ <string ] Affect only this farm (With --
ScalarizrMessaging, --Scaling)
--Scaling Roles scaling
--help Print this help


Now, by my count, there are 5 distinct cron jobs listed here. When I
check the Core Settings for Scalr in the administration panel, I see
the number it has chosen for the "Cron processes number" field is also
5.

As per the guide for 2.5 (located here:
http://wiki.scalr.net/index.php?title=Installing_Scalr/Scalr_2.5_%2B_Ubuntu_11.10_(Oneiric_Ocelot)
), I have specified 15 cron jobs total - coincidentally, I am missing
much of the functionality I would assume would be performed by the
'missing' 10 jobs. For instance, server farms will not run deploy jobs
on their own, a function I assume is done by DeployManager (I have no
problem manually starting a server out of a specific role, so it's not
a communication problem between AWS and Scalr), and I cannot seem to
create a snapshot of a running server and turn it into a new role
(something I assume is done by the BundleTasksManager job).

Checking cron.php, I can see that all the jobs specified in the guide
are *there* so my suspicion is that either the jobs have changed
between 2.2.1 (the version I am using) and 2.5 (the version the guide
apparently works for), or.. I have some other problem.

Any help on this issue would be much appreciated, thanks.

On May 9, 7:16 am, Srini <srinivasan.subraman...@gmail.com> wrote:

[Igor Savchenko]

2.5 is not the latest version. Latest one 3.0 available per request on
http://scalr.net

Regards,
igor

> --
> You received this message because you are subscribed to the Google Groups "scalr-discuss" group.
> To post to this group, send email to scalr-...@googlegroups.com.
> To unsubscribe from this group, send email to scalr-discus...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en.
>