Newbie question : Servers in a cycle of starting-initializing-terminating in self hosted Scalr

[Sreehari]

Hi All,
I'm trying to fix this problem for last 3 days, some help here would
be highly appreciated.

Context:
I downloaded the the latest scalr from github as a Zipped archive, and
followed the steps in listed here "http://wiki.scalr.net/pages/
viewpage.action?pageId=327743". Additionally, just so to be up to
date, I added the extra two crons listed here "http://wiki.scalr.net/
display/docs/Open-Source+Installation".
My Scalr is installed on a machine with public IP with following ports
open : 80, 8013, 8014.

Now when I create and launch a sample Lamp farm, my servers are caught
up in a cycle of start->intializing->pending terminate->terminated,
each cycle spawning its own server. (One server is terminated and
another is created, then terminated and so on). For each of the
created instance in Scalr, I can see the corresponding one being
created and then terminated in the AWS management console.

Anybody how met and resolved this issue before ?

What am I missing? Is there any DNS setting to be done ? Is there a
good/complete tutorial for self hosted/Open source Scalr deployment?
Please help !!

[Nick Toursky]

Hi there,

Did you check system/event logs?

--
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.

[Sree]

Hi Nick,
Thank you for the response.
The event log gives this message -  "Server '4569884f-1568-4209-8d11-69f12b285166' did not send 'hostUp' event in 2400 seconds after launch (Try increasing timeouts in role settings). Considering it broken. Terminating instance."

Regards
Sreehari

[Nick Toursky]

Connect to this instance using SSH and check Scalarizr logs to determine what's wrong there.

[Sree]

Also I get these log entries in the event log -
Jun 11, 2012 02:54:06
   
LAMP Farm Jun 8-A
   
/EventObserver
   
Cannot create volumeConfig for PromoteToMaster message: Scalr_Storage_Volume Volume # not found in database
   
 
   
   
Jun 11, 2012 02:52:06
   
LAMP Farm Jun 8-A
   
/FarmLog
   
Terminating server 'a07f5f1d-92d2-41d9-8988-3153b97f0897' (Platform: ec2) (Poller).
   
 
   
   
Jun 11, 2012 02:50:06
   
LAMP Farm Jun 8-A
   
/EventObserver
   
Cannot create volumeConfig for PromoteToMaster message: Scalr_Storage_Volume Volume # not found in database
   
 
   
   
Jun 11, 2012 02:49:55
   
LAMP Farm Jun 8-A
   
a07f5f1d-92d2-41d9-8988-3153b97f0897/scalarizr.handlers
   
Starting app

[Sree]

I'm attaching the system log from the terminated computer herwith (system.log)
Of this I notice this one which looks like an error -

* Starting Scalarizr  scalarizr      
    [80G
    [74G[ OK ]
landscape-client is not configured, please run landscape-config.



Any pointers ?

Regards
Sreehari

[Nick Toursky]

Scalarizr logs could be found at /var/log/scalarizr*

[Sree]

Yes Nick, I'm trying to ssh into one of these machines in this 15-20 minutes window. My admin password is not working. Will post scalarizr logs shortly.

[Sree]

Is there a default username and password to ssh into the vms ? i tried with root and the .pem file, which fails.

[Sree]

i'm trying to ssh into base centos 32 bit machine with user root and the key file. Asks for password. Tried my admin passwords on both scalr and AWS. Permission denied..

[Sree]

Could not ssh yet, certificate rejected. However could find that Ubuntu base 32 image doesn't have this problem and starts well.

[Sree]

Resolved. The server and database times were different due to difference in timezones. Put a timezone offset in Poller.php.

Thanks Nick for the concern and help.

[francesco cislaghi]

Hi,

i'm having the same issue. Can you post the code snippet you used to resolve this?

Thanks

Hi there,

To unsubscribe from this group, send email to scalr-discuss+unsubscribe@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en.

--
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-discuss+unsubscribe@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en.

--
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-discuss+unsubscribe@googlegroups.com.

For more options, visit this group at http://groups.google.com/group/scalr-discuss?hl=en.

--
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-discuss+unsubscribe@googlegroups.com.

[Sree]

HI francesco,
can you add another 1800 seconds to your timeout value for the role. If you continue to get the same error (Server xxx did not send 'hostUp' event in xxx seconds after launch) in the event log, most likely you too have the same timezone issue.

This case,
1. Go to Poller.php, find the line "$dtadded = strtotime($DBServer->dateAdded)" .
2. Modify this line to add the timezone difference between your db and app : $dtadded = strtotime($DBServer->dateAdded)+34351; (34351 seconds is around 10 hours, the timezone difference for me; yes this is a quick fix ). You can find the time difference for your setup by logging the difference between strtotime($DBServer->dateAdded) and time();

Regards
Sree

To view this discussion on the web visit https://groups.google.com/d/msg/scalr-discuss/-/nqVAn7-iwKMJ.

To post to this group, send email to scalr-...@googlegroups.com.

To unsubscribe from this group, send email to scalr-discus...@googlegroups.com.