Rundeck usage questions.

[Murali Bharadwaj]

I'm evaluating Rundeck to use as an application scheduler (not as source control build like Jenkins/Hudson) from Windows server and had few questions -

1. Can Rundeck be used as a file watcher (look for a trigger file and only when it's available start job/job group)? I looked into the docs but couldn't find any option that can serve as a file watcher. Do I need to write a plugin for achieving that feature and if so, can someone please point me to any doc that explains how to write a plugin and use it in the project.

2. Most of the jobs would be script based that will connect to SQL Server and execute stored procedures. Can this be done in Rundeck.

3. I want to use the email facility to send mail in case of success/failures. Since i'm behind a corporate Exchange server I couldn't get it working as the default port # of 25 was not accessible. While i can get the server settings from the exchange admin is there anything else i need to be doing besides configuring those settings in config file.

4. Is it possible to change db used by Rundeck to point to the application SQL Server and have the Rundeck logs written to custom tables and/or customize Rundeck to write to tables based on some condition.

5. Is it ok to run multiple instances of Rundeck or is it advisable to run only one instance?

I stumbled upon Rundeck only about a day back and still trying to read the docs and understand if it'll serve my purpose of an effective enterprise level scheduler (like Autosys, Tidal etc), so if any of my questions are rudimentary, please excuse me as i've not gone through the whole doc yet.

Any help/suggestion would be really helpful to me and I would like to thank in advance.

[Alex Honor]

Hi Murali,

My responses are in line:

On Sat, Jan 10, 2015 at 5:04 PM, Murali Bharadwaj <murali.b...@gmail.com> wrote:

I'm evaluating Rundeck to use as an application scheduler (not as source control build like Jenkins/Hudson) from Windows server and had few questions -

1. Can Rundeck be used as a file watcher (look for a trigger file and only when it's available start job/job group)? I looked into the docs but couldn't find any option that can serve as a file watcher. Do I need to write a plugin for achieving that feature and if so, can someone please point me to any doc that explains how to write a plugin and use it in the project.

Rundeck doesn't have a watcher. You could create a scheduled job that periodically checks for a file existing on a node and then invoke another job.

 

2. Most of the jobs would be script based that will connect to SQL Server and execute stored procedures. Can this be done in Rundeck.

Your command or script step could connect to the SQL server and execute the SPs. How are you doing this now?

 

3. I want to use the email facility to send mail in case of success/failures. Since i'm behind a corporate Exchange server I couldn't get it working as the default port # of 25 was not accessible. While i can get the server settings from the exchange admin is there anything else i need to be doing besides configuring those settings in config file.

See this page for email settings: http://rundeck.org/docs/administration/email-settings.html

You could also look at notification plugins to send messages to other tools.

 

4. Is it possible to change db used by Rundeck to point to the application SQL Server and have the Rundeck logs written to custom tables and/or customize Rundeck to write to tables based on some condition.

In 2.4.0, MSSQL is now supported. Change the dataSource configuration. See http://rundeck.org/docs/administration/setting-up-an-rdb-datasource.html

 

5. Is it ok to run multiple instances of Rundeck or is it advisable to run only one instance?

Yes you can run as many instances as you need. Typically there are at least two instances a "staging" and a "production" instance. Many have individual rundeck instances per user or team to develop jobs in their own environments.

 

I stumbled upon Rundeck only about a day back and still trying to read the docs and understand if it'll serve my purpose of an effective enterprise level scheduler (like Autosys, Tidal etc), so if any of my questions are rudimentary, please excuse me as i've not gone through the whole doc yet.

Any help/suggestion would be really helpful to me and I would like to thank in advance.

Take a look at the demos for working examples:

• https://github.com/rundeck/anvils-demo
  
• https://github.com/rundeck/guitars-demo 
  

 

--
You received this message because you are subscribed to the Google Groups "rundeck-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rundeck-discu...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

Alex Honor

[SimplifyOps, Inc | a...@simplifyops.com ]

Be sure to comment and vote on Rundeck Feature Development!

[Murali Bharadwaj]

Hi Alex,

Thanks for your reponse.

1. I'll try out and see how it works.

2. Currently i'm using SOS scheduler, however it's not very robust for regular use so trying out alternatives and that's how i came into Rundeck. It has templates which is enhanced for connecting to DB. So i was looking for something similar.

I'll take a look at the demos and also the docs and see how i can make better use of Rundeck.

Thanks for your help. 

Alex Honor

[SimplifyOps, Inc | a....@simplifyops.com ]

[tk yass]

Hi Alex,

I'm trying to evaluate rundeck as well and I was wondering if rundeck supports multiple job dependencies? if job3 is for example dependent on the completion of jobs 1 and 2 which both have different execution times, will I be able to check if both jobs 1 and 2 finished before I can trigger job3? tried using job state conditional to check for completion of executions but I found out that the state conditionals doesn't take into account the running state and execution outcome in the same workflow but it only looks for previous history so it wasn't helpful. I'm not sure if rundeck supports this feature? or I'm simply doesn't know how to implement it?

Thanks in advance!

On Sunday, January 11, 2015 at 12:28:27 PM UTC-5, Alex Honor wrote:

Alex Honor

[SimplifyOps, Inc | a....@simplifyops.com ]

[Alex Honor]

Hi tk,

For now you'll have to set up a scheduled job that uses the job conditional step to check if job1 and job2 completed and if so then execute job3. Use the error handler to run the Flow Control step to set a custom status if job1,2 did not complete yet.

Thanks

Thanks

To view this discussion on the web visit https://groups.google.com/d/msgid/rundeck-discuss/4a1cf3e1-dcd1-4629-acf6-6476f05ede6a%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

--

Alex Honor

[SimplifyOps, Inc | a...@simplifyops.com ]