Ask Your Question
0

[Tacker] VNFs stuck in PENDING_SCALE_xxx

asked 2017-01-31 21:42:42 -0500

updated 2017-01-31 21:43:15 -0500

I'm using Tacker in a manual Newton installation, testing scaling VNFs.

The problem is: whenever HEAT refuses to scale resources because of "no change in capacity" (min or max instances reached) or the scaling request came inside the "cooldown" timer (both valid, common situations), the VNF remains in PENDING_SCALE_xxx state for a long time.

I'm wondering if there's a way to alter the timer that Tacker waits before putting the VNF back to "ACTIVE" state.

Thanks.

PS - Related to https://answers.launchpad.net/tacker/...

edit retag flag offensive close merge delete

Comments

can you tell what is the version of tacker and the version of tacker-client that you are using?

DarkKnight gravatar imageDarkKnight ( 2017-02-02 22:49:54 -0500 )edit

You can try editing these values in the tacker.conf file "stack_retry_wait" and "stack_retries" under [openstack-vim] section and then restart the tackerd to reflect the changes.

DarkKnight gravatar imageDarkKnight ( 2017-02-02 22:52:35 -0500 )edit

@DarkKnight: thanks, there it is! but it seems Tacker's ''retries" are just monitor queries to Heat, to see if it already completed, they are not actual retries of the stack's creation/deletion. So at the end, it's just a timer (retries * retry_wait) that Tacker waits for Heat to complete anything.

gianpietro gravatar imagegianpietro ( 2017-02-02 23:56:58 -0500 )edit

If you notice, these parameters are defined under two different sections viz. [openstack-vim] and [tacker-heat]. Do you know how these are different?

DarkKnight gravatar imageDarkKnight ( 2017-02-03 01:42:07 -0500 )edit

@DarkKnight: I don't have the [tacker-heat] section, just the vim one, where I made the tests. Just noticed I'm not using the latest tacker-server version, I have 0.5.1.dev141, which one do you have?

gianpietro gravatar imagegianpietro ( 2017-02-04 23:39:19 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2017-02-14 08:55:49 -0500

Thanks to @DarkKnight, the answer to this challenge is to edit the "stack_retry_wait" and "stack_retries" parameters in tacker.conf .

It seems that through the retries*retry_wait timer, Tacker is just monitoring Heat until it finishes any creation/deletion/scaling task. If we configure the timer too short to avoid the VNF being locked in PENDING tasks, it may not wait enough time for the VNF actual creation and would throw an error. It would be ideal for Heat to signal Tacker whenever it couldn't perform the task, just as it seems to happen when a stack is successfully created.

If you consider this an issue, please add your comment here as well: https://answers.launchpad.net/tacker/...

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

3 followers

Stats

Asked: 2017-01-31 21:42:42 -0500

Seen: 145 times

Last updated: Feb 14 '17