teknologist's picture

Hi, Since yesterday I have an issue with tklbam restore.
I installed a Redmine 10.04 basedVM from the iso image.
Then I did a tklbam-restore of a previous Redmine 8.04 based.

it worked beautifully (except it still shows 8.04 in some palces and offers to do a do-release-upgrade to lucid).

Anyway, that is not bothering me.

What is bothering me is that since then, I can't see my backups finalizing in the hub.

On the VM side everything seems to works great, nevertheless when I got the hub webpage I see

TurnKey Redmine First backup in progress... 0 bytes
even after the backup is finished. And I don't see any details of available restores.


Also here are the outputs of tklbam-status and -list:

root@redmine /etc# tklbam-status TKLBAM: Backup ID #6, Updated Sat 2011-05-21 17:30



root@redmine:~# tklbam-list   
# ID  SKPP  Created     Updated     Size (MB)  Label
   1  No    2010-10-13  2010-10-13  566.79     TurnKey Redmine
   2  No    2010-10-13  2011-04-24  4480.01    TurnKey Redmine
   3  No    2011-04-26  2011-05-19  730.72     TurnKey Redmine
   4  No    2011-05-20  -           0.01       TurnKey Redmine
   5  No    2011-05-21  -           0.01       TurnKey Redmine
   6  No    2011-05-21  -           0.01       TurnKey Redmine


Obviously 4,5 and 6 are backups I have been attempting but seem to fail (again the backup process on the vm seems to run fine)

Thanks in advance for the help you may provide !

Eric

PS: attached is a full backup log of the latest incremental -------------------------------------------------

Forum: 
Tags: 
teknologist's picture

It seems that doing a tklbam-restore with the right ID (even the one that shows First Backup in progress  0bytes) work fine:

 

root@redmine:~# tklbam list 
# ID  SKPP  Created     Updated     Size (MB)  Label
   1  No    2010-10-13  2010-10-13  566.79     TurnKey Redmine
   2  No    2010-10-13  2011-04-24  4480.01    TurnKey Redmine
   3  No    2011-04-26  2011-05-19  730.72     TurnKey Redmine
   4  No    2011-05-20  -           0.01       TurnKey Redmine
   5  No    2011-05-21  -           0.01       TurnKey Redmine
   6  No    2011-05-21  -           0.01       TurnKey Redmine

 

 

root@redmine:~# tklbam-restore 6
Restoring duplicity archive from s3://s3-eu-west-1.amazonaws.com/tklbam-lmtixm4p3jxjnvje
Synchronizing remote metadata to local cache...
Copying duplicity-full-signatures.20110521T145118Z.sigtar to local cache.
Copying duplicity-full.20110521T145118Z.manifest to local cache.
Copying duplicity-inc.20110521T145118Z.to.20110521T151150Z.manifest to local cache.
Copying duplicity-inc.20110521T151150Z.to.20110521T151430Z.manifest to local cache.
Copying duplicity-inc.20110521T151430Z.to.20110521T151732Z.manifest to local cache.
Copying duplicity-inc.20110521T151732Z.to.20110521T153027Z.manifest to local cache.
Copying duplicity-inc.20110521T153027Z.to.20110521T154018Z.manifest to local cache.
Copying duplicity-new-signatures.20110521T145118Z.to.20110521T151150Z.sigtar to local cache.
Copying duplicity-new-signatures.20110521T151150Z.to.20110521T151430Z.sigtar to local cache.
Copying duplicity-new-signatures.20110521T151430Z.to.20110521T151732Z.sigtar to local cache.
Copying duplicity-new-signatures.20110521T151732Z.to.20110521T153027Z.sigtar to local cache.
Copying duplicity-new-signatures.20110521T153027Z.to.20110521T154018Z.sigtar to local cache.
Last full backup date: Sat May 21 16:51:18 2011
 
(...)
 
So it seems it's only related to tklbam-list and hub webapp showing wrong infos...
 
Can anyone help me please ?
 
Thanks in advance,
 
Eric
Jeremy Davis's picture

The devs often check (and post on) the forums but at times when they are deep in development they may be quite slow to respond. If you use the Hub's feedback feature that will get their attention and also makes it easier for them to see what's happening (they automatically know your Hub ID cause it's linked to your feedback). 

Chris Musty's picture

My images all stopped backing up autonomously 2 days ago? I placed feedback as suggested but this is kinda weired.

Chris Musty

Director

Specialised Technologies

Jeremy Davis's picture

But I do know that the core devs have been doing work on the Hub so perhaps it's been an unintended consequence of something they've done.

Hopefully they'll get back to you directly (seeing as you've done the 'feedback' thing) and post here too at some point soon.

Chris Musty's picture

OK without touching anything they just completed a backup.

Phew!

I am developing the snott out of an app that uses databases so backups are kinda mandatory.

Thanks

Chris Musty

Director

Specialised Technologies

Alon Swartz's picture

First, let me put you at ease. All backups were successful (as teknologist mentioned) and have not stopped - the issue lay with the Hub's backup record data being out of sync.

Let me explain. When TKLBAM completes a backup (full or incremental) it performs an API call to the Hub, which in turn triggers a background task to be executed. The background task queries Amazon for changes and updates its records (which is displayed in the Hub web interface and tklbam-list).

So what happened? The background task queue got "stuck", so the backup_updated tasks weren't being executed, resulting in stale data. I unblocked the queue and all queued tasks are currently being executed. If you're data is still not in sync, give it a few minutes.

We're investigating what caused the queue to get "stuck" (first time something like this has happened), so sorry about this and thanks for reporting the issue. 

@jeremy, thanks for mentioned the Hub feedback feature, that got my attention. And yeah - we're working on the Hub (new features coming soon), but in an offline development version. We haven't touched the production hub since the last update.

Chris Musty's picture

Thanks for the explanation, while I dont entirely understand what you did I trust you are working on it.

I plan on using this for clients so I am going to be testing it thoroughly first. So far so good! I have tried several applications that you guys dont have as standard appliances and they backed up seamlessly, migrated easily and were a snap to manage.

Seeing as you guys have saved me about 38190238928 hours trying to set up my own amazon appliances I am more than happy to support and hope that the 10% you guys earn goes to further development and or wages. When I bring clients on board I plan to donate to the cause as appreciation for your work. Keep it up!

EDIT: I can see on my servers that backups went ahead successfully and that it was a "hub" issue. Should have looked there first!

Chris Musty

Director

Specialised Technologies

Jeremy Davis's picture

I should've uessed that you'd be working on a development version of the Hub rather than the production/live Hub. You're way to professional to do stuff like that! :)

Glad to hear that you've got it worked out.

teknologist's picture

Thank you very much for such a fast response and solution!

Everything works great again!

Keep up the good work!

Cheers,

Eric

Add new comment